Tesi sul tema "Clusters detection"
Segui questo link per vedere altri tipi di pubblicazioni sul tema: Clusters detection.
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Vedi i top-50 saggi (tesi di laurea o di dottorato) per l'attività di ricerca sul tema "Clusters detection".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Vedi le tesi di molte aree scientifiche e compila una bibliografia corretta.
1
Farrens, S. "Optical detection of galaxy clusters". Thesis, University College London (University of London), 2011. http://discovery.ucl.ac.uk/1318077/.
Abstract (sommario):
This thesis first presents a relatively straight forward approach for detecting galaxy clusters using spectroscopic data. A friend-of-friends algorithm based on that of Huchra & Geller (1982) is implemented with linking parameters that take into account selection effects on the the 2dF-SDSS and QSO (2SLAQ) Luminous Red Galaxy Survey (Cannon et al. 2006). The linking parameters are constrained using a mock halo catalogue. The galaxy groups and clusters found with this method have an average velocity dispersion of \sigma v = 467:97 kms-1 and an average size of R clt = 0:78 h-1Mpc. Cluster masses are estimated using the cluster velocity dispersions and appear consistent with values expected for genuine structures. The spectroscopic cluster catalogue is then used to calibrate and compare with a more complex method for detecting clusters using photometric redshifts based on the method of Botzler et al. (2004). The spectroscopic cluster catalogue can be reproduced by around 38% and up to 80% if matching is made only to groups and clusters with six or more members. This code is also applied to the Megaz-LRG DR7 catalogue (Collister & Lahav 2004) producing two catalogues. One that appears to have a good level of completeness relative to the 2SLAQ spectroscopic catalogue. A spectroscopic follow up of some preliminary results from the photometric cluster finder was made using the Anglo-Australian Telescope, which show that the majority of the clusters analysed are genuine and approximate masses can be estimated from the cluster velocity dispersions. Finally, some initials results from on going work in the Dark Energy Survey collaboration are presented, which cover simulated galaxy photometric redshift and colour analysis as well as cluster detection.
2
Mundnich, Batic Karel Bogomir. "Early detection of high volatility clusters using particle filters". Tesis, Universidad de Chile, 2013. http://www.repositorio.uchile.cl/handle/2250/115486.
Abstract (sommario):
Ingeniero Civil ElectricistaEl presente trabajo explora y analiza el uso de herramientas de procesamiento de señales que son comunes en áreas de Ingeniería Eléctrica y Pronóstico y Gestión de Salud en el análisis de series de tiempo financieras. El objetivo principal de este trabajo es detectar eventos de alto riesgo en una etapa temprana. De esta forma, el algoritmo propuesto emplea la fuerte relación entre volatilidad y riesgo y detecta clusters de alta volatilidad mediante el uso de la información obtenida de los procesos de estimación a través de Filtro de Partículas. Para alcanzar el objetivo mencionado, se utiliza la representación de espacio-estado estocástica uGARCH para modelar la volatilidad de retornos compuestos continuamente. Dada la no-observabilidad de la volatilidad, se implementan dos esquemas de Filtro de Partículas para su estimación: los enfoques clásico y sensible al riesgo. Este último incluye el uso de una Distribución de Pareto Generalizada como propuesta para el funcional de riesgo (y distribución de importancia) para asegurar la asignación de partículas en regiones del espacio-estado que están asociadas a variaciones rápidas de volatilidad del sistema. Para evaluar correctamente el rendimiento de las rutinas de filtrado, se han generado seis conjuntos de datos, donde ambos el estado y las mediciones son conocidas. Además, se ha realizado un análisis de sensibilidad sobre los seis conjuntos de datos, para así obtener los parámetros que permiten la mejor estimación de volatilidad. De estos resultados, se calculan valores promedios de parámetros que son luego utilizados en el esquema de detección. La etapa de detección explora tres diferentes técnicas. Primero, se propone la utilización de un test de hipótesis entre las estimaciones a priori y a posteriori de las distribuciones de probabilidad del Filtro de Partículas Sensible al Riesgo. Segundo, se utiliza el Discriminante de Fisher para comparar las estimaciones a posteriori de las densidades entre el Filtro de Partículas Clásico y el Sensible al Riesgo. Finalmente, se utiliza la Divergencia de Kullback-Leibler de la misma forma que el Discriminante de Fisher. Los algoritmos propuestos son probados en los datos generados artificialmente y en datos de acciones de IBM. Los resultados demuestran que el Filtro de Partículas Sensible al Riesgo propuesto supera la precisión del Filtro de Partículas en momentos de alzas no esperadas de volatilidad. Por otra parte, el test de hipótesis empleado en el proceso de filtrado sensible al riesgo detecta correctamente la mayoría de las alzas repentinas de volatilidad que conducen a la detección temprana de clusters de alta volatilidad. Finalmente, los algoritmos de detección propuestos basados en Discriminante de Fisher y Divergencia de Kullback-Leibler llevan a resultados donde la detección no es posible.
3
Crawford, Carolin Susan. "The detection of distant cooling flows". Thesis, University of Cambridge, 1988. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.293490.
4
McLoughlin, Kirstin J. "Computer aided detection of microcalcification clusters in digital mammogram images". Thesis, University of Canterbury. Electrical and Computer Engineering, 2004. http://hdl.handle.net/10092/6536.
Abstract (sommario):
Recent advancements in computer technology have ensured that early detection of breast cancer, via computer aided detection (CAD) schemes, has become a rapidly expanding field of research. There is a desire to improve the detection accuracy of breast cancer without increasing the number of falsely identified cancers. The CAD scheme considered here is intended to assist radiologists in the detection of micro calcification clusters, providing a real contribution to the mammography screening process.
Factors that affect the detection accuracy of micro calcifications in digital mammograms include the presence of high spatial frequency noise, and locally linear high intensity structures known as curvilinear structures (CLS). The two issues considered are how to compensate for the high frequency image noise and how to detect CLS thus removing their influence on micro calcification detection. First, an adaptive approach to modelling the image noise is adopted. This is derived directly from each mammogram and is adaptable to varying imaging conditions. It is found that compensating for the high frequency image noise significantly improves micro calcification detection accuracy. Second, due to the varying size and orientation of CLS in mammogram images, a shape parameter is designed for their detection using a multiresolution wavelet filter bank. The shape parameter leads to an efficient way of distinguishing curvilinear structures from faint micro calcifications. This improves micro calcification detection performance by reducing the number of false positive detections related to CLS.
The detection and segmentation of micro calcification clusters is achieved by the development of a stochastic model, which classifies individual pixels within a mammogram into separate classes based on Bayesian decision theory. Both the high frequency noise model and CLS shape parameters are used as input to this segmentation process. The CAD scheme is specifically designed to be independent of the modality used, simultaneously exploiting the image data and prior knowledge available for micro calcification detection.
A new hybrid clustering scheme enables the distinction between individual and clustered micro calcifications, where clustered micro calcifications are considered more clinically suspicious. The scheme utilises the observed properties of genuine clusters (such as a uniform distribution) providing a practical approach to the clustering process. The results obtained are encouraging with a high percentage of genuine clusters detected at the expense of very few false positive detections.
An extensive performance evaluation of the CAD scheme helps determine the accuracy of the system and hence the potential contribution to the mammography screening process. Comparing the CAD scheme developed with previously developed micro calcification detection schemes shows that the performance of this method is highly competitive. The best results presented here give a sensitivity of 91% at an average false positive detection rate of 0.8 false positives per image.
5
Forsberg, Viktor. "AUTOMATIC ANOMALY DETECTION AND ROOT CAUSE ANALYSIS FOR MICROSERVICE CLUSTERS". Thesis, Umeå universitet, Institutionen för datavetenskap, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-164740.
Abstract (sommario):
Large microservice clusters deployed in the cloud can be very difficult to both monitor and debug. Monitoring theses clusters is a first step towards detection of anomalies, deviations from normal behaviour. Anomalies are often indicators that a component is failing or is about to fail and should hence be detected as soon as possible. There are often lots of metrics available to view. Furthermore, any errors that occur often propagate to other microservices making it hard to manually locate the root cause of an anomaly, because of this automatic methods are needed to detect and correct the problems. The goal of this thesis is to create a solution that can automatically monitor a microservice cluster, detect anomalies, and find a root cause. The anomaly detection is based on an unsupervised clustering algorithm that learns the normal behaviour of each service and then look for data that falls outside that behaviour. Once an anomaly is detected the proposed method tries to match the data against predefined root causes. The proposed solution is evaluated in a real microservice cluster deployed in the cloud, using Kubernetes together with a service mesh and several other tools to help gather metrics and trace requests in the system.
6
Burato, Dario <1993>. "Load balancing and fault early detection for Apache Kafka clusters". Master's Degree Thesis, Università Ca' Foscari Venezia, 2019. http://hdl.handle.net/10579/15159.
Abstract (sommario):
Apache Kafka is a publish-subscribe message system, producers publish data to a cluster and clients subscribes to receive data.
The messages are sent by their producers and stored in partitions, the load balancing is performed thanks to their distribution between the cluster's nodes.
The component which assign a message to a partition is called partitioner, located inside every producer. When partitions lacks intrinsic meaning, and are used purely for load-balancing purposes, the default partitioners available with Apache Kafka aim only to get the same amount of messages shared between partitions.
The most common Apache Kafka cluster configuration is based on multiple identical systems, when a cluster is updated with new more performing components the old ones are usually removed. Even if re-balancing tools exists, it would take time to properly adapt to an hybrid cluster configuration, this is caused by partitioners focus on data amount rather than node performance.
The problem could be solved by changing the amount of partitions in each old and new system, matching their performance ratio, thus tricking the default partitioner logic, but this actually could hurt client performance. A proper partitioner which knows the performance of each cluser's node is a correct solution, this document will present a formal method to detect problematic scenarios and a custom partitioner that adapts to them.
7
Toni, Greta. "Detection and characterization of galaxy clusters in the COSMOS field with the AMICO algorithm". Master's thesis, Alma Mater Studiorum - Università di Bologna, 2022. http://amslaurea.unibo.it/25229/.
Abstract (sommario):
In this work we made use of the AMICO algorithm to detect clusters in COSMOS2015, a photometric galaxy catalogue of the COSMOS field (Laigle+16). We divided our study in two different analyses being the cluster search on r-band photometry in the range 0
8
Marshall, J. Brooke. "Prospective Spatio-Temporal Surveillance Methods for the Detection of Disease Clusters". Diss., Virginia Tech, 2009. http://hdl.handle.net/10919/29639.
Abstract (sommario):
In epidemiology it is often useful to monitor disease occurrences prospectively to determine the location and time when clusters of disease are forming. This aids in the prevention of illness and injury of the public and is the reason spatio-temporal disease surveillance methods are implemented. Care must be taken in the design and implementation of these types of surveillance methods so that the methods provide accurate information on the development of clusters. Here two spatio-temporal methods for prospective disease surveillance are considered. These include the local Knox monitoring method and a new wavelet-based prospective monitoring method.
The local Knox surveillance method uses a cumulative sum (CUSUM) control chart for monitoring the local Knox statistic, which tests for space-time clustering each time there is an incoming observation. The detection of clusters of events occurring close together both temporally and spatially is important in finding outbreaks of disease within a specified geographic region. The local Knox surveillance method is based on the Knox statistic, which is often used in epidemiology to test for space-time clustering retrospectively. In this method, a local Knox statistic is developed for use with the CUSUM chart for prospective monitoring so that epidemics can be detected more quickly. The design of the CUSUM chart used in this method is considered by determining the in-control average run length (ARL) performance for different space and time closeness thresholds as well as for different control limit values. The effect of nonuniform population density and region shape on the in-control ARL is explained and some issues that should be considered when implementing this method are also discussed.
In the wavelet-based prospective monitoring method, a surface of incidence counts is modeled over time in the geographical region of interest. This surface is modeled using Poisson regression where the regressors are wavelet functions from the Haar wavelet basis. The surface is estimated each time new incidence data is obtained using both past and current observations, weighing current observations more heavily. The flexibility of this method allows for the detection of changes in the incidence surface, increases in the overall mean incidence count, and clusters of disease occurrences within individual areas of the region, through the use of control charts. This method is also able to incorporate information on population size and other covariates as they change in the geographical region over time. The control charts developed for use in this method are evaluated based on their in-control and out-of-control ARL performance and recommendations on the most appropriate control chart to use for different monitoring scenarios is provided.Ph. D.
9
Moreira, Gladston Juliano Prates. "The detection of spatial clusters: graph and dynamic programming based methods". Universidade Federal de Minas Gerais, 2011. http://hdl.handle.net/1843/BUOS-8MCG9A.
Abstract (sommario):
This thesis addresses the spatial and space-time cluster detection problem. Two algorithms to solve the typical problem for spatial data sets are proposed. A fast method for the detection and inference of point data set spatial and space-time disease clusters is presented, the Voronoi Based Scan (VBScan). A Voronoi diagram is built for points representing population individuals (cases and controls). The number of Voronoi cells boundaries intercepted by the line segment joining two cases points defines the Voronoi distance between those points. This distance is used to approximate the density of the heterogeneous population and build the Voronoi distance Minimum Spanning Tree (VMST) linking the cases. The successive removal of edges from the VMST generates sub-trees which are the potential clusters. Finally, those clusters are evaluated through the scan statistic. Monte Carlo replications of the original data are used to evaluate the significance of the clusters. The ability to promptly detect space-time clusters of disease outbreaks, when the number of individuals is large, was shown to be feasible, due to the reduced computational load of VBScan. Numerical simulations showed that VBScan has higher power of detection, sensitivity and positive predicted value than the Elliptic PST. Furthermore, an application for dengue fever in a small Brazilian city is presented. In a second approach, the typical spatial cluster detection problem is reformulated as a bi-objective combinatorial optimization problem. We propose an exact algorithm based on dynamic programming, Geographical Dynamic Scan, which empirically was able to solve instances up to large size within a reasonable computational time. We show that the set of nonv dominated solutions of the problem, computed efficiently, contains the solution that maximizes the Kulldorffs Spatial Scan Statistic. The method allows arbitrary shaped clusters, which can be a collection of disconnected or connected areas, taking into account a geometric constraint. Note that this is not a serious disadvantage, provided that there is not a huge gap between its component areas. We present an empirical comparison of detection and spatial accuracy between our algorithm and the classical Kulldorffs Circular Scan, using the data set of Chagas disease cases in puerperal women in Minas Gerais state, Brazil.Esta tese aborda o problema de detecção de clusters espaciais e espaços-temporais. Dois algoritmos para resolver o típico problema de conjuntos de dados com processos espaciais são propostos. Um método eficiente para a detecção e inferência de clusters de doenças espaciais e espaços-temporais de dados pontuais é apresentado, o Voronoi Based Scan (VBScan). Um diagrama de Voronoi é construído para os pontos que representam indivíduos da população (casos e controles). O número de células de Voronoi interceptadas pelo segmento de linha que une de dois pontos que representam dois casos define a distância de Voronoi entre esses pontos. Esta distância é usada para aproximar a densidade da população heterogenia e construir a árvore geradora m·nima baseada na distância de Voronoi (VMST) ligando os casos. A remoção sucessiva de arestas da VMST gera sub-arvores que são os clusters candidatos potenciais. Finalmente, os clusters são avaliados através da estatística scan de Kulldorff. Simulações de Monte Carlo dos dados originais são usados para avaliar a significância dos clusters. A capacidade de detectar rapidamente clusters de surtos da doença, quando o número de indivíduos é grande, mostrou-se viável, devido à redução da carga computacional obtida com o VBScan. As simulações numéricas mostraram que o VBScan tem maior poder de detecção, sensibilidade e valor preditivo positivo do que o scan elíptico. Além disso, uma aplicação de casos e controles georeferenciados de dengue em uma cidade do Brasil é apresentado. Numa segunda abordagem, o problema típico de detecção de clusters espaciais é reformulado como um problema bi-objetivo de otimização combinatória Nós propomos um algoritmo exato baseado em programação dinâmica, Geographical Dynamic Scan, que empiricamente foi capaz de resolver os casos até de grande porte dentro de tempo computacional aceitável. Nós mostramos que o conjunto de soluções não dominadas do problema, encontradas eficientemente, contém a solução que maximiza a estatística scan de Kulldorf. O método permite clusters de formatos arbitrários, que podem ser uma coleção de regiões desconectadas ou conectadas, tendo em conta uma restrição geográfica. Note-se que esta não é uma séria desvantagem, desde que não haja um grande espaçamento entre as suas áreas. Apresentamos uma comparação empírica de detecção e precisão espacial entre o nosso algoritmo e o clássico Scan circular, utilizando dados de casos de doença de Chagas em mulheres parturientes no estado de Minas Gerais, Brasil.
10
Oliveira, Fernando Luiz Pereira de. "Nonparametric intensity bounds for the detection and delineation of spatial clusters". Universidade Federal de Minas Gerais, 2011. http://hdl.handle.net/1843/ICED-8GQJAE.
Abstract (sommario):
There is considerable uncertainty in the disease rate estimation for aggregated area maps, especially for small population areas. As a consequence the delineation of local clustering is subject to substantial variation. Consider the most likely disease cluster produced by any given method, like SaTScan Kulldorff [2006], for the detection and inference of spatial clusters in a map divided into areas; if this cluster is found to be statistically signifcant, what could be said of the external areas adjacent to the cluster? Do we have enough information to exclude them from a health program of prevention? Do all the areas inside the cluster have the same importance from a practitioner perspective? We propose a criterion to measure the plausibility of each area being part of a possible localized anomaly in the map. In this work we assess the problem of finding error bounds for the delineation of spatial clusters in maps of areas with known populations and observed number of cases. A given map with the vector of real data (the number of observed cases for each area) shall be considered as just one of the possible realizations of the random variable vector with an unknown expected number of cases. In our methodology we perform m Monte Carlo replications: we consider that the simulated number of cases for each area is the realization of a random variable with average equal to the observed number of cases of the original map. Then the most likely cluster for each replicated map is detected and the corresponding m likelihood values obtained by means of the m replications are ranked. For each area, we determine the maximum likelihood value obtained among the most likely clusters containing that area. Thus, we construct the intensity function associated to each area's ranking of its respective likelihood value among the m obtained values. The method is tested in numerical simulations and applied for three different real data maps for sharply and diffusely delineated clusters. The intensity bounds found by the method re ect the geographic dispersion of the detected clusters. The proposed technique is able to detect irregularly shaped and multiple clusters, making use of simple tools like the circular scan. Intensity bounds for the delineation of spatial clusters are obtained and indicate the plausibility of each area belonging to the real cluster. This tool employs simple mathematical concepts and interpreting the intensity function is very intuitive in terms of the importance of each area in delineating the possible anomalies of the map of rates. The Monte Carlo simulation requires an effort similar to the circular scan algorithm, and therefore it is quite fast. We hope that this tool should be useful in public health decision making of which areas should be prioritized.texto completo
11
Kratsas, Sherry L. "Parallelization of ECG template-based abnormality detection". Morgantown, W. Va. : [West Virginia University Libraries], 2000. http://etd.wvu.edu/templates/showETD.cfm?recnum=1697.
Abstract (sommario):
Thesis (M.S.)--West Virginia University, 2000.Title from document title page. Document formatted into pages; contains vii, 62 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 61-62).
12
Terrell, Thomas. "Structural health monitoring for damage detection using wired and wireless sensor clusters". Master's thesis, University of Central Florida, 2011. http://digital.library.ucf.edu/cdm/ref/collection/ETD/id/5055.
Abstract (sommario):
Sensing and analysis of a structure for the purpose of detecting, tracking, and evaluating damage and deterioration, during both regular operation and extreme events, is referred to as Structural Health Monitoring (SHM). SHM is a multi-disciplinary field, with a complete system incorporating sensing technology, hardware, signal processing, networking, data analysis, and management for interpretation and decision making. However, many of these processes and subsequent integration into a practical SHM framework are in need of development. In this study, various components of an SHM system will be investigated. A particular focus is paid to the investigation of a previously developed damage detection methodology for global condition assessment of a laboratory structure with a decking system. First, a review of some of the current SHM applications, which relate to a current UCF Structures SHM study monitoring a full-scale movable bridge, will be presented in conjunction with a summary of the critical components for that project. Studies for structural condition assessment of a 4-span bridge-type steel structure using the SHM data collected from laboratory based experiments will then be presented. For this purpose, a time series analysis method using ARX models (Auto-Regressive models with eXogeneous input) for damage detection with free response vibration data will be expanded upon using both wired and wireless acceleration data. Analysis using wireless accelerometers will implement a sensor roaming technique to maintain a dense sensor field, yet require fewer sensors. Using both data types, this ARX based time series analysis method was shown to be effective for damage detection and localization for this relatively complex laboratory structure. Finally, application of the proposed methodologies on a real-life structure will be discussed, along with conclusions and recommendations for future work.ID: 029810361; System requirements: World Wide Web browser and PDF reader.; Mode of access: World Wide Web.; Thesis (M.S.C.E.)--University of Central Florida, 2011.; Includes bibliographical references (p. 102-114).
M.S.C.E.
Masters
Civil, Environmental and Construction Engineering
Engineering and Computer Science
Civil Engineering
13
Galeazzi, Alessandro. "Opinion Mining and Clusters Detection in Online Public Debates: a Quantitative Analysis". Doctoral thesis, Università degli studi di Brescia, 2022. http://hdl.handle.net/11379/555016.
Abstract (sommario):
La nascita di piattaforme online tramite cui condividere informazioni con una platea virtualmente illimitata ha radicalmente cambiato il modo di comunicare. Attraverso i social media, chiunque è in grado di creare contenuti che non solo sono fruiti quasi in tempo reale da migliaia di utenti, ma che, grazie alle funzioni offerte dalle varie piattaforme online, possono ottenere un feedback immediato tramite commenti e reazioni. Questa modalità di comunicazione veloce e disintermediata, da un lato, fornisce il mezzo perfetto per la proliferazione di dibattiti su temi controversi, dall'altro, grazie anche alla presenza di algoritmi che riducono la diversità dei contenuti a cui un utente è esposto, crea l'ambiente perfetto per la formazione di gruppi ideologicamente omogenei di persone, definiti echo chambers. In questi ambienti, grazie a ripetute interazioni con altri ideologicamente affini, gli utenti sono esposti a una visione parziale e omogenea dell'argomento dibattuto, che li porta a rinforzare la propria opinione preesistente e ignorare posizioni contrarie.
Questa tesi si pone l'obiettivo di analizzare molteplici aspetti che influenzano il dibattito online. In particolare, è stata studiata l'evoluzione del dibattito nei social media riguardo argomenti controversi quali elezioni politiche e pandemia, evoluzione della polarizzazione e impatto di notizie non verificate sulle elezioni presidenziali americane, la presenza di echo chambers in varie piattaforme e attorno diversi argomenti di dibattito; è stata inoltre misurata la magnitudo dell'"infodemia" concomitante con la recente pandemia. Lo studio dimostra come gli utenti, quando dibattono sui social media attorno ad argomenti controversi, tendono ad aggregarsi in fazioni ideologicamente opposte, consumando informazioni che rinforzano la loro visione e ignorando altri punti di vista. Questa caratteristica sembra dominare il consumo di informazioni online, influenzando la diffusione sia di contenuti fondati sia di informazioni non verificate.The advent of online platforms dramatically changed the way people create and communicate content. In online social media, users can easily share information that thousands of peers may consume almost immediately. Moreover, the unique features offered by online social platforms also allow immediate feedback and interactions, creating the perfect environment for the proliferation of an intense debate around controversial topics. Nevertheless, this new and disintermediated type of communication and platforms' feed algorithms may influence the dynamics of online discussion, creating a fertile environment for the formation of clusters of users reinforcing their opinion through repeated interactions called echo chambers. In this thesis, we study the debate around controversial topics in online social media, such as political elections and disease outbreaks, and analyze the factors influencing its dynamics. We also assess the impact of unsubstantiated rumors and measure the shift in polarization around political elections. Finally, we compare the effect of echo chambers around several topics and across different social media and quantify the online infodemic concurrent with the recent pandemic. In our studies, we find evidence that users tend to cluster together into groups with opposite opinions around debated topics and consume information adhering to their system of beliefs. This characteristic appears to dominate the information consumption dynamics in online social media, influencing the spread of both confirmed news and unsubstantiated rumors.
14
Tembey, Mugdha. "Computer-Aided Diagnosis for Mammographic Microcalcification Clusters". [Tampa, Fla.] : University of South Florida, 2003. http://purl.fcla.edu/fcla/etd/SFE0000168.
15
More, Surhud, Hironao Miyatake, Masahiro Takada, Benedikt Diemer, Andrey V. Kravtsov, Neal K. Dalal, Anupreeta More et al. "DETECTION OF THE SPLASHBACK RADIUS AND HALO ASSEMBLY BIAS OF MASSIVE GALAXY CLUSTERS". IOP PUBLISHING LTD, 2016. http://hdl.handle.net/10150/621397.
Abstract (sommario):
We show that the projected number density profiles of Sloan Digital Sky Survey photometric galaxies around galaxy clusters display strong evidence for the splashback radius, a sharp halo edge corresponding to the location of the first orbital apocenter of satellite galaxies after their infall. We split the clusters into two subsamples with different mean projected radial distances of their members, < R-mem >, at fixed richness and redshift. The sample with smaller < R-mem > has a smaller ratio of the splashback radius to the traditional halo boundary R-200m than the subsample with larger < R-mem >, indicative of different mass accretion rates for these subsamples. The same subsamples were recently used by Miyatake et al. to show that their large-scale clustering differs despite their similar weak lensing masses, demonstrating strong evidence for halo assembly bias. We expand on this result by presenting a 6.6 sigma difference in the clustering amplitudes of these samples using cluster-photometric galaxy cross-correlations. This measurement is a clear indication that halo clustering depends on parameters other than halo mass. If < R-mem > is related to the mass assembly history of halos, the measurement is a manifestation of the halo assembly bias. However, our measured splashback radii are smaller, while the strength of the assembly bias signal is stronger, than the predictions of collisionless. cold dark matter simulations. We show that dynamical friction, cluster mis-centering, or projection effects are not likely to be the sole source of these discrepancies. However, further investigations regarding unknown catastrophic weak lensing or cluster identification systematics are warranted.
16
Yam, Margaret. "Detection and analysis of microcalcification clusters in X-ray mammograms using the h_i_n_t representation". Thesis, University of Oxford, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.409946.
17
Bhowmik, Kowshik. "Comparing Communities & User Clusters in Twitter Network Data". University of Cincinnati / OhioLINK, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1573223960589.
18
Tärning, Jacob. "Troll Detection : A study of source usage between clusters of Twitter tweets todetect Internet trolls". Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-209436.
Abstract (sommario):
The purpose of this study was to examine whether it is possible to detect possibly malicioustweets posted by so-called trolls by inspecting the usage of sources such as url links,hashtags, user mentions and other media between clusters of tweets. This was done byutilizing the latent dirichlet allocation algorithm to find and assign topics to every tweet,clustering the tweets through their topics with the k-means algorithm. The resulting clusterswas iterated through and data fetch and summarized to examine any difference between theclusters. The results suggest that this method for finding trolls is, in combination with alexical study of the tweets text, plausible.Syftet bakom denna studie var undersöka ifall det är möjligt att detektera sannolikt illvilligatweets postad av så kallade troll genom att inspektera användandet av källor såsomurl-länkar, hashtaggar, omnämnande av användare och annan media mellan olika kluster avtweets. Detta utfördes med hjälp av latent dirichlet allocation algoritmen för att finna ochtilldela ämnen till varje tweet, där tweeten klustrades på deras ämnestilldelning med hjälpk-means metoden. De resulterande klustrena itererades igenom och data från tweetenhämtades och summerades för att undersöka skillnader mellan klustrena. Resultatenantyder att denna metod tillsammans med en analys av tweetens text är möjligtvis lämpligför att detektera troll.
19
Cassa, Christopher A. "Spatial outbreak detection analysis tool : a system to create sets of semi-synthetic geo-spatial clusters". Thesis, Massachusetts Institute of Technology, 2004. http://hdl.handle.net/1721.1/33124.
Abstract (sommario):
Thesis (M. Eng. and S.B.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2004.Includes bibliographical references (leaves 55-57).
Syndromic surveillance systems, especially software systems, have emerged as the leading outbreak detection mechanisms. Early outbreak detection systems can assist with medical and logistic decision support. One important concern for effectively testing these systems in practice is the scarcity of authentic outbreak health data. Because of this shortage, creating suitable geotemporal test clusters for surveillance algorithm validation is essential. Described is an automated tool that creates artificial patient clusters by varying a large variety of realistic outbreak parameters. The cluster creation tool is an open-source program that accepts a set of outbreak parameters and creates artificial geospatial patient data for a single cluster or a series of similar clusters. This helps automate the process of rigorous testing and validation of outbreak detection algorithms. Using the cluster generator, single patient clusters and series of patient clusters were created - as files and series of files containing patient longitude and latitude coordinates. These clusters were then tested and validated using a publicly-available GIS visualization program. All generated clusters were properly created within the ranges that were entered as parameters at program execution. Sample semi-synthetic datasets from the cluster creation tool were then used to validate a popular spatial outbreak detection algorithm, the M-Statistic.
by Christopher A. Casa.
M.Eng.and S.B.
20
Panebianco, Gabriele. "A new implementation of an optimal filter for the detection of galaxy clusters through weak lensing". Master's thesis, Alma Mater Studiorum - Università di Bologna, 2021. http://amslaurea.unibo.it/24444/.
Abstract (sommario):
We developed a new version of a C++ code, Get the Halo 2021, that implements the optimal linear matched filter presented in Maturi et al.(2005).
Our aim is to detect dark matter haloes of clusters of galaxies through their weak gravitational lensing signatures applying the filter to a catalogue of simulated galaxy ellipticities. The dataset represents typical data that will be available thanks to the Euclid mission, thus we are able to forecast the filter performances on weak lensing data obtained by Euclid. The linear matched filter is optimised to maximise the signal-to-noise ratio (S/N) of the detections and minimise the number of spurious detections caused by superposition of large-scale structures; this is achieved by suppressing those spatial frequencies dominated by the large-scale structure contamination.
We compared our detections with the true population of dark matter haloes used to produce the catalogue of ellipticities. We confirmed the expectations on the filter performance raised by Maturi et al.(2005) and Pace et al.(2007). We found that S/N 7 can be considered as a reliable threshold to detect haloes through weak lensing as 83% of our detections with S/N>7 were matched to the haloes; this is consistent with Pace et al.(2007). The purity of our catalogues of detections increases as a function of S/N and reaches 100% at S/N 10.5-11. We also confirmed that the filter selects preferentially haloes with redshift between 0.2 and 0.5, that have an intermediate distance between observer and background sources, condition that maximises the lensing effects. The completeness of our catalogues is a steadily growing function of the mass until 4-5Msun/h, where it reaches values 58-68%.
Our algorithm might be used to enhance the reliability of the detections of the AMICO code (Bellagamba et al.2018), the optimal linear matched filter implemented in the Euclid data analysis pipeline to identify galaxy clusters in photometric data (Euclid Collaboration et al.2019).
21
Castro, Ginard Alfred. "Detection, characterisation and use of open clusters in a Galactic context in a Big Data environment". Doctoral thesis, Universitat de Barcelona, 2021. http://hdl.handle.net/10803/671790.
Abstract (sommario):
Open clusters are groups of stars, gravitationally bound together, that were born from the same molecular cloud and, thus, share similar positions, kinematics, ages and metallicities. Traditional methods to detect open clusters rely in the visual inspection of regions of the sky to look for positional overdensities of stars, which then are checked to follow an isochrone pattern in a colour-magnitude diagram. The
publication of the second Gaia data release, with more than 1.3 billion stars with parallax and proper motion measurements together with mean photometry in three broadbands, boosted the development of novel machine learning-based techniques to automatise the search for open clusters, using both the astrometric and photometric information.
The characterised open clusters in the Galaxy are popular tracers of properties of the Galactic disc such as the structure and evolution of the spiral arms, or testbed for stellar evolution studies for instance, because their astrophysical parameters are estimated with greater precision than for field stars. Therefore, a good understanding of the open cluster population in the Milky Way is key for Galactic archaeology studies.
Our aim for this thesis is to transform classical methodologies to detect different kinds of patterns from astronomical data, that mostly relies on visual inspection, to an automatic data mining procedure to extract meaningful information from stellar catalogues. We also aim to use the result of the application of machine learning techniques to Gaia data, in a broader Galactic context.
We have developed a data mining methodology to blindly search for open clusters in the Galactic disc. First, we use a density-based clustering algorithm, DBSCAN, to search for overdensities in the five-dimensional astrometric parameter space in Gaia data.
The deployment of the clustering step in a Big Data environment, at the MareNostrum
supercomputer located in the Barcelona Supercomputing Center, prevents the search to be limited by computational limitations. Second, the detected overdensities are classified into mere statistical or physical overdensities using an artificial neural network trained to recognise the isochrone pattern that open cluster member stars follow in a colour-magnitude diagram.
We estimate astrophysical parameters such as ages, distances and line-of-sight extinctions for the whole open cluster population using an artificial neural network trained on well-known open clusters. We use this additional information, together with radial velocities gathered from different space-based and ground-based surveys, to trace the Galactic spiral present-day structure using GaussianMixtureModels to associate the young (< 30 Myr) open clusters to their mother spiral arms. We also describe the spiral arms evolution during the last 80 Myr to provide new insights into the nature of the Milky Way spiral structure.
The automatization of the open cluster detection procedure, together with its deployment in a Big Data environment, has resulted in more than 650 new open clusters detected with this methodology. The new UBC clusters (named after the
University of Barcelona) represent one-third of the actual open clusters census (2017 objects with Gaia DR2 parameters), and it is the largest single contribution to the open cluster catalogue.
We are able to add 264 young open clusters (< 30 Myr) to the 84 high-mass star- forming regions traditionally used to trace spiral arms, to increase the Galactocentric azimuth range where the Milky Way spiral arms are defined, and better estimate their present-day parameters. By analysing the age distribution of the open clusters across the Galactic spiral arms, and computing the spiral arms pattern speeds following the open clusters orbits from their birthplaces, we are able to disfavour classical density waves as the main mechanism for the formation of the Milky Way spiral arms, favouring a transient behaviour.
This thesis has shown that the use of machine learning, with proper treatment of the computational resources, has a long journey ahead in a data-dominated future for Astronomy.Els cúmuls estel·lars oberts són conjunts d'estels, lligats gravitatòriament, nascuts al mateix núvol molecular que tenen propietats similars. Aquests cúmuls són traçadors populars de la estructura del disc Galàctic, com ara els braços espirals. El segon llançament de dades de Gaia, amb més de 1300 milions d'estels, impossibilita la detecció de cúmuls a partir de mètodes tradicionals degut al gran volum del catàleg. Per això, el desenvolupament de tècniques automàtiques per aquest fi ha crescut juntament amb el volums dels catàlegs a analitzar. Hem desenvolupat una metodologia per a la cerca a cegues de cúmuls oberts al disc Galàctic. Hem utilitzat un algoritme de clustering, DBSCAN, per trobar sobredensitats en l'espai astromètric de cinc dimensions de Gaia. La implementació del mètode de clustering a un entorn de Big Data, al superordinador MareNostrum, ens permet cercar cúmuls oberts basant-nos en les seves propietats físiques. Les sobredensitats detectades s'identifiquen com a cúmuls oberts reals per mitjà d'una xarxa neuronal artificial que reconeix isòcrones en un diagrama de color-magnitud. L'automatització del procediment de detecció amb l'ús de tècniques de Big Data, ha resultat en més de 650 nous cúmuls. Aquests nous cúmul representen un terç de la població actual, i és la contribució individual més gran al catàleg. Hem pogut estimar les propietats físiques dels cúmuls com distància, edat i extinció, fent servir una xarxa neuronal artificial entrenada sobre cúmuls coneguts. Fem servir aquesta informació, juntament amb mesures de velocitat radial, per traçar l'estructura espiral actual de la nostra Galàxia associant els cúmuls oberts més joves (< 30 milions d'anys) al braç espiral on s'han format. Amb això, hem augmentat el nombre de traçadors de braços espirals, afegint 264 cúmuls joves als traçadors utilitzats tradicionalment. Això ens ha permès estimar millor els paràmetres actuals d'aquests braços. Analitzant la distribució en edat dels cúmuls dins dels braços espirals, i calculant la velocitat en la que aquests braços es mouen a partir de l'orbita dels cúmuls, hem pogut desfavorir la teoria clàssica d'ona de densitat com a mecanisme principal de formació de l'estructura espiral, trobant un comportament més transitori dels braços.
22
Evanko, Liberty Rae. "Development of an H alpha index for the detection of PMS candidates in young open clusters /". Diss., CLICK HERE for online access, 2007. http://contentdm.lib.byu.edu/ETD/image/etd1715.pdf.
23
Soergel, B., S. Flender, K. T. Story, L. Bleem, T. Giannantonio, G. Efstathiou, E. Rykoff et al. "Detection of the kinematic Sunyaev–Zel'dovich effect with DES Year 1 and SPT". OXFORD UNIV PRESS, 2016. http://hdl.handle.net/10150/621727.
Abstract (sommario):
We detect the kinematic Sunyaev-Zel'dovich (kSZ) effect with a statistical significance of 4.2 sigma by combining a cluster catalogue derived from the first year data of the Dark Energy Survey with cosmic microwave background temperature maps from the South Pole Telescope Sunyaev-Zel'dovich Survey. This measurement is performed with a differential statistic that isolates the pairwise kSZ signal, providing the first detection of the large-scale, pairwise motion of clusters using redshifts derived from photometric data. By fitting the pairwise kSZ signal to a theoretical template, we measure the average central optical depth of the cluster sample, (tau) over bar (e) = (3.75 +/- 0.89) x 10(-3). We compare the extracted signal to realistic simulations and find good agreement with respect to the signal to noise, the constraint on (tau) over bar (e), and the corresponding gas fraction. High-precision measurements of the pairwise kSZ signal with future data will be able to place constraints on the baryonic physics of galaxy clusters, and could be used to probe gravity on scales greater than or similar to 100 Mpc.
24
Dupree, A. K., A. Dotter, C. I. Johnson, A. F. Marino, A. P. Milone, J. I. Bailey, J. D. Crane, M. Mateo e E. W. Olszewski. "NGC 1866: First Spectroscopic Detection of Fast-rotating Stars in a Young LMC Cluster". IOP PUBLISHING LTD, 2017. http://hdl.handle.net/10150/625815.
Abstract (sommario):
High-resolution spectroscopic observations were taken of 29 extended main-sequence turnoff (eMSTO) stars in the young (similar to 200 Myr) Large Magellanic Cloud (LMC) cluster, NGC 1866, using the Michigan/Magellan Fiber System and MSpec spectrograph on the Magellan-Clay 6.5 m telescope. These spectra reveal the first direct detection of rapidly rotating stars whose presence has only been inferred from photometric studies. The eMSTO stars exhibit Ha emission (indicative of Be-star decretion disks), others have shallow broad H alpha absorption (consistent with rotation. greater than or similar to 150 km s(-1)), or deep Ha core absorption signaling lower rotation velocities (less than or similar to 150 km s(-1)). The spectra appear consistent with two populations of stars-one rapidly rotating, and the other, younger and slowly rotating.
25
Ramos, Ceja Miriam Elizabeth [Verfasser]. "Studying galaxy clusters through X-rays and the Sunyaev-Zel'dovich effect: simulations, detection and characterisation / Miriam Elizabeth Ramos Ceja". Bonn : Universitäts- und Landesbibliothek Bonn, 2016. http://d-nb.info/1113688289/34.
26
JUNIOR, MARCOS PAULINO RORIZ. "DG2CEP: AN ON-LINE ALGORITHM FOR REAL-TIME DETECTION OF SPATIAL CLUSTERS FROM LARGE DATA STREAMS THROUGH COMPLEX EVENT PROCESSING". PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO, 2017. http://www.maxwell.vrac.puc-rio.br/Busca_etds.php?strSecao=resultado&nrSeq=30249@1.
Abstract (sommario):
PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIROCONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICO
FUNDAÇÃO DE APOIO À PESQUISA DO ESTADO DO RIO DE JANEIRO
BOLSA NOTA 10
Clusters (ou concentrações) de objetos móveis, como veículos e seres humanos, é um padrão de mobilidade relevante para muitas aplicações. Uma detecção rápida deste padrão e de sua evolução, por exemplo, se o cluster está encolhendo ou crescendo, é útil em vários cenários, como detectar a formação de engarrafamentos ou detectar uma rápida dispersão de pessoas em um show de música. A detecção on-line deste padrão é uma tarefa desafiadora porque requer algoritmos que sejam capazes de processar de forma contínua e eficiente o alto volume de dados enviados pelos objetos móveis em tempo hábil. Atualmente, a maioria das abordagens para a detecção destes clusters operam em lote. As localizações dos objetos móveis são armazenadas durante um determinado período e depois processadas em lote por uma rotina externa, atrasando o resultado da detecção do cluster até o final do período ou do próximo lote. Além disso, essas abordagem utilizam extensivamente estruturas de dados e operadores espaciais, o que pode ser problemático em cenários de grande fluxos de dados. Com intuito de abordar estes problemas, propomos nesta tese o DG2CEP, um algoritmo que combina o conhecido algoritmo de aglomeração por densidade (DBSCAN) com o paradigma de processamento de fluxos de dados (Complex Event Processing) para a detecção contínua e rápida dos aglomerados. Nossos experimentos com dados reais indicam que o DG2CEP é capaz de detectar a formação e dispersão de clusters rapidamente, em menos de alguns segundos, para milhares de objetos móveis. Além disso, os resultados obtidos indicam que o DG2CEP possui maior similaridade com DBSCAN do que abordagens baseadas em lote.
Spatial concentrations (or spatial clusters) of moving objects, such as vehicles and humans, is a mobility pattern that is relevant to many applications. A fast detection of this pattern and its evolution, e.g., if the cluster is shrinking or growing, is useful in numerous scenarios, such as detecting the formation of traffic jams or detecting a fast dispersion of people in a music concert. An on-line detection of this pattern is a challenging task because it requires algorithms that are capable of continuously and efficiently processing the high volume of position updates in a timely manner. Currently, the majority of approaches for spatial cluster detection operate in batch mode, where moving objects location updates are recorded during time periods of certain length and then batch-processed by an external routine, thus delaying the result of the cluster detection until the end of the time period. Further, they extensively use spatial data structures and operators, which can be troublesome to maintain or parallelize in on-line scenarios. To address these issues, in this thesis we propose DG2CEP, an algorithm that combines the well-known density-based clustering algorithm DBSCAN with the data stream processing paradigm Complex Event Processing (CEP) to achieve continuous and timely detection of spatial clusters. Our experiments with real world data streams indicate that DG2CEP is able to detect the formation and dispersion of clusters with small latency while having a higher similarity to DBSCAN than batch-based approaches.
27
Júnior, Evanivaldo Castro Silva. "Modelo de processamento de imagens mamográficas para detecção de agrupamentos de microcalcificações". Universidade de São Paulo, 2009. http://www.teses.usp.br/teses/disponiveis/18/18152/tde-06052009-095239/.
Abstract (sommario):
O objetivo principal deste projeto foi desenvolver um modelo para a detecção de clusters de microcalcificações para o processamento de imagens mamográficas inteiras. O modelo foi subdividido em três etapas sendo na primeira realizado um pré-processamento para a melhoria da qualidade das imagens mamográficas no que se refere à remoção de ruídos e alargamento de contraste. Na segunda etapa do processamento, um conjunto de algoritmos foi aplicado visando-se a detecção propriamente dita de regiões de interesse nas imagens as quais possivelmente representariam os agrupamentos de microcalcificações. A terceira etapa destinou-se à classificação das regiões pré-selecionadas na etapa anterior para a determinação final dos achados verdadeiro-positivos (VP), buscando-se, assim, a diminuição da taxa de achados falso-positivos (FP). Em cada etapa do desenvolvimento do modelo, testes computacionais foram realizados a fim de auxiliarem na análise de resultados preliminares. Por fim, vários testes computacionais foram realizados em três conjuntos de imagens com composições distintas sendo o primeiro formado por regiões de interesse (RI) de phantoms, o segundo por RI de mamografias e o terceiro por imagens mamográficas inteiras. Propõe-se a integração das técnicas propostas ao sistema CAD em desenvolvimento pelo grupo de pesquisa do LAPIMO (Laboratório de Análise e Processamento de Imagens Médicas e Oftalmológicas) da Escola de Engenharia de São Carlos do presente instituto.The main purpose of this project was to develop a new model for the detection of microcalcifications clusters for image processing in full mammograms. The model was subdivided in three stages being in the first accomplished a pre-processing for the improvement of the quality of the mammographic images through the removal of noise and contrast enlargement. In the second stage of the processing, a group of algorithms was applied being sought the detection properly said of regions of interest (ROI\'s) in the images which possibly would represent the microcalcifications clusters. The third stage was destined to the classification of the pre-selected areas in the previous stage for the final determination of the true-positive findies (TP), being looked for, like this, the decrease of the rate of false-positive (FP) ones. In each stage of the development of the model, computational tests was accomplished in order to analyze the preliminary results. Finally, several computational tests was accomplished in three groups of images with different compositions being the first formed by ROI\'s of phantoms, the second by ROI\'s of mammograms and the third for full mammograms. Is proposed too the integration of the techniques proposed to the CAD scheme in development for the group of research of LAPIMO (Laboratory of Analysis and Processing of Medical and Ophthalmology Images) of the University of São Paulo, São Carlos of the present institute.
28
Matzinger, (geb Jandrasits) Christine [Verfasser]. "Computational Pan-genomics for Detection of Transmission Clusters in Molecular Surveillance with Application in the Epidemiology of Tuberculosis / Christine Matzinger (geb.Jandrasits)". Berlin : Freie Universität Berlin, 2019. http://d-nb.info/1202041744/34.
29
Aros, Pinochet Francisco Ignacio [Verfasser], e de Ven Glenn [Akademischer Betreuer] van. "Towards a robust detection of intermediate-mass black holes in globular clusters. / Francisco Ignacio Aros Pinochet ; Betreuer: Glenn van de Ven". Heidelberg : Universitätsbibliothek Heidelberg, 2021. http://d-nb.info/1237270847/34.
30
With, Meike de. "Search for neutrinos from annihilating dark matter in galaxies and galaxy clusters with the IceCube detector". Doctoral thesis, Humboldt-Universität zu Berlin, 2018. http://dx.doi.org/10.18452/19462.
Abstract (sommario):
In dieser Dissertation werden Daten aus drei Jahren vom vollständig fertiggestellten IceCube-Detektor benutzt um nach Neutrinos zu suchen, die in Dunkle-Materie-Annihilationen in fünf nahegelegen Zwerggalaxien, der M31 Galaxie und dem Virgo Galaxienhaufen produziert wurden. Um die Messung durchzuführen, wurde zunächst eine Ereignis-Selektion
angewandt, die es ermöglicht, die von aus der Atmosphäre stammenden Teilchen dominierte Rate der Ereignisse von zirka 100 Hz auf 0.5 mHz zu reduzieren. Danach wird eine Maximum-Likelihood-Schätzer eingesetzt um zu bestimmen ob es ein Überschuss von Neutrinos aus der Richtung der jeweiligen Quellen gibt, der mit einen Energie-Spektrum
übereinstimmt das mann für Dunkle-Materie-Annihilationen erwartet. Für die M31 Galaxie und den Virgo Galaxienhaufen wurde zusätzlich zu dieser Suche nach einer Punktquelle auch eine Suche für ein erweitertes Signal durchgeführt. In allen untersuchten Fällen ist das Ergebnis der Analyse vereinbar mit einer Messung der Hintergrund-Hypothese, und daraus wurden Limits für den über die Geschwindigkeit gemittelten Wirkungsquerschnitt für Dunkle-Materie-Annihilation für verschiedene Endprodukte bestimmt. Für hohe Dunkle-Materie-Massen gibt es ein Überschuss von Neutrinos aus drei der Zwerggalaxien. Dieser Überschuss hat einen globalen p-Wert von 4.9% und ist damit nicht statistisch signifikant. Die Suche für ein erweitertes Signal von der M31 Galaxie und dem Virgo Galaxienhaufen ergab keinen Überschuss. Die Limits auf den über die Geschwindigkeit gemittelten Wirkungsquerschnitt für Dunkle-Materie-Annihilation haben sich im Vergleich zu vorherigen IceCube-Analysen signifikant verbessert, um bis zu einer Größenordnung. Diese ist teilweise auf Grund Verbesserungen für diese Analyse besonders: eine verbesserte Ereignis-Selektion, und für die Analyse ist eine Maximum-Likelihood-Schätzer eingesetzt statt eine Analyse in ein Suchfenster.In this thesis, three years worth of data from the completed IceCube detector is used to search for neutrinos produced in dark matter annihilations in five nearby dwarf galaxies, the M31 galaxy and the Virgo cluster. To do this, an event selection which was developed for this analysis is applied to the data sample to reduce the atmospheric background rate from approximately 100 Hz to less than 0.5 mHz. Then, an unbinned maximum likelihood method is used to determine whether there is an excess of neutrinos from the direction of the considered galaxies or galaxy cluster that has an energy spectrum that matches the spectrum expected from dark matter annihilations. For the M31 galaxy and the Virgo cluster an extended signal with a two-dimensional Gaussian shape and width up to 5 degrees is also considered. In all cases, the results of the analysis are compatible with the background-only hypothesis and limits are set on the velocity-averaged dark matter annihilation cross section for different annihilation channels. For high dark matter masses there is an excess of neutrinos from three of the five dwarf galaxies. This excess has a global p-value of 4.9%, so it is not statistically significant. The search for an extended emission from the direction of the M31 galaxy and the Virgo cluster also did not result in an excess: in both cases the global p-value is larger than 50%. The limits on the velocity-averaged dark matter annihilation cross section have improved significantly (up to an order of magnitude) with respect to the previous IceCube analysis considering these same targets. This is partially due to improvements to this analysis specifically: an improved event selection was used to select the final data sample and an unbinned maximum likelihood method was used for the final analysis instead of a binned analysis method.
31
Hofmann, Florian [Verfasser], e Kirpal [Akademischer Betreuer] Nandra. "Turbulence and direct dark matter detection in the X-ray halo of galaxy clusters : implications for eROSITA / Florian Hofmann ; Betreuer: Kirpal Nandra". München : Universitätsbibliothek der Ludwig-Maximilians-Universität, 2016. http://d-nb.info/1120302129/34.
32
Guttmann, Aline. "Evaluation des méthodes statistiques en épidémiologie spatiale : cas des méthodes locales de détection d'agrégats". Thesis, Clermont-Ferrand 1, 2014. http://www.theses.fr/2014CLF1MM21/document.
Abstract (sommario):
L'évaluation des performances des méthodes de détection d'agrégats de maladie est fondamentale dans le domaine de l'épidémiologie spatiale et, paradoxalement, on déplore une absence de consensus quant à sa conduite. Cette problématique est d'autant plus importante que les nouvelles technologies de partage d'informations promettent une évolution importante des signaux disponibles pour l'épidémiologie et la veille sanitaire. Les spécialistes du domaine ont adopté un mode d'évaluation fondé sur l'utilisation concomitante de plusieurs indicateurs de performances complémentaires tels que des indicateurs dérivés de l'évaluation des méthodes diagnostiques ou encore diverses définitions de puissance conditionnelle. Cependant, ces évaluations issues de schémas de simulation classiques reposent sur le choix de quelques hypothèses alternatives particulières et ne permettent qu'une interprétation limitée à ces hypothèses. De plus, la démultiplication des indicateurs évaluant la performance, différents selon les protocoles, gêne la comparaison des études entres elles et complique l'interprétation des résultats. Notre travail propose et évalue plusieurs indicateurs de performance prenant en compte à la fois puissance et précision de localisation. Leur intérêt dans l'évaluation spatiale systématique des méthodes est illustré par la création de cartes de performance. En complément de l'évaluation des performances lorsqu'une détection est attendue, nous proposons également une méthode d'évaluation de la répartition spatiale de l'erreur de type I complétée par la construction d'une nouvelle inférence statistique testant l'éventualité d'un effet de bordAlthough performance assessment of cluster detection tests is a critical issue in spatial epidemiology, there is a lack of consensus regarding how it should be carried out. Nowadays, with the spread of new technologies in network systems, data sources for epidemiology are undergoing radical changes that will increase the need for performance evaluation. Field specialists are currently evaluating cluster detection tests with multiple complementary performance indicators such as conditional powers or indicators derived from the field of diagnostic tools evaluation. These evaluations are performed following classical protocols for power assessment and are often limited to a few number of simulated alternative hypotheses, thus restricting results interpretation and scope. Furthermore, with the use of multiple varying indicators, comparisons between studies is difficult at best. This work proposes and compares different global performance indicators that take into account both usual power and location accuracy. Their benefit for cluster detection tests evaluation is illustrated with a systematic spatial assessment enabling performance mapping. In addition to the evaluation of performance when clusters exist, we also propose a method for the spatial evaluation of type I error, together with a new statistical test for edge effect
33
Wen, Shihua. "Semiparametric cluster detection". College Park, Md. : University of Maryland, 2007. http://hdl.handle.net/1903/7204.
Abstract (sommario):
Thesis (Ph. D.) -- University of Maryland, College Park, 2007.Thesis research directed by: Mathematical Statistics. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also available in paper.
34
Hill, Evelyn June. "Applying statistical and syntactic pattern recognition techniques to the detection of fish in digital images". University of Western Australia. School of Mathematics and Statistics, 2004. http://theses.library.uwa.edu.au/adt-WU2004.0070.
Abstract (sommario):
This study is an attempt to simulate aspects of human visual perception by automating the detection of specific types of objects in digital images. The success of the methods attempted here was measured by how well results of experiments corresponded to what a typical human’s assessment of the data might be. The subject of the study was images of live fish taken underwater by digital video or digital still cameras. It is desirable to be able to automate the processing of such data for efficient stock assessment for fisheries management. In this study some well known statistical pattern classification techniques were tested and new syntactical/ structural pattern recognition techniques were developed. For testing of statistical pattern classification, the pixels belonging to fish were separated from the background pixels and the EM algorithm for Gaussian mixture models was used to locate clusters of pixels. The means and the covariance matrices for the components of the model were used to indicate the location, size and shape of the clusters. Because the number of components in the mixture is unknown, the EM algorithm has to be run a number of times with different numbers of components and then the best model chosen using a model selection criterion. The AIC (Akaike Information Criterion) and the MDL (Minimum Description Length) were tested.The MDL was found to estimate the numbers of clusters of pixels more accurately than the AIC, which tended to overestimate cluster numbers. In order to reduce problems caused by initialisation of the EM algorithm (i.e. starting positions of mixtures and number of mixtures), the Dynamic Cluster Finding algorithm (DCF) was developed (based on the Dog-Rabbit strategy). This algorithm can produce an estimate of the locations and numbers of clusters of pixels. The Dog-Rabbit strategy is based on early studies of learning behaviour in neurons. The main difference between Dog-Rabbit and DCF is that DCF is based on a toroidal topology which removes the tendency of cluster locators to migrate to the centre of mass of the data set and miss clusters near the edges of the image. In the second approach to the problem, data was extracted from the image using an edge detector. The edges from a reference object were compared with the edges from a new image to determine if the object occurred in the new image. In order to compare edges, the edge pixels were first assembled into curves using an UpWrite procedure; then the curves were smoothed by fitting parametric cubic polynomials. Finally the curves were converted to arrays of numbers which represented the signed curvature of the curves at regular intervals. Sets of curves from different images can be compared by comparing the arrays of signed curvature values, as well as the relative orientations and locations of the curves. Discrepancy values were calculated to indicate how well curves and sets of curves matched the reference object. The total length of all matched curves was used to indicate what fraction of the reference object was found in the new image. The curve matching procedure gave results which corresponded well with what a human being being might observe.
35
Engelin, Martin, e Silva Felix De. "Troll detection : A comparative study in detecting troll farms on Twitter using cluster analysis". Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-186406.
Abstract (sommario):
The purpose of this research is to test whether clustering algorithmscan be used to detect troll farms in social networks. Troll farms are profes-sional organizations that spread disinformation online via fake personas.The research involves a comparative study of two different clustering algo-rithms and a dataset of Twitter users and posts that includes a fabricatedtroll farm. By comparing the results and the implementations of the K-means as well as the DBSCAN algorithm we have concluded that clusteranalysis can be used to detect troll farms and that DBSCAN is bettersuited for this particular problem compared to K-means.Målet med denna rapport är att testa om klusteringalgoritmer kananvändas för att identifiera trollfarmer på sociala medier. Trollfarmer ärprofessionella organisationer som sprider desinformation online med hjälpav falska identiteter. Denna rapport är en jämförande studie med två olikaklusteringalgoritmer och en datamängd av Twitteranvändare och tweetssom inkluderar en fabrikerad trollfarm. Genom att jämföra resultaten ochimplementationerna av algoritmerna K-means och DBSCAN får vi framslutsatsen att klusteralgoritmer kan användas för att identifiera trollfar-mar och att DBSCAN är bättre lämpad för detta problem till skillnadfrån K-means.
36
Sreenivasulu, Ajay. "Evaluation of cluster based Anomaly detection". Thesis, Högskolan i Skövde, Institutionen för informationsteknologi, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-18053.
Abstract (sommario):
Anomaly detection has been widely researched and used in various application domains such as network intrusion, military, and finance, etc. Anomalies can be defined as an unusual behavior that differs from the expected normal behavior. This thesis focuses on evaluating the performance of different clustering algorithms namely k-Means, DBSCAN, and OPTICS as an anomaly detector. The data is generated using the MixSim package available in R. The algorithms were tested on different cluster overlap and dimensions. Evaluation metrics such as Recall, precision, and F1 Score were used to analyze the performance of clustering algorithms. The results show that DBSCAN performed better than other algorithms when provided low dimensional data with different cluster overlap settings but it did not perform well when provided high dimensional data with different cluster overlap. For high dimensional data k-means performed better compared to DBSCAN and OPTICS with different cluster overlaps
37
Mazumdar, Soumya. "Shape and scale in detecting disease clusters". Diss., University of Iowa, 2008. https://ir.uiowa.edu/etd/208.
Abstract (sommario):
This dissertation offers a new cluster detection method. This method looks at the cluster detection problem from a new perspective. I change the question of "What do real clusters look like?" to the question of "What do spurious clusters look like?" and "How do spurious clusters affect the ability to recover real clusters?" Spurious clusters can be identified from their geographical characteristics. These are related to the spatial distribution of people at risk, the shape and scale of the geographic units used to aggregate these people, the shape and scale of the spatial configurations that the disease mapping or cluster detection method may impose on the data and the shape and scale of the area of increased risk. The statistical testing process may also create spurious clusters. I propose that the problem of spurious clusters can be resolved using a computational geographic approach. I argue that Monte Carlo simulations can be used to estimate the patterns of spurious clusters in any situation of interest given knowledge of the first three of these four determinants of spurious clusters. Then, given these determinants, where real measurements of disease or mortality are known, it is possible to show those areas of increased risk that are true clusters as opposed to those that are spurious clusters. The extent of similarity (or dissimilarity) of a cluster to the simulated spurious cluster influences whether it can be recovered. These experiments show that this method is successful in detecting clusters. This method can also predict with reasonable certainty which clusters can be recovered, and which cannot. I compare this method with Rogerson's Score statistic method. These comparisons expose the weaknesses of Rogerson's method. Finally these two methods and the Spatial Scan Statistic are applied to searching for possible clusters of prostate cancer incidence in Iowa. The implications of the findings are discussed.
38
Soon, Shih Chung. "On detection of extreme data points in cluster analysis". Connect to resource, 1987. http://rave.ohiolink.edu/etdc/view.cgi?acc%5Fnum=osu1262886219.
39
Alodat, Moh'd. "Detecting conjunctions using cluster volumes". Thesis, McGill University, 2004. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=84981.
Abstract (sommario):
In brain mapping, the regions of the brain that are 'activated' by a task or external stimulus are detected by thresholding an image of test statistics. Often the experiment is repeated on several different subjects or for several different stimuli on the same subject, and the researcher is interested in the common points in the brain where 'activation' occurs in all test statistic images. The conjunction is thus defined as those points in the brain that show 'activation' in all images. We are interested in which parts of the conjunction are noise, and which show true activation in all test statistic images. We would expect truly activated regions to be larger than usual, so our test statistic is based on the volume of clusters (connected components) of the conjunction. Our main result is an approximate P-value for this in the case of the conjunction of two Gaussian or chi2 test statistic images. The results are applied to a functional magnetic resonance experiment in pain perception.
40
Svedberg, Oskar. "Automatic detection of ULF waves in Cluster data". Thesis, KTH, Rymd- och plasmafysik, 2007. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-91550.
41
Bozkus, Nebahat. "Cluster detection by lifting with application to phylogenetics". Thesis, University of Leeds, 2018. http://etheses.whiterose.ac.uk/21300/.
Abstract (sommario):
In this thesis, we propose a new algorithm which automatically detects the number of clusters in a tree structure data set by denoising some generalized node values in the tree using lifting “one coefficient at a time” (LOCAAT) algorithm introduced by Jansen et al. (2001). Our algorithm can be applied to any multidimensional data set using compactness value as a node value or to phylogenetic data sets, DNA sequences, using either compactness value or dissimilarity score as a node value. Compactness value is defined as the average distance from the centroid of each possible cluster in the tree, and the dissimilarity score is the average number of loci, where at least one of them does not share the same nucleotide between sequences under the node of interest. For multidimensional data sets, we consider each node in the tree as a possible location of a cluster after denoising the tree by LOCAAT. Thus, for each possible cluster, we check how much departure we can allow from the centroid of the cluster to assign the objects under the node of interest as a cluster. Then if a node and all its child nodes are denoised less than or equal to the allowed amount of departure from the centroid of their clusters, a cluster is located at this node. We also propose another version of our algorithm based on non-decimated lifting (Knight & Nason, 2009) in which we generate a probability of being clustered for each node. If a node and all its child nodes have a probability of being clustered less than or equal to the probability of acceptance, θ∈[0; 1], a cluster is located at this node. We provide a comparison study between our algorithms and some available internal cluster validity indices (CVIs) in the literature using some artificial data sets and a real data set. In addition, we compare the performance of each method using some available external cluster validity scores. For phylogenetic data sets, we check the performance of our algorithms and other CVIs using both compactness value and dissimilarity score as a node value. To be able to compute compactness value for a phylogenetic tree, we need to find the position of each specie in Rp using multidimensional scaling (MDS), and then we can find which species share the similar features using our algorithm. If we use the dissimilarity score as a node value, we will cluster similar species together by finding how much difference we can allow between species. We check the performance of our algorithms using some artificial and a real data sets. In the final part of our thesis, we propose a visualization tool for cophylogenetic data sets. We only consider the associated two phylogenetic trees case, and we apply our algorithm to both host and parasite trees separately to provide a summary of these data sets. We check the performance of our algorithm using two well-known cophylogenetic data sets.
42
Licitra, Rossella. "Galaxy cluster detection with optical and infrared imaging". Paris 7, 2014. http://www.theses.fr/2014PA077149.
Abstract (sommario):
En tant que structures gravitationnellement liées les plus massives, les amas de galaxies permettent de poser des contraintes fortes sur les structures à grande échelle prédites par le modèle cosmologique standard. Ils permettent aussi de comprendre l'influence de l'environnement sur l'évolution des galaxies. Pour mener ce type d'étude et obtenir des résultats robustes, il est impératif de construire des catalogues d'amas complets et purs. Dans le présent manuscrit, je décris l'algorithme de détection d'amas que j'ai développé lors de ma thèse de doctorat -Red GOLD ainsi que les résultats que j'ai obtenus en l'appliquant aux relevés multi-longueur d'onde. Mon algorithme est fondé sur la détection de surdensités de galaxies et la caractérisation de leur séquence rouge : il détecte les surdensités de galaxies rouges par rapport à la distribution moyenne des galaxies. Je sélectionne les galaxies rouges à l'aide des couleurs prédites par les modèles de population stellaire, en imposant des coupes en couleur en fonction du redshift. Parmi ces galaxies, j'identifie celles ayant un type spectral correspondant à des galaxies de type précoce. J'ai appliqué Red-GOLD à des données dans le visible venant de deux relevés diférents, le Next Generation Virgo Cluster Survey (NGVS) et le Canada-France-Hawaii Telescope Lensing Survey (CFHTLS) et j'ai détecté des candidats amas jusqu'à z~1. J'ai estimé les performances de mon algorithme en l'appliquant aux catalogues de galaxies simulées issus des simulations Millenium. Mon catalogue d'amas est complet à ~ 80 % jusqu'à z=1 et pur à 81%Being galaxy clusters the most massive bound structures in the Universe, they represent a powerful tool to probe the large-scale structure predicted by the standard cosmological model, and to understand how environmental effects affect galaxy evolution. To conduct these studies and obtain reliable results, it is important to build complete and pure cluster catalogs. The use of these catalogs for cosmology requires accurate estimates of cluster mass. In this work, I describe the cluster detection algorithm that I developed during my PhD thesis : Red-GOLD, and the results that I obtained by applying i to current multi-wavelength surveys. My algorithm is based on the detection of galaxy overdensities and the characterisation of their red-sequence. The algorithm finds red galaxy overdensities with respect to the mean background. I select red galaxies using color predictions given by stellar population synthesis models and impose color limits as a function of redshift. Among those galaxies, I discern the early-type galaxies from their spectral type. I then identify cluster members using accurate photometric redshifts, and estimate the cluster candidate richness. I applied Red-GOLD to optical data coming from two different surveys, the Next Generatiôn Virgo Cluster Survey (NGVS) and the Canada-France-Hawaii Telescope Lensing Survey (CFHTLS) and detected galaxy cluster candidates up to redshift z=1. I assessed the performances of my algorithm by applying it to simulated galaxy catalogs from the Millennium simulations. My cluster catalogue is complete at the 80% up to redshift z=1 and pure at 81%
43
Bigdeli, Elnaz. "Incremental Anomaly Detection Using Two-Layer Cluster-based Structure". Thesis, Université d'Ottawa / University of Ottawa, 2016. http://hdl.handle.net/10393/34299.
Abstract (sommario):
Anomaly detection algorithms face several challenges, including processing
speed and dealing with noise in data. In this thesis, a two-layer cluster-
based anomaly detection structure is presented which is fast, noise-resilient
and incremental. In this structure, each normal pattern is considered as
a cluster, and each cluster is represented using a Gaussian Mixture Model
(GMM). Then, new instances are presented to the GMM to be labeled as
normal or abnormal.
The proposed structure comprises three main steps. In the first step, the
data are clustered. The second step is to represent each cluster in a way
that enables the model to classify new instances. The Summarization based
on Gaussian Mixture Model (SGMM) proposed in this thesis represents each
cluster as a GMM.
In the third step, a two-layer structure efficiently updates clusters using
GMM representation while detecting and ignoring redundant instances. A
new approach, called Collective Probabilistic Labeling (CPL) is presented
to update clusters in a batch mode. This approach makes the updating
phase noise-resistant and fast. The collective approach also introduces a new
concept called 'rag bag' used to store new instances. The new instances
collected in the rag bag are clustered and summarized by GMMs. This
enables online systems to identify nearby clusters in the existing and new
clusters, and merge them quickly, despite the presence of noise to update the model.
An important step in the updating is the merging of new clusters with ex-
isting ones. To this end, a new distance measure is proposed, which is a mod-
i ed Kullback-Leibler distance between two GMMs. This modi ed distance
allows accurate identi cation of nearby clusters. After finding neighboring
clusters, they are merged, quickly and accurately. One of the reasons that
GMM is chosen to represent clusters is to have a clear and valid mathematical
representation for clusters, which eases further cluster analysis.
In most real-time anomaly detection applications, incoming instances are
often similar to previous ones. In these cases, there is no need to update
clusters based on duplicates, since they have already been modeled in the
cluster distribution. The two-layer structure is responsible for identifying
redundant instances. In this structure, redundant instance are ignored, and
the remaining new instances are used to update clusters. Ignoring redundant
instances, which are typically in the majority, makes the detection phase fast.
Each part of the general structure is validated in this thesis. The experiments include, detection rates, clustering goodness, time, memory usage
and the complexity of the algorithms. The accuracy of the clustering and
summarization of clusters using GMMs is evaluated, and compared to that of
other methods. Using Davies-Bouldin (DB) and Dunn indexes, the distances
for original and regenerated clusters using GMMs is almost zero with SGMM
method while this value for ABACUS is around 0:01. Moreover, the results
show that the SGMM algorithm is 3 times faster than ABACUS in running
time, using one-third of the memory used by ABACUS.
The CPL method, used to label new instances, is found to collectively
remove the effect of noise, while increasing the accuracy of labeling new
instances. In a noisy environment, the detection rate of the CPL method
is 5% higher than other algorithms such as one-class SVM. The false alarm rate is decreased by 10% on average. Memory use is 20 times lesser that that
of the one-class SVM.
The proposed method is found to lower the false alarm rate, which is
one of the basic problems for the one-class SVM. Experiments show the false
alarm rate is decreased from 5% to 15% among different datasets, while the
detection rate is increased from 5% to 10% in di erent datasets with two-
layer structure. The memory usage for the two-layer structure is 20 to 50
times less than that of one-class SVM. One-class SVM uses support vectors in
labeling new instances, while the labeling of the two-layer structure depends
on the number of GMMs. The experiments show that the two-layer structure
is 20 to 50 times faster than the one-class SVM in labeling new instances.
Moreover, the updating time of two-layer structure is 2 to 3 times less than
one-layer structure. This reduction is the direct result of ignoring redundant
instances and using two-layer structure.
44
Schäfer, Björn Malte. "Methods for detecting and characterising clusters of galaxies". Diss., lmu, 2005. http://nbn-resolving.de/urn:nbn:de:bvb:19-40652.
45
Brolin, Morgan, e Erik Ledin. "Detecting trolls on twitterthrough cluster analysis". Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-208354.
Abstract (sommario):
The social media platform Twitter is designed to allow users to efficiently spread informationthrough short messages that are broadcast to the world. The efficient way to spreadinformation that is in no way controlled or edited brings inherent problems with the spreadingof misinformation and other malicious activity as it can often be very difficult to establishwhat information can be considered reliable. This study seeks to showcase these problemsas well as find out whether it is possible to identify these malicious users by filtering tweetsby keywords, clustering the tweets based on similarity and analyzing these clusters alongwith user data such as amount of followers, number of accounts followed, and geolocationbeing turned off. The tweets were gathered using the Twitter streaming API and theclustering was done through the use of k-means clustering using a tf-idf approach.Approximately 2000 tweets were gathered for every keyword, and roughly 4000 using nofilter, to allow us to discern which topics contain higher and lower percentages of likely trollsor malicious users. The results show that highly political and controversial topics such as“ISIS”, “Russia”, and “Putin” have significantly higher percentages of likely trolls andmalicious users when compared to tweets that are not filtered by any keyword, which in turnhas higher amounts than more neutral keywords such as “cat”, “happy” and “car”. Howeverthe results also show that it would be very difficult to use clustering alone to find trolls ormalicious users, and that the analysis of user data does not paint a complete picture andmay give both false positives as well as false negatives. However clustering in combinationwith other techniques such as user data analysis can be used to successfully analyze howmalicious users are spread through different topics on Twitter.Den sociala nätverkstjänsten Twitter är utformad för att låta användare effektivt och snabbtsprida information via korta meddelanden som sänds ut till världen. Denna typ av effektivaspridning av information som inte kontrolleras eller redigeras bär med sig problem i formenav spridning av misinformation och annan skadlig aktivitet, då det kan vara mycket svårt attsäkerställa vilken information som är pålitlig. Denna studie försöker klargöra dessa problemoch ta reda på om det är möjligt att identifiera dessa skadliga användare genom att filtreratweets på nyckelord, klustra dessa tweets baserat på likhet och analysera klustren isamband med användardata såsom antal följare, antal konton följda och att geolocation äravstängt. Tweetsen hämtades med hjälp av Twitters streaming API och klustringen gjordesmed tf-idf k-means clustering. Uppskattningsvis 2000 tweets hämtades för varje nyckelord,och cirka 4000 ofiltrerade tweets, för att möjliggöra att skilja på vilka ämnen som har störreoch mindre andelar potentiellt skadliga användare. Resultaten visar på att politiska ochkontroversiella ämnen såsom “ISIS”, “Ryssland” och “Putin” har märkbart högre andelarpotentiellt skadliga användare, jämfört med tweets som inte filtrerats baserat på någotnyckelord, vilka i sin tur har högre andelar än mer neutrala nyckelord såsom “cat”, “happy”och “car”. Resultaten tyder på att det är svårt att enbart använda klustring för att hittaskadliga användare och att analysen av användardata inte alltid visar den hela bilden ochkan ge felaktiga resultat åt båda håll. Trots det kan klustring i kombination med andratekniker såsom data analys användas för att analysera hur skadliga användare är spriddagenom olika ämnen på twitter.
46
Schmidt, Thomas. "Efficient algorithms for gene cluster detection in prokaryotic genomes". [S.l.] : [s.n.], 2005. http://deposit.ddb.de/cgi-bin/dokserv?idn=976473356.
47
Dodoo, Nii Lartey 1977. "Selecting predicates for conditional invariant detection using cluster analysis". Thesis, Massachusetts Institute of Technology, 2002. http://hdl.handle.net/1721.1/87425.
Abstract (sommario):
Thesis (M.Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2002.Includes bibliographical references (p. 59-61).
by Nii Lartey Dodoo.
M.Eng.
48
Liu, Zhen. "A lightweight intrusion detection system for the cluster environment". Master's thesis, Mississippi State : Mississippi State University, 2003. http://sun.library.msstate.edu/ETD-db/theses/available/etd-07102003-152642/unrestricted/ZhenLiu%5Fthesis.pdf.
49
Buttery, H. J. "New methods for detecting high-redshift clusters of galaxies". Thesis, University of Cambridge, 2004. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.597196.
Abstract (sommario):
The detection of high-redshift clusters of galaxies is important in under-standing the evolution of our Universe. Computer simulations, assuming hierarchical structure formation within a Ωλ » 0.7, ΩM » 0.3 universe, predict the galaxy clusters should be detected out to redshifts of z = 1. Traditional methods for finding clusters of galaxies, for example using optical plates, are not well to cluster detection beyond a redshift of z = 0.3, because they suffer from contamination of foreground sources. A new method of searching for galaxy clusters has been devised. It uses the hypothesis that at high redshifts radiosources trace high-density regions of our Universe. This would imply that high-redshift groupings of radiosources would be preferentially found in the high-density environments of galaxy clusters. I have used the Sydney University Molonglo Sky Survey (SUMSS) to search for groupings of five radiosources within a seven-arcminute circle. In this thesis I present the work extracting 60 cluster candidates from SUMSS. I also present the radio, infrared and optical follow-up observations that were undertaken and the implications of these. This thesis also discusses the Sunyaev Zel’dovich Effect (SZE), which is the inverse-Compton scattering of the Cosmic Microwave Background (CMB) by a plasma. This is another new method for detecting high-redshift clusters of galaxies. It is particularly important because the magnitude of the effect is independent of redshift. I include the observations of a galaxy cluster where this effect is apparent carried out at the Ryle Telescope (RT) in Cambridge.
50
Cau, Massimo <1968>. "New detections and statistics of diffuse radio sources in galaxy clusters". Doctoral thesis, Alma Mater Studiorum - Università di Bologna, 2019. http://amsdottorato.unibo.it/9066/1/Cau_Massimo_tesi.pdf.
Abstract (sommario):
Galaxy clusters are the largest gravitationally bound structures of the Universe, formed from density fluctuations and grown hierarchically through the extreme process of merging and mass accretion. They represent very interesting tools to study the cosmology and the evolution of large scale structures. Synchrotron non-thermal emission detected in the forms of radio halos, relics and minihalos according to their morphology, size and location is linked to the different dynamical state of the hosting clusters. Halos and relics are detected in clusters characterized by a strong merger activity and a dynamical disturbed state, while minihalos are present only in relaxed cool-core clusters.
In spite of many encouraging results obtained up to now, the occurrence and the luminosity function of diffuse radio sources with the redshift are still unknown: present data are strongly limited to nearby clusters (z < 0.2 − 0.3).
The aim of this PhD thesis is mainly to investigate the evolutionary history of non-thermal properties of galaxy clusters and to determine whether the correlations observed at low redshift evolve with time. To attempt this ambitious goal we selected an homogeneous sample of 44 massive and high X-ray luminous galaxy clusters in the redshift range
0.3 ≤ z < 0.7, extracted from the Ebeling MAssive Cluster Survey (MACS). We have undertaken an observational campaign on this sample with the JVLA in L-band, C and D configurations.