Journal articles on the topic 'Black Box trees'
To see the other types of publications on this topic, follow the link: Black Box trees.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Black Box trees.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Veugen, Thijs, Bart Kamphorst, and Michiel Marcus. "Privacy-Preserving Contrastive Explanations with Local Foil Trees." Cryptography 6, no. 4 (October 28, 2022): 54. http://dx.doi.org/10.3390/cryptography6040054.
Full textAbstract:
We present the first algorithm that combines privacy-preserving technologies and state-of-the-art explainable AI to enable privacy-friendly explanations of black-box AI models. We provide a secure algorithm for contrastive explanations of black-box machine learning models that securely trains and uses local foil trees. Our work shows that the quality of these explanations can be upheld whilst ensuring the privacy of both the training data and the model itself.
2
Stone, C., and PE Bacon. "Influence of Insect Herbivory on the Decline of Black Box (Eucalyptus largiflorens)." Australian Journal of Botany 43, no. 6 (1995): 555. http://dx.doi.org/10.1071/bt9950555.
Full textAbstract:
The contribution of insect herbivory to the canopy decline of Eucalyptus largiflorens F.Muell. (black box) was assessed on nine irrigated properties around Deniliquin in southern central New South Wales. Fully expanded leaves less than 1 year old were sampled from 36 mature trees in June 1993 and again in June 1994 after half the trees had been treated with a systemic insecticide in November 1993. Insect herbivory in treated trees fell significantly from 27 to 9%. It also fell, but to a lesser extent (28-19%, P < 0.05), in the untreated trees. The fall in insect herbivory in control trees corresponded to a decrease in rainfall in 1994 when the rainfall was 50% of that for 1993. There was a significant linear relationship between insect herbivory and trunk diameter increment in the untreated trees. There was no consistent relationship between insect herbivory and the visual assessment of crown condition. Although E. largiflorens is described as having both narrow adult and juvenile foliage, adjacent trees in this study differed significantly in their leaf length:breadth ratios. Canopies with a dominance of broader foliage had significantly higher levels of herbivory. Individual trees tended to replace foliage with leaves of similar morphology. It is suggested that this variation in leaf shape may be genetic rather than environmental. If so, landholders could select for trees with narrower foliage which may result in reduced impact of insect herbivory.
3
McTavish, Hayden, Chudi Zhong, Reto Achermann, Ilias Karimalis, Jacques Chen, Cynthia Rudin, and Margo Seltzer. "Fast Sparse Decision Tree Optimization via Reference Ensembles." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 9 (June 28, 2022): 9604–13. http://dx.doi.org/10.1609/aaai.v36i9.21194.
Full textAbstract:
Sparse decision tree optimization has been one of the most fundamental problems in AI since its inception and is a challenge at the core of interpretable machine learning. Sparse decision tree optimization is computationally hard, and despite steady effort since the 1960's, breakthroughs have been made on the problem only within the past few years, primarily on the problem of finding optimal sparse decision trees. However, current state-of-the-art algorithms often require impractical amounts of computation time and memory to find optimal or near-optimal trees for some real-world datasets, particularly those having several continuous-valued features. Given that the search spaces of these decision tree optimization problems are massive, can we practically hope to find a sparse decision tree that competes in accuracy with a black box machine learning model? We address this problem via smart guessing strategies that can be applied to any optimal branch-and-bound-based decision tree algorithm. The guesses come from knowledge gleaned from black box models. We show that by using these guesses, we can reduce the run time by multiple orders of magnitude while providing bounds on how far the resulting trees can deviate from the black box's accuracy and expressive power. Our approach enables guesses about how to bin continuous features, the size of the tree, and lower bounds on the error for the optimal decision tree. Our experiments show that in many cases we can rapidly construct sparse decision trees that match the accuracy of black box models. To summarize: when you are having trouble optimizing, just guess.
4
Welchowski, Thomas, Kelly O. Maloney, Richard Mitchell, and Matthias Schmid. "Techniques to Improve Ecological Interpretability of Black-Box Machine Learning Models." Journal of Agricultural, Biological and Environmental Statistics 27, no. 1 (October 28, 2021): 175–97. http://dx.doi.org/10.1007/s13253-021-00479-7.
Full textAbstract:
AbstractStatistical modeling of ecological data is often faced with a large number of variables as well as possible nonlinear relationships and higher-order interaction effects. Gradient boosted trees (GBT) have been successful in addressing these issues and have shown a good predictive performance in modeling nonlinear relationships, in particular in classification settings with a categorical response variable. They also tend to be robust against outliers. However, their black-box nature makes it difficult to interpret these models. We introduce several recently developed statistical tools to the environmental research community in order to advance interpretation of these black-box models. To analyze the properties of the tools, we applied gradient boosted trees to investigate biological health of streams within the contiguous USA, as measured by a benthic macroinvertebrate biotic index. Based on these data and a simulation study, we demonstrate the advantages and limitations of partial dependence plots (PDP), individual conditional expectation (ICE) curves and accumulated local effects (ALE) in their ability to identify covariate–response relationships. Additionally, interaction effects were quantified according to interaction strength (IAS) and Friedman’s $$H^2$$ H 2 statistic. Interpretable machine learning techniques are useful tools to open the black-box of gradient boosted trees in the environmental sciences. This finding is supported by our case study on the effect of impervious surface on the benthic condition, which agrees with previous results in the literature. Overall, the most important variables were ecoregion, bed stability, watershed area, riparian vegetation and catchment slope. These variables were also present in most identified interaction effects. In conclusion, graphical tools (PDP, ICE, ALE) enable visualization and easier interpretation of GBT but should be supported by analytical statistical measures. Future methodological research is needed to investigate the properties of interaction tests. Supplementary materials accompanying this paper appear on-line.
5
Barbosa, Pedro, Astrid Caldas, and Gaden Robinson. "Host Plant Associations among Species in Two Macrolepidopteran Assemblages." Journal of Entomological Science 38, no. 1 (January 1, 2003): 41–47. http://dx.doi.org/10.18474/0749-8004-38.1.41.
Full textAbstract:
Host plant associations of macrolepidopteran species in assemblages on box elder, Acer negundo L., and black willow, Salix nigra (Marsh), were characterized. Almost 90% of the macrolepidoptera collected on these two riparian tree species of the mid-Atlantic area of the United States were new host records. Larvae of 87 species (and another nine specimens identified to genus) were collected on box elder and black willow. About one-fifth of the species were found exclusively on box elder, one-third exclusively on black willow, and about one-half of the macrolepidopteran species were found on both tree species. Although many macrolepidopterawere found on both tree species, they were not equally abundant on both trees, suggesting a predominantly favored tree species. However, there was no statistically significant asymmetry in host tree species use.
6
Wagers, Steven, Guillermo Castilla, Michelle Filiatrault, and G. Arturo Sanchez-Azofeifa. "Using TLS-Measured Tree Attributes to Estimate Aboveground Biomass in Small Black Spruce Trees." Forests 12, no. 11 (November 4, 2021): 1521. http://dx.doi.org/10.3390/f12111521.
Full textAbstract:
Research Highlights: This study advances the effort to accurately estimate the biomass of trees in peatlands, which cover 13% of Canada’s land surface. Background and Objectives: Trees remove carbon from the atmosphere and store it as biomass. Terrestrial laser scanning (TLS) has become a useful tool for modelling forest structure and estimating the above ground biomass (AGB) of trees. Allometric equations are often used to estimate individual tree AGB as a function of height and diameter at breast height (DBH), but these variables can often be laborious to measure using traditional methods. The main objective of this study was to develop allometric equations using TLS-measured variables and compare their accuracy with that of other widely used equations that rely on DBH. Materials and Methods: The study focusses on small black spruce trees (<5 m) located in peatland ecosystems of the Taiga Plains Ecozone in the Northwest Territories, Canada. Black spruce growing in peatlands are often stunted when compared to upland black spruce and having models specific to them would allow for more precise biomass estimates. One hundred small trees were destructively sampled from 10 plots and the dry weight of each tree was measured in the lab. With this reference data, we fitted biomass models specific to peatland black spruce using DBH, crown diameter, crown area, height, tree volume, and bounding box volume as predictors. Results: Our best models had crown size and height as predictors and outperformed established AGB equations that rely on DBH. Conclusions: Our equations are based on predictors that can be measured from above, and therefore they may enable the plotless creation of accurate biomass reference data for a prominent tree species in a common ecosystem (treed peatlands) in North America’s boreal.
7
Shahpouri, Saeid, Armin Norouzi, Christopher Hayduk, Reza Rezaei, Mahdi Shahbakhti, and Charles Robert Koch. "Hybrid Machine Learning Approaches and a Systematic Model Selection Process for Predicting Soot Emissions in Compression Ignition Engines." Energies 14, no. 23 (November 24, 2021): 7865. http://dx.doi.org/10.3390/en14237865.
Full textAbstract:
The standards for emissions from diesel engines are becoming more stringent and accurate emission modeling is crucial in order to control the engine to meet these standards. Soot emissions are formed through a complex process and are challenging to model. A comprehensive analysis of diesel engine soot emissions modeling for control applications is presented in this paper. Physical, black-box, and gray-box models are developed for soot emissions prediction. Additionally, different feature sets based on the least absolute shrinkage and selection operator (LASSO) feature selection method and physical knowledge are examined to develop computationally efficient soot models with good precision. The physical model is a virtual engine modeled in GT-Power software that is parameterized using a portion of experimental data. Different machine learning methods, including Regression Tree (RT), Ensemble of Regression Trees (ERT), Support Vector Machines (SVM), Gaussian Process Regression (GPR), Artificial Neural Network (ANN), and Bayesian Neural Network (BNN) are used to develop the black-box models. The gray-box models include a combination of the physical and black-box models. A total of five feature sets and eight different machine learning methods are tested. An analysis of the accuracy, training time and test time of the models is performed using the K-means clustering algorithm. It provides a systematic way for categorizing the feature sets and methods based on their performance and selecting the best method for a specific application. According to the analysis, the black-box model consisting of GPR and feature selection by LASSO shows the best performance with test R2 of 0.96. The best gray-box model consists of SVM-based method with physical insight feature set along with LASSO for feature selection with test R2 of 0.97.
8
Wongvibulsin, Shannon, Katherine C. Wu, and Scott L. Zeger. "Improving Clinical Translation of Machine Learning Approaches Through Clinician-Tailored Visual Displays of Black Box Algorithms: Development and Validation." JMIR Medical Informatics 8, no. 6 (June 9, 2020): e15791. http://dx.doi.org/10.2196/15791.
Full textAbstract:
Background Despite the promise of machine learning (ML) to inform individualized medical care, the clinical utility of ML in medicine has been limited by the minimal interpretability and black box nature of these algorithms. Objective The study aimed to demonstrate a general and simple framework for generating clinically relevant and interpretable visualizations of black box predictions to aid in the clinical translation of ML. Methods To obtain improved transparency of ML, simplified models and visual displays can be generated using common methods from clinical practice such as decision trees and effect plots. We illustrated the approach based on postprocessing of ML predictions, in this case random forest predictions, and applied the method to data from the Left Ventricular (LV) Structural Predictors of Sudden Cardiac Death (SCD) Registry for individualized risk prediction of SCD, a leading cause of death. Results With the LV Structural Predictors of SCD Registry data, SCD risk predictions are obtained from a random forest algorithm that identifies the most important predictors, nonlinearities, and interactions among a large number of variables while naturally accounting for missing data. The black box predictions are postprocessed using classification and regression trees into a clinically relevant and interpretable visualization. The method also quantifies the relative importance of an individual or a combination of predictors. Several risk factors (heart failure hospitalization, cardiac magnetic resonance imaging indices, and serum concentration of systemic inflammation) can be clearly visualized as branch points of a decision tree to discriminate between low-, intermediate-, and high-risk patients. Conclusions Through a clinically important example, we illustrate a general and simple approach to increase the clinical translation of ML through clinician-tailored visual displays of results from black box algorithms. We illustrate this general model-agnostic framework by applying it to SCD risk prediction. Although we illustrate the methods using SCD prediction with random forest, the methods presented are applicable more broadly to improving the clinical translation of ML, regardless of the specific ML algorithm or clinical application. As any trained predictive model can be summarized in this manner to a prespecified level of precision, we encourage the use of simplified visual displays as an adjunct to the complex predictive model. Overall, this framework can allow clinicians to peek inside the black box and develop a deeper understanding of the most important features from a model to gain trust in the predictions and confidence in applying them to clinical care.
9
Fernando, Denise R., Jonathan P. Lynch, Meredith T. Hanlon, and Alan T. Marshall. "Foliar elemental microprobe data and leaf anatomical traits consistent with drought tolerance in Eucalyptus largiflorens (Myrtaceae)." Australian Journal of Botany 69, no. 4 (2021): 215. http://dx.doi.org/10.1071/bt20170.
Full textAbstract:
In food-productive river basins, ecosystems reliant on natural flows are affected by climate change and water removal. One such example is Australia’s Murray–Darling Basin (MDB), to which the ecologically important black box tree Eucalyptus largiflorens (Myrtaceae) is unique. Little is known about its mineral nutrition and response to flooding. A field study conducted at Hattah Kulkyne National Park on the MDB examined nutrient and Al distribution in mature and young foliage of trees whose status varied with respect to the presence of surface floodwaters. Black box is also of interest due to emerging evidence of its capacity to accumulate high foliar salt concentrations. Here, cryo scanning electron microscopy alone (SEM), combined with energy dispersive spectroscopy (SEM-EDS) and X-ray fluorescence (XRF) spectroscopy were applied to evaluate leaf anatomy and elemental patterns at the cellular and whole-leaf levels. Variation in whole-leaf elemental levels across flooded and dry trees aligned with known nutritional fluctuations in this drought-tolerant species reliant on occasional infrequent flooding. The microprobe data provide evidence of drought tolerance by demonstrating that extended conditions of lack of water to trees do not elicit leaf anatomical changes nor changes to leaf cellular storage of these elements. Foliar Na concentrations of ~2000–6000mgkg–1 DW were found co-localised with Cl in mesophyll and dermal cells of young and mature leaves, suggesting vacuolar salt disposal as a detoxification strategy.
10
Duryea, Mary, George Blakeslee, William Hubbard, and Ricardo Vasquez. "Wind and Trees: A Survey of Homeowners After Hurricane Andrew." Arboriculture & Urban Forestry 22, no. 1 (January 1, 1996): 44–50. http://dx.doi.org/10.48044/jauf.1996.006.
Full textAbstract:
The destructive winds of Hurricane Andrew dramatically changed the urban forest in Dade County, Florida on August 24,1992. Overnight, the tree canopy was replaced by a landscape of broken, uprooted, defoliated and severely damaged trees. To assist communities in reforestation efforts, scientists at the University of Florida conducted a homeowner survey to determine how different tree species responded to strong winds. Native tree species, such as box leaf stopper, sabal palm gumbo limbo, and live oak were the best survivors of the winds. Other palms such as areca, cabada, and Alexander were also highly wind resistant. In general, fruit trees such as navel orange, mango, avocado and grapefruit were severely damaged. Black olive, live oak, and gumbo limbo trees that were pruned survived the hurricane better than unpruned trees. Only 18% of all the trees that fell caused property damage. Hurricane-susceptible communities should consider wind resistance as one of their criteria in tree species selection.
11
Jespersen, Christian Kragh, Miles Cranmer, Peter Melchior, Shirley Ho, Rachel S. Somerville, and Austen Gabrielpillai. "Mangrove: Learning Galaxy Properties from Merger Trees." Astrophysical Journal 941, no. 1 (December 1, 2022): 7. http://dx.doi.org/10.3847/1538-4357/ac9b18.
Full textAbstract:
Abstract Efficiently mapping baryonic properties onto dark matter is a major challenge in astrophysics. Although semianalytic models (SAMs) and hydrodynamical simulations have made impressive advances in reproducing galaxy observables across cosmologically significant volumes, these methods still require significant computation times, representing a barrier to many applications. Graph neural networks have recently proven to be the natural choice for learning physical relations. Among the most inherently graph-like structures found in astrophysics are the dark matter merger trees that encode the evolution of dark matter halos. In this paper, we introduce a new, graph-based emulator framework, Mangrove, and show that it emulates the galactic stellar mass, cold gas mass and metallicity, instantaneous and time-averaged star formation rate, and black hole mass—as predicted by an SAM—with an rms error up to 2 times lower than other methods across a (75 Mpc/h)3 simulation box in 40 s, 4 orders of magnitude faster than the SAM. We show that Mangrove allows for quantification of the dependence of galaxy properties on merger history. We compare our results to the current state of the art in the field and show significant improvements for all target properties. Mangrove is publicly available: https://github.com/astrockragh/Mangrove.
12
Jaafreh, Russlan, Jung-Gu Kim, and Kotiba Hamad. "Interpretable Machine Learning Analysis of Stress Concentration in Magnesium: An Insight beyond the Black Box of Predictive Modeling." Crystals 12, no. 9 (September 2, 2022): 1247. http://dx.doi.org/10.3390/cryst12091247.
Full textAbstract:
In the present work, machine learning (ML) was employed to build a model, and through it, the microstructural features (parameters) affecting the stress concentration (SC) during plastic deformation of magnesium (Mg)-based materials are determined. As a descriptor for the SC, the kernel average misorientation (KAM) was used, and starting from the microstructural features of pure Mg and AZ31 Mg alloy, as recorded using electron backscattered diffraction (EBSD), the ML model was trained and constructed using various types of ML algorithms, including Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), Naive Bayes Classifier (NBC), K-Nearest Neighbor (KNN), Multilayer Perceptron (MLP), and Extremely Randomized Trees (ERT). The results show that the accuracy of the ERT-based model was higher compared to other models, and accordingly, the nine most-important features in the ERT-based model, those with a Gini impurity higher than 0.025, were extracted. The feature importance showed that the grain size is the most effective microstructural parameter for controlling the SC in Mg-based materials, and according to the relative Accumulated Local Effects (ALE) plot, calculated to show the relationship between KAM and grain size, it was found that SC occurs with a lower probability in the fine range of grain size. All findings from the ML-based model built in the present work were experimentally confirmed through EBSD observations.
13
Afdhal, Muhammad, Vicky Ariandi, and Rita Rita. "Memprediksi Penjualan Pada Toko Hanifah Metode C.45." Jurnal Teknologi Dan Sistem Informasi Bisnis 4, no. 2 (July 1, 2022): 248–55. http://dx.doi.org/10.47233/jteksis.v4i1.460.
Full textAbstract:
Drug stocks in pharmacies are important information for the sales process. The existing stock is not in accordance with the needs of consumers, the distribution of drugs that are less needed in the stock will cause losses because the drug has expired due to too long stored in the warehouse. Another problem is that obat cannot predict drugs that are needed a lot, to overcome these problems needed a prediction system. These problems can be solved by the decision tree method for the prediction of drug supplies. The concept of the Decision Tree Algorithm is to convert data into decision trees and decision rules. System development with Waterfall model, using PHP programming language, My SQL database, system design using object oriented approach, system testing using Black Box for functionality test, validity testing with rapid miner tools. The result of the development of the system is a prediction of drug sales at pharmacies. The result of black-box testing is that all developed systems work properly. The results of the validity test by comparing the old system with the new system with the rapid miner, using 30 transaction samples the accuracy is 89% which means the system has good performance.
14
Hermsen, Eden, Anne Kerle, and Julie M. Old. "Diet of an inland population of the common ringtail possum (Pseudocheirus peregrinus)." Australian Mammalogy 38, no. 1 (2016): 130. http://dx.doi.org/10.1071/am15008.
Full textAbstract:
Populations of the common ringtail possum (Pseudocheirus peregrinus) in inland New South Wales have declined or disappeared. Habitat requirements and diet of these populations are poorly understood. Determining the diet of inland ringtail possums is crucial to understanding the factors limiting their survival, and was the focus of this study. Spotlighting surveys were conducted to locate ringtail possums, and scat and vegetation samples were collected for microhistological analysis. Ringtail possums were most frequently observed in red stringybark followed by bundy box and black cypress pine trees, and this correlated with the most common dietary items consumed.
15
de Lima, Robson Borges, Eric Bastos Görgens, Anderson Pedro Bernardina Batista, Diego Armando Silva da Silva, Cinthia Pereira de Oliveira, and Carla Samara Campelo de Sousa. "Diversity and Big Trees Patterns in the Brazilian Amazon." Diversity 14, no. 7 (June 22, 2022): 503. http://dx.doi.org/10.3390/d14070503.
Full textAbstract:
The increasing availability of field data presents an opportunity to understand the fundamental ecological relationships and functions of large trees in tropical forests at regional and global scales. However, it is not always clear what the relationships or patterns of diversity and structure are among sites in different biogeographic regions. We evaluated the relationship of the biomass and diameter of the largest trees with a diversity of species and compared, between the sites, the attributes of structure, diversity, and the influence of the 50 hyperdominant species in each site, aiming at the potential formation of groups by sites with characteristics and patterns of similar diversity within biogeographic regions. The average wood density together with the diversity of genera and families are the most important attributes to discriminate biogeographic regions when considering all forest information. Large trees play a fundamental role in forest ecology and seem to express regional environmental characteristics. The upper canopy of tropical forests remains one of the least studied environments in all terrestrial biomes, and is often referred to as “the last biotic frontier” or a “black box,” and large trees are also part of this mysterious frontier.
16
Morales Sánchez, Damián, Antonio Moreno, and María Dolores Jiménez López. "A White-Box Sociolinguistic Model for Gender Detection." Applied Sciences 12, no. 5 (March 4, 2022): 2676. http://dx.doi.org/10.3390/app12052676.
Full textAbstract:
Within the area of Natural Language Processing, we approached the Author Profiling task as a text classification problem. Based on the author’s writing style, sociodemographic information, such as the author’s gender, age, or native language can be predicted. The exponential growth of user-generated data and the development of Machine-Learning techniques have led to significant advances in automatic gender detection. Unfortunately, gender detection models often become black-boxes in terms of interpretability. In this paper, we propose a tree-based computational model for gender detection made up of 198 features. Unlike the previous works on gender detection, we organized the features from a linguistic perspective into six categories: orthographic, morphological, lexical, syntactic, digital, and pragmatics-discursive. We implemented a Decision-Tree classifier to evaluate the performance of all feature combinations, and the experiments revealed that, on average, the classification accuracy increased up to 3.25% with the addition of feature sets. The maximum classification accuracy was reached by a three-level model that combined lexical, syntactic, and digital features. We present the most relevant features for gender detection according to the trees generated by the classifier and contextualize the significance of the computational results with the linguistic patterns defined by previous research in relation to gender.
17
Fernando, Denise R., Jonathan P. Lynch, Suzie M. Reichman, Gary J. Clark, Rebecca E. Miller, and Tanya M. Doody. "Inundation of a floodplain lake woodlands system: nutritional profiling and benefit to mature Eucalyptus largiflorens (Black Box) trees." Wetlands Ecology and Management 26, no. 5 (September 1, 2018): 961–75. http://dx.doi.org/10.1007/s11273-018-9623-x.
Full text
18
Deptuła, Adam, Michał Stosiak, Mykola Karpenko, and Mariusz Łapka. "The Concept of Dependency Game Tree Graphs as a Black Box in the Analysis of Automatic Transmissions." Transport and Telecommunication Journal 23, no. 3 (June 1, 2022): 207–19. http://dx.doi.org/10.2478/ttj-2022-0017.
Full textAbstract:
Abstract The methodology of graph theory has long had applications in mechanics. Graphs allow the simultaneous consideration and synthesis of the structure of a real system and the idealized structure of an equivalent model of a mechanical system. In particular, there are many applications of graph theory in the analysis of automatic transmissions. This paper presents continuing research on applications of game tree graphs in the analysis of automatic transmissions. Black box ideas for dependency graphs are presented. This makes it possible to automatically generate simplified models from given physical models of an automatic transmission and to determine the optimal number of teeth. The developed methodology allows also to perform real-time simulations. In a further stage, the methods of decision logic trees can be applied to analyze the functional diagrams of selected gears.
19
Dufour-Pelletier, Samuel, Junior A. Tremblay, Christian Hébert, Thibault Lachat, and Jacques Ibarzabal. "Testing the Effect of Snag and Cavity Supply on Deadwood-Associated Species in a Managed Boreal Forest." Forests 11, no. 4 (April 9, 2020): 424. http://dx.doi.org/10.3390/f11040424.
Full textAbstract:
Standing deadwood is an important attribute of old-growth boreal forests and it provides essential microhabitats for deadwood-associated species. In managed boreal forests, short rotations tend to limit the amount and diversity of standing deadwood. This study evaluates if the anthropogenic supply of deadwood attributes through tree girdling or by providing nest boxes may favor deadwood-associated species. We studied the short-term response of saproxylic beetles, foraging woodpeckers, and secondary cavity users to snag and cavity supply in 50 to 70-year-old black spruce stands. In spring 2015, we girdled 8000 black spruce according to two spatial distributions (uniform and clustered), and we also installed 450 nest boxes of six different sizes at three distances from the forest edge. Using trunk window traps, we captured significantly more beetles in sites with girdled trees than in control sites in both 2015 and 2016. We also recorded a trend of a greater abundance of beetles in clusters of girdled trees than within uniformly distributed girdled trees. Trypodendron lineatum (Oliver) dominated beetle assemblages, representing 88.5% of all species in 2015 and 74.6% in 2016. The number of beetles captured was 7× higher in 2015 than in 2016. In contrast, we observed greater amounts of woodpecker foraging marks in fall 2016 than in either fall 2015 or spring 2016. Woodpeckers foraged significantly more in clusters of girdled trees than within uniformly distributed girdled trees. Woodpeckers’ foraging mark presence was positively associated with the proportion of recent cuts at 1 km around the study sites. Five Boreal Chickadee (Poecile hudsonicus Forster) pairs used nest boxes and occupied smaller box sizes that were located away from the forest edge. Our study showed that structural enrichment can be effective in rapidly attracting deadwood-associated species within managed forest stands.
20
He, Qi, and Bin Hu. "Research on the Influencing Factors of Film Consumption and Box Office Forecast in the Digital Era: Based on the Perspective of Machine Learning and Model Integration." Wireless Communications and Mobile Computing 2021 (October 14, 2021): 1–10. http://dx.doi.org/10.1155/2021/6094924.
Full textAbstract:
The film industry is one of the core industries of the digital creative industry, which has great positive externalities to the digital creative economy. Movie box office revenue is an important indicator to measure the realization of the market value of movie consumption, and it is also the basic guarantee for the sustainable development of the movie industry. This paper relies on the professional database of the Maoyan movie market to use Python software to collect a total of 830 domestic movie-related consumption characteristic data from 2017 to 2019. In this study, the stacking method in the machine learning ensemble algorithm combines the fivefold crossfolding training method based on distributed random forest, extremely randomized trees, and generalized linear models. The model is good at handling different data types. It has higher fitting and model accuracy in feature mining and model construction, so as to effectively grasp the relevant feature factors affecting movie consumption and accurately predict the future movie box office. Based on the innovative design method of model fusion, the extracted feature vector is used to build a more accurate movie box office prediction model through stacking with a fivefold crossfolding training method. It is aimed at opening the black box that affects the realization of the value of the film content consumption market in the digital age and putting forward corresponding countermeasures and suggestions.
21
Laud, Peeter. "Parallel Oblivious Array Access for Secure Multiparty Computation and Privacy-Preserving Minimum Spanning Trees." Proceedings on Privacy Enhancing Technologies 2015, no. 2 (June 1, 2015): 188–205. http://dx.doi.org/10.1515/popets-2015-0011.
Full textAbstract:
AbstractIn this paper, we describe efficient protocols to perform in parallel many reads and writes in private arrays according to private indices. The protocol is implemented on top of the Arithmetic Black Box (ABB) and can be freely composed to build larger privacypreserving applications. For a large class of secure multiparty computation (SMC) protocols, our technique has better practical and asymptotic performance than any previous ORAM technique that has been adapted for use in SMC.Our ORAM technique opens up a large class of parallel algorithms for adoption to run on SMC platforms. In this paper, we demonstrate how the minimum spanning tree (MST) finding algorithm by Awerbuch and Shiloach can be executed without revealing any details about the underlying graph (beside its size). The data accesses of this algorithm heavily depend on the location and weight of edges (which are private) and our ORAM technique is instrumental in their execution. Our implementation is the first-ever realization of a privacypreserving MST algorithm with sublinear round complexity.
22
Gjærum, Vilde B., Inga Strümke, Ole Andreas Alsos, and Anastasios M. Lekkas. "Explaining a Deep Reinforcement Learning Docking Agent Using Linear Model Trees with User Adapted Visualization." Journal of Marine Science and Engineering 9, no. 11 (October 26, 2021): 1178. http://dx.doi.org/10.3390/jmse9111178.
Full textAbstract:
Deep neural networks (DNNs) can be useful within the marine robotics field, but their utility value is restricted by their black-box nature. Explainable artificial intelligence methods attempt to understand how such black-boxes make their decisions. In this work, linear model trees (LMTs) are used to approximate the DNN controlling an autonomous surface vessel (ASV) in a simulated environment and then run in parallel with the DNN to give explanations in the form of feature attributions in real-time. How well a model can be understood depends not only on the explanation itself, but also on how well it is presented and adapted to the receiver of said explanation. Different end-users may need both different types of explanations, as well as different representations of these. The main contributions of this work are (1) significantly improving both the accuracy and the build time of a greedy approach for building LMTs by introducing ordering of features in the splitting of the tree, (2) giving an overview of the characteristics of the seafarer/operator and the developer as two different end-users of the agent and receiver of the explanations, and (3) suggesting a visualization of the docking agent, the environment, and the feature attributions given by the LMT for when the developer is the end-user of the system, and another visualization for when the seafarer or operator is the end-user, based on their different characteristics.
23
Fera Agurini, Aulia Desy Nur Utomo, and Fahrudin Mukti Wibowo. "Designing a Fallen Tree Disaster Reporting Application Based on Mobile Android Case Study: Regional Disaster Management Agency (BPBD) Banyumas Regency." Jurnal E-Komtek (Elektro-Komputer-Teknik) 6, no. 1 (June 30, 2022): 114–28. http://dx.doi.org/10.37339/e-komtek.v6i1.913.
Full textAbstract:
Banyumas Regency is prone to falling trees in the rainy season, but the report on fallen tree disasters is still not focused. The large number of social media used and the regent's complaint booth confuses the reporting process because it is not centralized in one system, making data recording difficult. The authorities in handling the fallen tree disaster are the Regional Disaster Management Agency (BPBD) of Banyumas district. With these problems, an android application was designed that functions to facilitate the process of reporting disasters from the community, as well as the process of recording fallen tree disaster data. The design stage uses the Waterfall method, and the testing uses UAT and black box testing. The testing was carried out by three users, namely the Community, Admin, and Officers. The respondents from the community as many as 120 people, as many as 20 officers, and admin as many as 3 people. The results of the UAT application test are that the application is very helpful for the disaster reporting process; the menu display is attractive; the application can be understood; the type, font size, and color is easy to read; and the users do not find it difficult to make an account, and the application menu meets the users’ needs. In the black box testing, it was found that the designed application was very helpful with an average validation percentage of 98% for each menu.
24
Mashayekhi, Morteza, and Robin Gras. "Rule Extraction from Decision Trees Ensembles: New Algorithms Based on Heuristic Search and Sparse Group Lasso Methods." International Journal of Information Technology & Decision Making 16, no. 06 (November 2017): 1707–27. http://dx.doi.org/10.1142/s0219622017500055.
Full textAbstract:
Decision trees are examples of easily interpretable models whose predictive accuracy is normally low. In comparison, decision tree ensembles (DTEs) such as random forest (RF) exhibit high predictive accuracy while being regarded as black-box models. We propose three new rule extraction algorithms from DTEs. The RF[Formula: see text]DHC method, a hill climbing method with downhill moves (DHC), is used to search for a rule set that decreases the number of rules dramatically. In the RF[Formula: see text]SGL and RF[Formula: see text]MSGL methods, the sparse group lasso (SGL) method, and the multiclass SGL (MSGL) method are employed respectively to find a sparse weight vector corresponding to the rules generated by RF. Experimental results with 24 data sets show that the proposed methods outperform similar state-of-the-art methods, in terms of human comprehensibility, by greatly reducing the number of rules and limiting the number of antecedents in the retained rules, while preserving the same level of accuracy.
25
Kennedy, Ann-Marie, Sommer Kapitan, Neha Bajaj, Angelina Bakonyi, and Sean Sands. "Uncovering wicked problem’s system structure: seeing the forest for the trees." Journal of Social Marketing 7, no. 1 (January 3, 2017): 51–73. http://dx.doi.org/10.1108/jsocm-05-2016-0029.
Full textAbstract:
Purpose This paper aims to use systems thinking, systems theory and Camillus’ framework for responding to wicked problems to provide social marketers with a theoretically based framework for approaching strategy formation for wicked problems. The paper treats fast fashion as an illustrative case and takes a step back from implementation to provide a framework for analysing and gaining understanding of wicked problem system structure for social marketers to then plan more effective interventions. The proposed approach is intended as a theory-based tool for social marketing practitioners to uncover system structure and analyse the wicked problems they face. Design/methodology/approach Following Layton, this work provides theoretically based guidelines for analysing the black box of how to develop and refine strategy as first proposed in Camillus’ (2008) framework for responding to wicked issues. Findings The prescription thus developed for approaching wicked problems’ system structure revolves around identifying the individuals, groups or entities that make up the system involved in the wicked problem, and then determining which social mechanisms most clearly drive each entity and which outcomes motivate these social mechanisms, before determining which role the entities play as either incumbent, challenger or governance and which social narratives drive each role’s participation in the wicked problem. Originality/value This paper shows that using systems thinking can help social marketers to gain big picture thinking and develop strategy for responding to complex issues, while considering the consequences of interventions.
26
Han, Tao, and Gerardo Arturo Sánchez-Azofeifa. "A Deep Learning Time Series Approach for Leaf and Wood Classification from Terrestrial LiDAR Point Clouds." Remote Sensing 14, no. 13 (July 1, 2022): 3157. http://dx.doi.org/10.3390/rs14133157.
Full textAbstract:
The accurate separation between leaf and woody components from terrestrial laser scanning (TLS) data is vital for the estimation of leaf area index (LAI) and wood area index (WAI). Here, we present the application of deep learning time series separation of leaves and wood from TLS point clouds collected from broad-leaved trees. First, we use a multiple radius nearest neighbor approach to obtain a time series of the geometric features. Second, we compare the performance of Fully Convolutional Neural Network (FCN), Long Short-Term Memory Fully Convolutional Neural Network (LSTM-FCN), and Residual Network (ResNet) on leaf and wood classification. We also compare the effect of univariable (UTS) and multivariable (MTS) time series on classification accuracy. Finally, we explore the utilization of a class activation map (CAM) to reduce the black-box effect of deep learning. The average overall accuracy of the MTS method across the training data is 0.96, which is higher than the UTS methods (0.67 to 0.88). Meanwhile, ResNet spent much more time than FCN and LSTM-FCN in model development. When testing our method on an independent dataset, the MTS models based on FCN, LSTM-FCN, and ResNet all demonstrate similar performance. Our method indicates that the CAM can explain the black-box effect of deep learning and suggests that deep learning algorithms coupled with geometric feature time series can accurately separate leaf and woody components from point clouds. This provides a good starting point for future research into estimation of forest structure parameters.
27
Horry, Michael, Subrata Chakraborty, Biswajeet Pradhan, Manoranjan Paul, Douglas Gomes, Anwaar Ul-Haq, and Abdullah Alamri. "Deep Mining Generation of Lung Cancer Malignancy Models from Chest X-ray Images." Sensors 21, no. 19 (October 7, 2021): 6655. http://dx.doi.org/10.3390/s21196655.
Full textAbstract:
Lung cancer is the leading cause of cancer death and morbidity worldwide. Many studies have shown machine learning models to be effective in detecting lung nodules from chest X-ray images. However, these techniques have yet to be embraced by the medical community due to several practical, ethical, and regulatory constraints stemming from the “black-box” nature of deep learning models. Additionally, most lung nodules visible on chest X-rays are benign; therefore, the narrow task of computer vision-based lung nodule detection cannot be equated to automated lung cancer detection. Addressing both concerns, this study introduces a novel hybrid deep learning and decision tree-based computer vision model, which presents lung cancer malignancy predictions as interpretable decision trees. The deep learning component of this process is trained using a large publicly available dataset on pathological biomarkers associated with lung cancer. These models are then used to inference biomarker scores for chest X-ray images from two independent data sets, for which malignancy metadata is available. Next, multi-variate predictive models were mined by fitting shallow decision trees to the malignancy stratified datasets and interrogating a range of metrics to determine the best model. The best decision tree model achieved sensitivity and specificity of 86.7% and 80.0%, respectively, with a positive predictive value of 92.9%. Decision trees mined using this method may be considered as a starting point for refinement into clinically useful multi-variate lung cancer malignancy models for implementation as a workflow augmentation tool to improve the efficiency of human radiologists.
28
Patil, Shruti, Vijayakumar Varadarajan, Siddiqui Mohd Mazhar, Abdulwodood Sahibzada, Nihal Ahmed, Onkar Sinha, Satish Kumar, Kailash Shaw, and Ketan Kotecha. "Explainable Artificial Intelligence for Intrusion Detection System." Electronics 11, no. 19 (September 27, 2022): 3079. http://dx.doi.org/10.3390/electronics11193079.
Full textAbstract:
Intrusion detection systems are widely utilized in the cyber security field, to prevent and mitigate threats. Intrusion detection systems (IDS) help to keep threats and vulnerabilities out of computer networks. To develop effective intrusion detection systems, a range of machine learning methods are available. Machine learning ensemble methods have a well-proven track record when it comes to learning. Using ensemble methods of machine learning, this paper proposes an innovative intrusion detection system. To improve classification accuracy and eliminate false positives, features from the CICIDS-2017 dataset were chosen. This paper proposes an intrusion detection system using machine learning algorithms such as decision trees, random forests, and SVM (IDS). After training these models, an ensemble technique voting classifier was added and achieved an accuracy of 96.25%. Furthermore, the proposed model also incorporates the XAI algorithm LIME for better explainability and understanding of the black-box approach to reliable intrusion detection. Our experimental results confirmed that XAI LIME is more explanation-friendly and more responsive.
29
Abbas, Amel H., and Marwa A. Shamel. "Identify and Classify Normal and Defects of Prunus_armeniaca Using Imaging Techniques." Kurdistan Journal of Applied Research 2, no. 3 (August 27, 2017): 1–6. http://dx.doi.org/10.24017/science.2017.3.11.
Full textAbstract:
The Prunus_armeniaca fruit is classified manually in wholesale markets, supermarkets and food processing plants on a normal or defects basis. The aim of this research is to replace the manual sorting techniques using computer vision techniques and applications by proposing techniques for identify and recognitions patterns through the use of 150 fruits of Prunus_armeniaca, 10 for the testing stage in fresh and 10 for testing stage in case of defects. The fruits Prunus_armeniaca collected from growing trees in the large fields of Salah al-Din province\Iraq. The system designed for classification based on the color image taken inside a black box used camera pixel resolution of (13 mega) with a constant intensity of light. . Used K-mean in phase segmentations and only computed 13 features derive statistics from GLCM .classification phase used SVM classify fruit into two class, either (normal or defects) .Results the system success rate reach 100%.The work done using MATLAB R2016a.
30
Rezazadeh, Alireza, Yasamin Jafarian, and Ali Kord. "Explainable Ensemble Machine Learning for Breast Cancer Diagnosis Based on Ultrasound Image Texture Features." Forecasting 4, no. 1 (February 13, 2022): 262–74. http://dx.doi.org/10.3390/forecast4010015.
Full textAbstract:
Image classification is widely used to build predictive models for breast cancer diagnosis. Most existing approaches overwhelmingly rely on deep convolutional networks to build such diagnosis pipelines. These model architectures, although remarkable in performance, are black-box systems that provide minimal insight into the inner logic behind their predictions. This is a major drawback as the explainability of prediction is vital for applications such as cancer diagnosis. In this paper, we address this issue by proposing an explainable machine learning pipeline for breast cancer diagnosis based on ultrasound images. We extract first- and second-order texture features of the ultrasound images and use them to build a probabilistic ensemble of decision tree classifiers. Each decision tree learns to classify the input ultrasound image by learning a set of robust decision thresholds for texture features of the image. The decision path of the model predictions can then be interpreted by decomposing the learned decision trees. Our results show that our proposed framework achieves high predictive performance while being explainable.
31
Carnegie, Nicole Bohme, and James Wu. "Variable Selection and Parameter Tuning for BART Modeling in the Fragile Families Challenge." Socius: Sociological Research for a Dynamic World 5 (January 2019): 237802311982588. http://dx.doi.org/10.1177/2378023119825886.
Full textAbstract:
Our goal for the Fragile Families Challenge was to develop a hands-off approach that could be applied in many settings to identify relationships that theory-based models might miss. Data processing was our first and most time-consuming task, particularly handling missing values. Our second task was to reduce the number of variables for modeling, and we compared several techniques for variable selection: least absolute selection and shrinkage operator, regression with a horseshoe prior, Bayesian generalized linear models, and Bayesian additive regression trees (BART). We found minimal differences in final performance based on the choice of variable selection method. We proceeded with BART for modeling because it requires minimal assumptions and permits great flexibility in fitting surfaces and based on previous success using BART in black-box modeling competitions. In addition, BART allows for probabilistic statements about the predictions and other inferences, which is an advantage over most machine learning algorithms. A drawback to BART, however, is that it is often difficult to identify or characterize individual predictors that have strong influences on the outcome variable.
32
Rizalno, Muhammad Fando, Asahar Johar, and Funny Farady Coastera. "Analisis Prediksi Masa Studi Mahasiswa Menggunakan Metode Decision Tree Dengan Penerapan Algoritme Cart (Classification and Regression Trees) (Studi Kasus Data Alumni Fakultas Teknik Universitas Bengkulu)." Rekursif: Jurnal Informatika 10, no. 1 (April 24, 2022): 96–106. http://dx.doi.org/10.33369/rekursif.v10i1.21362.
Full textAbstract:
Salah satu suber daya besar yang dimiliki universitas adalah basis data, namun basis data yang besar ini hanya disimpan dalam gudang data. Padahal data yang terkumpul dan berukuran besar tersebut merupakan aset yang dapat dimanfaatkan untuk dianalisis yang hasilnya berupa pengetahuan atau informasi berharga. Melihat kondisi tersebut diperlukan penelitian untuk menggali data yang dimiliki oleh Universitas Bengkulu untuk melihat parameter yang paling berpengaruh pada lama masa studi mahasiswa, data yang akan dimanfaatkan disini adalah data akademik dan data wisudawan. Metode yang digunakan untuk menganalisinya adalah Decission tree dengan menggunakan algoritme CART. Melakukan prediksi dibutuhkan dataset yang teratur, Dataset yang digunakan masih mengandung missing values sehingga dalam penelitian ini tahap prepocessing data dilakukan. Tahap preprocessing menggunakan sistem data warehouse dengan penerapa ETL. Aplikasi ini dibuat dengan menggunakan Visual studio code bahasa pemrograman PHP framewort laravel. Hasil dari penelitian ini berupa aplikasi pengelompokkan data alumni lulus tepat waktu dan prediksi lama masa studi utuk mahasiswa baru. Pengujian software menggunakan metode Evaluasi Accuracy dan Black box. Hasil dari penelitian ini adalah nilai parameter yang berpengaruh pada kelulusan mahasiswa dan dapat digunakan untuk proses prediksi masa studi mahasiswa.
33
Kim, Sangwon, Byoung-Chul Ko, and Jaeyeal Nam. "Model Simplification of Deep Random Forest for Real-Time Applications of Various Sensor Data." Sensors 21, no. 9 (April 25, 2021): 3004. http://dx.doi.org/10.3390/s21093004.
Full textAbstract:
The deep random forest (DRF) has recently gained new attention in deep learning because it has a high performance similar to that of a deep neural network (DNN) and does not rely on a backpropagation. However, it connects a large number of decision trees to multiple layers, thereby making analysis difficult. This paper proposes a new method for simplifying a black-box model of a DRF using a proposed rule elimination. For this, we consider quantifying the feature contributions and frequency of the fully trained DRF in the form of a decision rule set. The feature contributions provide a basis for determining how features affect the decision process in a rule set. Model simplification is achieved by eliminating unnecessary rules by measuring the feature contributions. Consequently, the simplified and transparent DRF has fewer parameters and rules than before. The proposed method was successfully applied to various DRF models and benchmark sensor datasets while maintaining a robust performance despite the elimination of a large number of rules. A comparison with state-of-the-art compressed DNNs also showed the proposed model simplification’s higher parameter compression and memory efficiency with a similar classification accuracy.
34
Drobnič, Franc, Andrej Kos, and Matevž Pustišek. "On the Interpretability of Machine Learning Models and Experimental Feature Selection in Case of Multicollinear Data." Electronics 9, no. 5 (May 6, 2020): 761. http://dx.doi.org/10.3390/electronics9050761.
Full textAbstract:
In the field of machine learning, a considerable amount of research is involved in the interpretability of models and their decisions. The interpretability contradicts the model quality. Random Forests are among the best quality technologies of machine learning, but their operation is of “black box” character. Among the quantifiable approaches to the model interpretation, there are measures of association of predictors and response. In case of the Random Forests, this approach usually consists of calculating the model’s feature importances. Known methods, including the built-in one, are less suitable in settings with strong multicollinearity of features. Therefore, we propose an experimental approach to the feature selection task, a greedy forward feature selection method with least-trees-used criterion. It yields a set of most informative features that can be used in a machine learning (ML) training process with similar prediction quality as the original feature set. We verify the results of the proposed method on two known datasets, one with small feature multicollinearity and another with large feature multicollinearity. The proposed method also allows for a domain expert help with selecting among equally important features, which is known as the human-in-the-loop approach.
35
Qiu, Mengning, Mashor Housh, and Avi Ostfeld. "A Two-Stage LP-NLP Methodology for the Least-Cost Design and Operation of Water Distribution Systems." Water 12, no. 5 (May 12, 2020): 1364. http://dx.doi.org/10.3390/w12051364.
Full textAbstract:
This paper presents a two-stage method for simultaneous least-cost design and operation of looped water distribution systems (WDSs). After partitioning the network into a chord and spanning trees, in the first stage, a reformulated linear programming (LP) method is used to find the least cost design of a WDS for a given set of flow distribution. In the second stage, a non-linear programming (NLP) method is used to find a new flow distribution that reduces the cost of the WDS operation given the WDS design obtained in stage one. The following features of the proposed two-stage method make it more appealing compared to other methods: (1) the reformulated LP stage can consistently reduce the penalty cost when designing a WDS under multiple loading conditions; (2) robustness as the number of loading conditions increases; (3) parameter tuning is not required; (4) the method reduces the computational burden significantly when compared to meta-heuristic methods; and (5) in oppose to an evolutionary “black box” based methodology such as a genetic algorithm, insights through analytical sensitivity analysis, while the algorithm progresses, are handy. The efficacy of the proposed methodology is demonstrated using two WDSs case studies.
36
Zhao, Linan (Frank). "Data-Driven Approach for Predicting and Explaining the Risk of Long-Term Unemployment." E3S Web of Conferences 214 (2020): 01023. http://dx.doi.org/10.1051/e3sconf/202021401023.
Full textAbstract:
Long-term unemployment has significant societal impact and is of particular concerns for policymakers with regard to economic growth and public finances. This paper constructs advanced ensemble machine learning models to predict citizens’ risks of becoming long-term unemployed using data collected from European public authorities for employment service. The proposed model achieves 81.2% accuracy on identifying citizens with high risks of long-term unemployment. This paper also examines how to dissect black-box machine learning models by offering explanations at both a local and global level using SHAP, a state-of-the-art model-agnostic approach to explain factors that contribute to long-term unemployment. Lastly, this paper addresses an under-explored question when applying machine learning in the public domain, that is, the inherent bias in model predictions. The results show that popular models such as gradient boosted trees may produce unfair predictions against senior age groups and immigrants. Overall, this paper sheds light on the recent increasing shift for governments to adopt machine learning models to profile and prioritize employment resources to reduce the detrimental effects of long-term unemployment and improve public welfare.
37
Granata, Francesco, Stefano Papirio, Giovanni Esposito, Rudy Gargano, and Giovanni De Marinis. "Machine Learning Algorithms for the Forecasting of Wastewater Quality Indicators." Water 9, no. 2 (February 9, 2017): 105. http://dx.doi.org/10.3390/w9020105.
Full textAbstract:
Stormwater runoff is often contaminated by human activities. Stormwater discharge into water bodies significantly contributes to environmental pollution. The choice of suitable treatment technologies is dependent on the pollutant concentrations. Wastewater quality indicators such as biochemical oxygen demand (BOD5), chemical oxygen demand (COD), total suspended solids (TSS), and total dissolved solids (TDS) give a measure of the main pollutants. The aim of this study is to provide an indirect methodology for the estimation of the main wastewater quality indicators, based on some characteristics of the drainage basin. The catchment is seen as a black box: the physical processes of accumulation, washing, and transport of pollutants are not mathematically described. Two models deriving from studies on artificial intelligence have been used in this research: Support Vector Regression (SVR) and Regression Trees (RT). Both the models showed robustness, reliability, and high generalization capability. However, with reference to coefficient of determination R2 and root‐mean square error, Support Vector Regression showed a better performance than Regression Tree in predicting TSS, TDS, and COD. As regards BOD5, the two models showed a comparable performance. Therefore, the considered machine learning algorithms may be useful for providing an estimation of the values to be considered for the sizing of the treatment units in absence of direct measures.
38
Baressi Šegota, Sandi, Ivan Lorencin, Mario Šercer, and Zlatan Car. "Determining residuary resistance per unit weight of displacement with Symbolic Regression and Gradient Boosted Tree algorithms." Pomorstvo 35, no. 2 (December 22, 2021): 287–96. http://dx.doi.org/10.31217/p.35.2.11.
Full textAbstract:
Determining the residuary resistance per unit weight of displacement is one of the key factors in the design of vessels. In this paper, the authors utilize two novel methods – Symbolic Regression (SR) and Gradient Boosted Trees (GBT) to achieve a model which can be used to calculate the value of residuary resistance per unit weight, of displacement from the longitudinal position of the center of buoyancy, prismatic coefficient, length-displacement ratio, beam-draught ratio, length-beam ratio, and Froude number. This data is given as results of 308 experiments provided as a part of a publicly available dataset. The results are evaluated using the coefficient of determination (R2) and Mean Absolute Percentage Error (MAPE). Pre-processing, in the shape of correlation analysis combined with variable elimination and variable scaling, is applied to the dataset. The results show that while both methods achieve regression results, the result of regression of SR is relatively poor in comparison to GBT. Both methods provide slightly poorer, but comparable results to previous research focussing on the use of “black-box” methods, such as neural networks. The elimination of variables does not show a high influence on the modeling performance in the presented case, while variable scaling does achieve better results compared to the models trained with the non-scaled dataset.
39
Wu, Mike, Sonali Parbhoo, Michael C. Hughes, Volker Roth, and Finale Doshi-Velez. "Optimizing for Interpretability in Deep Neural Networks with Tree Regularization." Journal of Artificial Intelligence Research 72 (September 14, 2021): 1–37. http://dx.doi.org/10.1613/jair.1.12558.
Full textAbstract:
Deep models have advanced prediction in many domains, but their lack of interpretability remains a key barrier to the adoption in many real world applications. There exists a large body of work aiming to help humans understand these black box functions to varying levels of granularity – for example, through distillation, gradients, or adversarial examples. These methods however, all tackle interpretability as a separate process after training. In this work, we take a different approach and explicitly regularize deep models so that they are well-approximated by processes that humans can step through in little time. Specifically, we train several families of deep neural networks to resemble compact, axis-aligned decision trees without significant compromises in accuracy. The resulting axis-aligned decision functions uniquely make tree regularized models easy for humans to interpret. Moreover, for situations in which a single, global tree is a poor estimator, we introduce a regional tree regularizer that encourages the deep model to resemble a compact, axis-aligned decision tree in predefined, human-interpretable contexts. Using intuitive toy examples, benchmark image datasets, and medical tasks for patients in critical care and with HIV, we demonstrate that this new family of tree regularizers yield models that are easier for humans to simulate than L1 or L2 penalties without sacrificing predictive power.
40
Silva, Pedro, Leon Cao, and Wayne Hayes. "SpArcFiRe: Enhancing Spiral Galaxy Recognition Using Arm Analysis and Random Forests." Galaxies 6, no. 3 (September 5, 2018): 95. http://dx.doi.org/10.3390/galaxies6030095.
Full textAbstract:
Automated quantification of galaxy morphology is necessary because the size of upcoming sky surveys will overwhelm human volunteers. Existing classification schemes are inadequate because (a) their uncertainty increases near the boundary of classes and astronomers need more control over these uncertainties; (b) galaxy morphology is continuous rather than discrete; and (c) sometimes we need to know not only the type of an object, but whether a particular image of the object exhibits visible structure. We propose that regression is better suited to these tasks than classification, and focus specifically on determining the extent to which an image of a spiral galaxy exhibits visible spiral structure. We use the human vote distributions from Galaxy Zoo 1 (GZ1) to train a random forest of decision trees to reproduce the fraction of GZ1 humans who vote for the “Spiral” class. We prefer the random forest model over other black box models like neural networks because it allows us to trace post hoc the precise reasoning behind the regression of each image. Finally, we demonstrate that using features from SpArcFiRe—a code designed to isolate and quantify arm structure in spiral galaxies—improves regression results over and above using traditional features alone, across a sample of 470,000 galaxies from the Sloan Digital Sky Survey.
41
Afrabandpey, Homayun, Tomi Peltola, Juho Piironen, Aki Vehtari, and Samuel Kaski. "A decision-theoretic approach for model interpretability in Bayesian framework." Machine Learning 109, no. 9-10 (September 2020): 1855–76. http://dx.doi.org/10.1007/s10994-020-05901-8.
Full textAbstract:
Abstract A salient approach to interpretable machine learning is to restrict modeling to simple models. In the Bayesian framework, this can be pursued by restricting the model structure and prior to favor interpretable models. Fundamentally, however, interpretability is about users’ preferences, not the data generation mechanism; it is more natural to formulate interpretability as a utility function. In this work, we propose an interpretability utility, which explicates the trade-off between explanation fidelity and interpretability in the Bayesian framework. The method consists of two steps. First, a reference model, possibly a black-box Bayesian predictive model which does not compromise accuracy, is fitted to the training data. Second, a proxy model from an interpretable model family that best mimics the predictive behaviour of the reference model is found by optimizing the interpretability utility function. The approach is model agnostic—neither the interpretable model nor the reference model are restricted to a certain class of models—and the optimization problem can be solved using standard tools. Through experiments on real-word data sets, using decision trees as interpretable models and Bayesian additive regression models as reference models, we show that for the same level of interpretability, our approach generates more accurate models than the alternative of restricting the prior. We also propose a systematic way to measure stability of interpretabile models constructed by different interpretability approaches and show that our proposed approach generates more stable models.
42
Post, Christian, Christian Rietz, Wolfgang Büscher, and Ute Müller. "Using Sensor Data to Detect Lameness and Mastitis Treatment Events in Dairy Cows: A Comparison of Classification Models." Sensors 20, no. 14 (July 10, 2020): 3863. http://dx.doi.org/10.3390/s20143863.
Full textAbstract:
The aim of this study was to develop classification models for mastitis and lameness treatments in Holstein dairy cows as the target variables based on continuous data from herd management software with modern machine learning methods. Data was collected over a period of 40 months from a total of 167 different cows with daily individual sensor information containing milking parameters, pedometer activity, feed and water intake, and body weight (in the form of differently aggregated data) as well as the entered treatment data. To identify the most important predictors for mastitis and lameness treatments, respectively, Random Forest feature importance, Pearson’s correlation and sequential forward feature selection were applied. With the selected predictors, various machine learning models such as Logistic Regression (LR), Support Vector Machine (SVM), K-nearest neighbors (KNN), Gaussian Naïve Bayes (GNB), Extra Trees Classifier (ET) and different ensemble methods such as Random Forest (RF) were trained. Their performance was compared using the receiver operator characteristic (ROC) area-under-curve (AUC), as well as sensitivity, block sensitivity and specificity. In addition, sampling methods were compared: Over- and undersampling as compensation for the expected unbalanced training data had a high impact on the ratio of sensitivity and specificity in the classification of the test data, but with regard to AUC, random oversampling and SMOTE (Synthetic Minority Over-sampling) even showed significantly lower values than with non-sampled data. The best model, ET, obtained a mean AUC of 0.79 for mastitis and 0.71 for lameness, respectively, based on testing data from practical conditions and is recommended by us for this type of data, but GNB, LR and RF were only marginally worse, and random oversampling and SMOTE even showed significantly lower values than without sampling. We recommend the use of these models as a benchmark for similar self-learning classification tasks. The classification models presented here retain their interpretability with the ability to present feature importances to the farmer in contrast to the “black box” models of Deep Learning methods.
43
Lürig, Christoph. "Learning Machine Learning with a Game." European Conference on Games Based Learning 16, no. 1 (September 29, 2022): 316–23. http://dx.doi.org/10.34190/ecgbl.16.1.481.
Full textAbstract:
AIs playing strategic games have always fascinated humans. Specifically, the reinforcement learning technique Alpha Zero (D.Silver, 2016) has gained much attention for its capability to play Go, which was hard to crack problem for AI for a long time. Additionally, we see the rise of explainable AI (xAI), which tries to address the problem that many modern AI decision techniques are black-box approaches and incomprehensible to humans. Combining a board game AI for the relatively simple game Connect-Four with explanation techniques offers the possibility of learning something about an AI's inner workings and the game itself. This paper explains how to combine an Alpha-Zero-based AI with known explanation techniques used in supervised learning. Additionally, we combine this with known visualization approaches for trees. Alpha-Zero combines a neuronal network and a Monte-Carlo-Search-Tree. The approach we present in this paper focuses on two explanations. The first explanation is a dynamic analysis of the evolving situation, primarily based on the tree aspect, and works with a radial tree representation (Yee et al., 2001). The second explanation is a static analysis that tries to identify the relevant situation elements using the Lime (Local Interpretable Model Agnostic Explanations) approach (Christoforos Anagnostopoulos, 2020). This technique focuses primarily on the neuronal network aspect. The straightforward application of Lime towards the Monte-Carlo-Search-Tree approach would be too compute-intensive for interactive applications. We suggest a modification to accommodate search trees and sacrifice the model agnosticism specifically. We use a weighted Lasso-based approach on the different board constellations analyzed in the search tree by the neuronal network to get a final static explanation of the situation. Finally, we visually interpret the resulting linear weights from the Lasso analysis on the game board. The implementation is done in Python using the PyGame library for visualization and interaction implementation. We implemented the neuronal networks with PyTorch and the Lasso analysis with Scikit Learn. This paper provides implementation details on an experimental approach to learning something about a game and how machines learn to play a game.
44
Adams, Solomon M., Habiba Feroze, Tara Nguyen, Seenae Eum, Cyrille Cornelio, and Arthur F. Harralson. "Genome Wide Epistasis Study of On-Statin Cardiovascular Events with Iterative Feature Reduction and Selection." Journal of Personalized Medicine 10, no. 4 (November 7, 2020): 212. http://dx.doi.org/10.3390/jpm10040212.
Full textAbstract:
Predicting risk for major adverse cardiovascular events (MACE) is an evidence-based practice that incorporates lifestyle, history, and other risk factors. Statins reduce risk for MACE by decreasing lipids, but it is difficult to stratify risk following initiation of a statin. Genetic risk determinants for on-statin MACE are low-effect size and impossible to generalize. Our objective was to determine high-level epistatic risk factors for on-statin MACE with GWAS-scale data. Controlled-access data for 5890 subjects taking a statin collected from Vanderbilt University Medical Center’s BioVU were obtained from dbGaP. We used Random Forest Iterative Feature Reduction and Selection (RF-IFRS) to select highly informative genetic and environmental features from a GWAS-scale dataset of patients taking statin medications. Variant-pairs were distilled into overlapping networks and assembled into individual decision trees to provide an interpretable set of variants and associated risk. 1718 cases who suffered MACE and 4172 controls were obtained from dbGaP. Pathway analysis showed that variants in genes related to vasculogenesis (FDR = 0.024), angiogenesis (FDR = 0.019), and carotid artery disease (FDR = 0.034) were related to risk for on-statin MACE. We identified six gene-variant networks that predicted odds of on-statin MACE. The most elevated risk was found in a small subset of patients carrying variants in COL4A2, TMEM178B, SZT2, and TBXAS1 (OR = 4.53, p < 0.001). The RF-IFRS method is a viable method for interpreting complex “black-box” findings from machine-learning. In this study, it identified epistatic networks that could be applied to risk estimation for on-statin MACE. Further study will seek to replicate these findings in other populations.
45
Sun, Jia, and Yanrong Jiao. "Enterprise Financial Risk Analysis Based on Improved Model C-Means Clustering Algorithm." Security and Communication Networks 2022 (July 12, 2022): 1–12. http://dx.doi.org/10.1155/2022/1109813.
Full textAbstract:
As a provider of loans to SMEs, banks should prudently examine loan risks while ensuring that they provide loans to SMEs from the perspective of cooperating with policy implementation and controlling their own risks. The existing loan risk measurement tools include multiple discriminant analysis models, multiple regression models, and machine learning methods. Most machine learning methods have higher prediction accuracy than traditional models when using historical data for calculation, but the existence of problems such as overfitting seriously affects the robustness of machine learning methods. A similar method is introduced into the loan default risk prediction of SMEs, and the mean clustering method is used to preset penalty items to reduce overfitting and high accuracy to help banks effectively identify the default probability of SMEs during the loan period. This study will use the mean clustering method to iteratively train 900,000 SME credit records published by the US Small and Medium Business Administration, with 27 dimensions of data provided by Small Business Administration (SBA) to provide partial guarantees. A regression tree evaluates the data, combining the scores of multiple regression trees to produce a final prediction of the probability of credit default on the input data. The research results show that the mean clustering method can effectively improve the prediction accuracy of traditional machine learning methods and multiple linear regression in the scenario of SME loan default prediction and reduce the overfitting and black-box properties. As a supplementary loan default risk measurement tool, it can strengthen the ability of commercial banks to control the risk of loan business and can also promote the development of small- and medium-sized enterprises and the market economy to a certain extent.
46
Kato, Daiki, Kenya Yoshitsugu, Naoki Maeda, Toshiki Hirogaki, Eiichi Aoyama, and Kenichi Takahashi. "Positioning Error Calibration of Industrial Robots Based on Random Forest." International Journal of Automation Technology 15, no. 5 (September 5, 2021): 581–89. http://dx.doi.org/10.20965/ijat.2021.p0581.
Full textAbstract:
Because most industrial robots are taught using the direct teaching and playback method, they are unsuitable for variable production systems. Alternatively, the offline teaching method has limited applications because of the low accuracy of the position and posture of the end-effector. Therefore, many studies have been conducted to calibrate the position and posture. Positioning errors of robots can be divided into kinematic and non-kinematic errors. In some studies, kinematic errors are calibrated by kinematic models, and non-kinematic errors are calibrated by neural networks. However, the factor of the positioning errors has not been identified because the neural network is a black box. In another machine learning method, a random forest is constructed from decision trees, and its structure can be visualized. Therefore, we used a random forest method to construct a calibration model for the positioning errors and to identify the positioning error factors. The proposed calibration method is based on a simulation of many candidate points centered on the target point. A large industrial robot was used, and the 3D coordinates of the end-effector were obtained using a laser tracker. The model predicted the positioning error from end-effector coordinates, joint angles, and joint torques using the random forest method. As a result, the positioning error was predicted with a high accuracy. The random forest analysis showed that joint 2 was the primary factor of the X- and Z-axis errors. This suggests that the air cylinder used as an auxiliary to the servo motor of joint 2, which is unique to large industrial robots, is the error factor. With the proposed calibration, the positioning error norm was reduced at all points.
47
Silvina, Fetmi, Arnis en Yulia, and Noreza Masri. "PEMBERIAN BERBAGAI PUPUK ORGANIK TERHADAP PERTUMBUHAN DAN PRODUKSI BEBERAPA VARIETAS PADI GOGO (Oryza sativa L.) YANG DITANAM DIANTARA TANAMAN KELAPA SAWIT BELUM MENGHASILKAN." DINAMIKA PERTANIAN 33, no. 3 (September 24, 2019): 231–42. http://dx.doi.org/10.25299/dp.2017.vol33(3).3836.
Full textAbstract:
The study aimed to determine the effect of various organic fertilizers on growth and yield of several varieties of upland rice, and the response of upland rice varieties to organic fertilizers under the stands of oil palm trees have not produced (TBM). The researches conducted in March until July 2016. This study was a randomized block design factorial trial with two factors and three replications, the first factor was some upland rice variety consist of ; Inpago 8 (V1), Situ Bagendit (V2), and Inpago 9 (V3) and the second factor was the variety of organic fertilizers such as; without organic fertilizer (BO0), compost of oil palm empty fruit bunches (BO1), chicken manure (BO2), Kirinyuh or green manure (BO3). Data were analyzed by ANOVA and a further test of Duncan’s multiple range test (DNMRT) level of 5%. The results showed that the varieties Situ Bagendit gave the best response to organic fertilizer by weight of dry milled grain at 4.6 tonnes/ha in the giving of chicken manure.
48
Murari, Andrea, Emmanuele Peluso, Michele Lungaroni, Riccardo Rossi, and Michela Gelfusa. "Investigating the Physics of Tokamak Global Stability with Interpretable Machine Learning Tools." Applied Sciences 10, no. 19 (September 24, 2020): 6683. http://dx.doi.org/10.3390/app10196683.
Full textAbstract:
The inadequacies of basic physics models for disruption prediction have induced the community to increasingly rely on data mining tools. In the last decade, it has been shown how machine learning predictors can achieve a much better performance than those obtained with manually identified thresholds or empirical descriptions of the plasma stability limits. The main criticisms of these techniques focus therefore on two different but interrelated issues: poor “physics fidelity” and limited interpretability. Insufficient “physics fidelity” refers to the fact that the mathematical models of most data mining tools do not reflect the physics of the underlying phenomena. Moreover, they implement a black box approach to learning, which results in very poor interpretability of their outputs. To overcome or at least mitigate these limitations, a general methodology has been devised and tested, with the objective of combining the predictive capability of machine learning tools with the expression of the operational boundary in terms of traditional equations more suited to understanding the underlying physics. The proposed approach relies on the application of machine learning classifiers (such as Support Vector Machines or Classification Trees) and Symbolic Regression via Genetic Programming directly to experimental databases. The results are very encouraging. The obtained equations of the boundary between the safe and disruptive regions of the operational space present almost the same performance as the machine learning classifiers, based on completely independent learning techniques. Moreover, these models possess significantly better predictive power than traditional representations, such as the Hugill or the beta limit. More importantly, they are realistic and intuitive mathematical formulas, which are well suited to supporting theoretical understanding and to benchmarking empirical models. They can also be deployed easily and efficiently in real-time feedback systems.
49
Proshkin, B. V., and A. V. Klimov. "SEED PRODUCTIVITY AND DEVELOPMENT OF PLANTLETS POPULUS × JRTYSCHENSIS CH. Y. YANG." Bulletin of NSAU (Novosibirsk State Agrarian University), no. 2 (July 23, 2019): 51–57. http://dx.doi.org/10.31677/2072-6724-2019-51-2-51-57.
Full textAbstract:
The research explores the seed productivity and plantlets growth in the free pollination of the natural hybrid taxon P. × jrtyschensis. Fruits of P. × jrtyschensis were selected from four plants that grow in the collection ofResearchCenter“EducationalBotanical Garden” ofKemerovoStateUniversity. Four P. nigra model trees, randomly selected from theTomRiverfloodplain population, were applied as a control group. The authors used 30 fruit-bearing amentumsfrom each model. The researchers measured set of fruit (capsule); number of ovules per fruit; number of seeds per fruit; set of seeds.. Laboratory germination was determined by sowing Petri dishes on wet filter paper. The authors found out sowing germination by sowing 100 seeds in a box with soil and drainage. The energy of germination was determined on the second day while germination - on the fifth day. P. × jrtyschensis is characterized by a lower level of seed productivity (15-30%) compared to P. nigra. In terms of laboratory germination of seeds, the descendants of hybrids surpassed many P. nigra models, but their soil germination was 20-30% lower than that of black poplar. The observed variability in reproductive indices of both P. × jrtyschensis and P. nigra is mainly caused by specific features of their genotypes. Plantlets being developed, the authors observed no significant differences among the descendants of P. nigra and hybrids. The researchers highlighted plantlets that can stop growing and even more abnormal plants with one, three or four seeds in P. × jrtyschensis. This may be caused by underdevelopment of hypocotyl or germ root. The authors observed breaches in development of P. nigra just once. They outline high plantlets destruction when sowing hybrids on the first day after germination The share of destructed plants within a month (from the beginning of the experiment) reaches 66,0 %, and in P. nigra it does not exceed 40,0 %.
50
JGENTI, Lali, Gia BOLKVADZE, Maradi IAKOBADZE, and Inga DIASAMIDZE. "Flora and Fauna Conservation in Machakhela National Park Georgia." Eurasia Proceedings of Health, Environment and Life Sciences 5 (November 2, 2022): 35–39. http://dx.doi.org/10.55549/ephels.32.
Full textAbstract:
Biodiversity conservation is the protection and management of biodiversity to obtain resources for sustainable development. The diverse topography and climate has provided conditions to develop a remarkably wide array of landscapes and plant formations. They include two features of plants and plant associations that date back millions of years: the Colchic refugium in the eastern Black Sea basin and the Hyrcanic region on the southern Caspian Sea coast. These “refugia”/refugial forests harbour many locally endemic plants - species that are found nowhere else. They include relict and endemic oaks (such as Quercus imeretina, Q. hartwissiana), Medvedev’s birch (Betula medwedewii), Ungern’s and Smirow’s rhododendrons (Rhododendron ungernii, R. smirnowii) in the Colchic. Machakhela National Park is located 30 km away from Batumi in the foothills of the Lesser Caucasus. Close to the Turkish border, Machakhela expands the protection of the unique ecosystems of the Colchic forests – rich tropical and sub-tropical habitats (temperate rain forests with peat bogs) which contain unrivaled biodiversity, and are rich in relics of the tertiary period: Colchic bot box, chestnut, nut, hazel-nut, and bot trees abound. Trails are being developed and since this park has only been recently established, you can still be one of the first to witness its wet beauty. At the same time these unique forests can mostly be classified as temperate rainforests, due to the same principal reasons as for other temperate rainforest regions: relevant slopes of barriermountains located along coastlines that trap a large portion of the humidity from sea air masses. Montane barriers also contribute to a warm and humid climate that has been present since the late Tertiary and is the primary reason that the Caucasus has acted as a shelter for humid- and warm-requiring (hygro-thermophilous) relicts during the ice age.