Artigos de revistas sobre o tema "Convolutional code-word"
Siga este link para ver outros tipos de publicações sobre o tema: Convolutional code-word.
Crie uma referência precisa em APA, MLA, Chicago, Harvard, e outros estilos
Veja os 34 melhores artigos de revistas para estudos sobre o assunto "Convolutional code-word".
Ao lado de cada fonte na lista de referências, há um botão "Adicionar à bibliografia". Clique e geraremos automaticamente a citação bibliográfica do trabalho escolhido no estilo de citação de que você precisa: APA, MLA, Harvard, Chicago, Vancouver, etc.
Você também pode baixar o texto completo da publicação científica em formato .pdf e ler o resumo do trabalho online se estiver presente nos metadados.
Veja os artigos de revistas das mais diversas áreas científicas e compile uma bibliografia correta.
1
Sidorenko, A. A. "Decoding of the turbo code created on the basis of the block code using the syndrome grid". Journal of Physics: Conference Series 2388, n.º 1 (1 de dezembro de 2022): 012029. http://dx.doi.org/10.1088/1742-6596/2388/1/012029.
Texto completo da fonteResumo:
Abstract Turbo codes came closest to the Shannon limit. They can be built on the basis of both block and convolutional codes. When decoding a block code using a syndrome lattice, it is possible to use well-studied algorithms for decoding convolutional codes with soft input and soft output. This publication demonstrates the application of the maximum a posteriori information decoding algorithm. The process of decoding a code sequence of a turbo code generated by an encoder consisting of two encoders of a block code and an interleaving device, in the presence of three errors in the received code word, is considered in detail.
2
Wang, Yilin, Siqing Xue e Jun Song. "A Malicious Webpage Detection Method Based on Graph Convolutional Network". Mathematics 10, n.º 19 (25 de setembro de 2022): 3496. http://dx.doi.org/10.3390/math10193496.
Texto completo da fonteResumo:
In recent years, with the rapid development of the Internet and information technology, video websites, shopping websites, and other portals have grown rapidly. However, malicious webpages can disguise themselves as benign websites and steal users’ private information, which seriously threatens network security. Current detection methods for malicious webpages do not fully utilize the syntactic and semantic information in the web source code. In this paper, we propose a GCN-based malicious webpage detection method (GMWD), which constructs a text graph to describe and then a GCN model to learn the syntactic and semantic correlations within and between webpage source codes. We replace word nodes in the text graph with phrase nodes to better maintain the syntactic and semantic integrity of the webpage source code. In addition, we use the URL links appearing in the source code as auxiliary detection information to further improve the detection accuracy. The experiments showed that the proposed method can achieve 99.86% accuracy and a 0.137% false negative rate, achieving a better performance than other related malicious webpage detection methods.
3
Ramanna, Dasari, e V. Ganesan. "Low-Power VLSI Implementation of Novel Hybrid Adaptive Variable-Rate and Recursive Systematic Convolutional Encoder for Resource Constrained Wireless Communication Systems". International Journal of Electrical and Electronics Research 10, n.º 3 (30 de setembro de 2022): 523–28. http://dx.doi.org/10.37391/ijeer.100320.
Texto completo da fonteResumo:
In the modern wireless communication system, digital technology has tremendous growth, and all the communication channels are slowly moving towards digital form. Wireless communication has to provide the reliable and efficient transfer of information between transmitter and receiver over a wireless channel. The channel coding technique is the best practical approach to delivering reliable communication for the end-users. Many conventional encoder and decoder units are used as error detection and correction codes in the digital communication system to overcome the multiple transient errors. The proposed convolutional encoder consists of both Recursive Systematic Convolutional (RSC) Encoder and Adaptive Variable-Rate Convolutional (AVRC) encoder. Adaptive Variable-Rate Convolutional encoder improves the bit error rate performance and is more suitable for a power-constrained wireless system to transfer the data. Recursive Systematic Convolutional encoder also reduces the bit error rate and improves the throughput by employing the trellis termination strategy. Here, AVRC encoder ultimately acquires the channel state information and feeds the data into a fixed rate convolutional encoder and rate adaptor followed by a buffer device. A hybrid encoder combines the AVRC encoder and RSC encoder output serially and parallel, producing the solid encoded data for the modulator in the communication system. A modified turbo code is also obtained by placing interleaver between the two encoder units and building the stronger code word for the system. Finally, the conventional encoder system is compared and analyzed with the proposed method regarding the number of LUT’s, gates, clock cycle, slices, area, power, bit error rate, and throughput.
4
Et.al, Vishaal Saravanan. "Automated Web Design And Code Generation Using Deep Learning". Turkish Journal of Computer and Mathematics Education (TURCOMAT) 12, n.º 6 (10 de abril de 2021): 364–73. http://dx.doi.org/10.17762/turcomat.v12i6.1401.
Texto completo da fonteResumo:
Excited by ground-breaking progress in automatic code generation, machine translation, and computer vision, further simplify web design workflow by making it easier and productive. A Model architecture is proposed for the generation of static web templates from hand-drawn images. The model pipeline uses the word-embedding technique succeeded by long short-term memory (LSTM) for code snippet prediction. Also, canny edge detection algorithm fitted with VGG19 convolutional neural net (CNN) and attention-based LSTM for web template generation. Extracted features are concatenated, and a terminal LSTM with a SoftMax function is called for final prediction. The proposed model is validated with a benchmark based on the BLUE score, and performance improvement is compared with the existing image generation algorithms.
5
Farid, Ahmed Bahaa, Enas Mohamed Fathy, Ahmed Sharaf Eldin e Laila A. Abd-Elmegid. "Software defect prediction using hybrid model (CBIL) of convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM)". PeerJ Computer Science 7 (16 de novembro de 2021): e739. http://dx.doi.org/10.7717/peerj-cs.739.
Texto completo da fonteResumo:
In recent years, the software industry has invested substantial effort to improve software quality in organizations. Applying proactive software defect prediction will help developers and white box testers to find the defects earlier, and this will reduce the time and effort. Traditional software defect prediction models concentrate on traditional features of source code including code complexity, lines of code, etc. However, these features fail to extract the semantics of source code. In this research, we propose a hybrid model that is called CBIL. CBIL can predict the defective areas of source code. It extracts Abstract Syntax Tree (AST) tokens as vectors from source code. Mapping and word embedding turn integer vectors into dense vectors. Then, Convolutional Neural Network (CNN) extracts the semantics of AST tokens. After that, Bidirectional Long Short-Term Memory (Bi-LSTM) keeps key features and ignores other features in order to enhance the accuracy of software defect prediction. The proposed model CBIL is evaluated on a sample of seven open-source Java projects of the PROMISE dataset. CBIL is evaluated by applying the following evaluation metrics: F-measure and area under the curve (AUC). The results display that CBIL model improves the average of F-measure by 25% compared to CNN, as CNN accomplishes the top performance among the selected baseline models. In average of AUC, CBIL model improves AUC by 18% compared to Recurrent Neural Network (RNN), as RNN accomplishes the top performance among the selected baseline models used in the experiments.
6
Hsu, Jia-Lien, Teng-Jie Hsu, Chung-Ho Hsieh e Anandakumar Singaravelan. "Applying Convolutional Neural Networks to Predict the ICD-9 Codes of Medical Records". Sensors 20, n.º 24 (11 de dezembro de 2020): 7116. http://dx.doi.org/10.3390/s20247116.
Texto completo da fonteResumo:
The International Statistical Classification of Disease and Related Health Problems (ICD) is an international standard system for categorizing and reporting diseases, injuries, disorders, and health conditions. Most previously-proposed disease predicting systems need clinical information collected by the medical staff from the patients in hospitals. In this paper, we propose a deep learning algorithm to classify disease types and identify diagnostic codes by using only the subjective component of progress notes in medical records. In this study, we have a dataset, consisting of about one hundred and sixty-eight thousand medical records, from a medical center, collected during 2003 and 2017. First, we apply standard text processing procedures to parse the sentences and word embedding techniques for vector representations. Next, we build a convolution neural network model on the medical records to predict the ICD-9 code by using a subjective component of the progress note. The prediction performance is evaluated by ten-fold cross-validation and yields an accuracy of 0.409, recall of 0.409 and precision of 0.436. If we only consider the “chapter match” of ICD-9 code, our model achieves an accuracy of 0.580, recall of 0.580, and precision of 0.582. Since our diagnostic code prediction model is solely based on subjective components (mainly, patients’ self-report descriptions), the proposed approach could serve as a remote and self-diagnosis assistance tool, prior to seeking medical advice or going to the hospital. In addition, our work may be used as a primary evaluation tool for discomfort in the rural area where medical resources are restricted.
7
Banerjee, Suman, e Mitesh M. Khapra. "Graph Convolutional Network with Sequential Attention for Goal-Oriented Dialogue Systems". Transactions of the Association for Computational Linguistics 7 (novembro de 2019): 485–500. http://dx.doi.org/10.1162/tacl_a_00284.
Texto completo da fonteResumo:
Domain-specific goal-oriented dialogue systems typically require modeling three types of inputs, namely, (i) the knowledge-base associated with the domain, (ii) the history of the conversation, which is a sequence of utterances, and (iii) the current utterance for which the response needs to be generated. While modeling these inputs, current state-of-the-art models such as Mem2Seq typically ignore the rich structure inherent in the knowledge graph and the sentences in the conversation context. Inspired by the recent success of structure-aware Graph Convolutional Networks (GCNs) for various NLP tasks such as machine translation, semantic role labeling, and document dating, we propose a memory-augmented GCN for goal-oriented dialogues. Our model exploits (i) the entity relation graph in a knowledge-base and (ii) the dependency graph associated with an utterance to compute richer representations for words and entities. Further, we take cognizance of the fact that in certain situations, such as when the conversation is in a code-mixed language, dependency parsers may not be available. We show that in such situations we could use the global word co-occurrence graph to enrich the representations of utterances. We experiment with four datasets: (i) the modified DSTC2 dataset, (ii) recently released code-mixed versions of DSTC2 dataset in four languages, (iii) Wizard-of-Oz style CAM676 dataset, and (iv) Wizard-of-Oz style MultiWOZ dataset. On all four datasets our method outperforms existing methods, on a wide range of evaluation metrics.
8
Rao, Jinfeng, Wei Yang, Yuhao Zhang, Ferhan Ture e Jimmy Lin. "Multi-Perspective Relevance Matching with Hierarchical ConvNets for Social Media Search". Proceedings of the AAAI Conference on Artificial Intelligence 33 (17 de julho de 2019): 232–40. http://dx.doi.org/10.1609/aaai.v33i01.3301232.
Texto completo da fonteResumo:
Despite substantial interest in applications of neural networks to information retrieval, neural ranking models have mostly been applied to “standard” ad hoc retrieval tasks over web pages and newswire articles. This paper proposes MP-HCNN (Multi-Perspective Hierarchical Convolutional Neural Network), a novel neural ranking model specifically designed for ranking short social media posts. We identify document length, informal language, and heterogeneous relevance signals as features that distinguish documents in our domain, and present a model specifically designed with these characteristics in mind. Our model uses hierarchical convolutional layers to learn latent semantic soft-match relevance signals at the character, word, and phrase levels. A poolingbased similarity measurement layer integrates evidence from multiple types of matches between the query, the social media post, as well as URLs contained in the post. Extensive experiments using Twitter data from the TREC Microblog Tracks 2011–2014 show that our model significantly outperforms prior feature-based as well as existing neural ranking models. To our best knowledge, this paper presents the first substantial work tackling search over social media posts using neural ranking models. Our code and data are publicly available.1
9
Simistira Liwicki, Foteini, Vibha Gupta, Rajkumar Saini, Kanjar De e Marcus Liwicki. "Rethinking the Methods and Algorithms for Inner Speech Decoding and Making Them Reproducible". NeuroSci 3, n.º 2 (19 de abril de 2022): 226–44. http://dx.doi.org/10.3390/neurosci3020017.
Texto completo da fonteResumo:
This study focuses on the automatic decoding of inner speech using noninvasive methods, such as Electroencephalography (EEG). While inner speech has been a research topic in philosophy and psychology for half a century, recent attempts have been made to decode nonvoiced spoken words by using various brain–computer interfaces. The main shortcomings of existing work are reproducibility and the availability of data and code. In this work, we investigate various methods (using Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU), Long Short-Term Memory Networks (LSTM)) for the detection task of five vowels and six words on a publicly available EEG dataset. The main contributions of this work are (1) subject dependent vs. subject-independent approaches, (2) the effect of different preprocessing steps (Independent Component Analysis (ICA), down-sampling and filtering), and (3) word classification (where we achieve state-of-the-art performance on a publicly available dataset). Overall we achieve a performance accuracy of 35.20% and 29.21% when classifying five vowels and six words, respectively, in a publicly available dataset, using our tuned iSpeech-CNN architecture. All of our code and processed data are publicly available to ensure reproducibility. As such, this work contributes to a deeper understanding and reproducibility of experiments in the area of inner speech detection.
10
Zhang, Min, Yujin Yan, Hai Wang e Wei Zhao. "An Algorithm for Natural Images Text Recognition Using Four Direction Features". Electronics 8, n.º 9 (31 de agosto de 2019): 971. http://dx.doi.org/10.3390/electronics8090971.
Texto completo da fonteResumo:
Irregular text has widespread applications in multiple areas. Different from regular text, irregular text is difficult to recognize because of its various shapes and distorted patterns. In this paper, we develop a multidirectional convolutional neural network (MCN) to extract four direction features to fully describe the textual information. Meanwhile, the character placement possibility is extracted as the weight of the four direction features. Based on these works, we propose the encoder to fuse the four direction features for the generation of feature code to predict the character sequence. The whole network is end-to-end trainable due to using images and word-level labels. The experiments on standard benchmarks, including the IIIT-5K, SVT, CUTE80, and ICDAR datasets, demonstrate the superiority of the proposed method on both regular and irregular datasets. The developed method shows an increase of 1.2% in the CUTE80 dataset and 1.5% in the SVT dataset, and has fewer parameters than most existing methods.
11
Masud, Jakir Hossain Bhuiyan, Chen-Cheng Kuo, Chih-Yang Yeh, Hsuan-Chia Yang e Ming-Chin Lin. "Applying Deep Learning Model to Predict Diagnosis Code of Medical Records". Diagnostics 13, n.º 13 (6 de julho de 2023): 2297. http://dx.doi.org/10.3390/diagnostics13132297.
Texto completo da fonteResumo:
The International Classification of Diseases (ICD) code is a diagnostic classification standard that is frequently used as a referencing system in healthcare and insurance. However, it takes time and effort to find and use the right diagnosis code based on a patient’s medical records. In response, deep learning (DL) methods have been developed to assist physicians in the ICD coding process. Our findings propose a deep learning model that utilized clinical notes from medical records to predict ICD-10 codes. Our research used text-based medical data from the outpatient department (OPD) of a university hospital from January to December 2016. The dataset used clinical notes from five departments, and a total of 21,953 medical records were collected. Clinical notes consisted of a subjective component, objective component, assessment, plan (SOAP) notes, diagnosis code, and a drug list. The dataset was divided into two groups: 90% for training and 10% for test cases. We applied natural language processing (NLP) technique (word embedding, Word2Vector) to process the data. A deep learning-based convolutional neural network (CNN) model was created based on the information presented above. Three metrics (precision, recall, and F-score) were used to calculate the achievement of the deep learning CNN model. Clinically acceptable results were achieved through the deep learning model for five departments (precision: 0.53–0.96; recall: 0.85–0.99; and F-score: 0.65–0.98). With a precision of 0.95, a recall of 0.99, and an F-score of 0.98, the deep learning model performed the best in the department of cardiology. Our proposed CNN model significantly improved the prediction performance for an automated ICD-10 code prediction system based on prior clinical information. This CNN model could reduce the laborious task of manual coding and could assist physicians in making a better diagnosis.
12
Sabbah, Ahmed F., e Abualsoud A. Hanani. "Self-admitted technical debt classification using natural language processing word embeddings". International Journal of Electrical and Computer Engineering (IJECE) 13, n.º 2 (1 de abril de 2023): 2142. http://dx.doi.org/10.11591/ijece.v13i2.pp2142-2155.
Texto completo da fonteResumo:
<p>Recent studies show that it is possible to detect technical dept automatically from source code comments intentionally created by developers, a phenomenon known as self-admitted technical debt. This study proposes a system by which a comment or commit is classified as one of five dept types, namely, requirement, design, defect, test, and documentation. In addition to the traditional term frequency-inverse document frequency (TF-IDF), several word embeddings methods produced by different pre-trained language models were used for feature extraction, such as Word2Vec, GolVe, bidirectional encoder representations from transformers (BERT), and FastText. The generated features were used to train a set of classifiers including naive Bayes (NB), random forest (RF), support vector machines (SVM), and two configurations of convolutional neural network (CNN). Two datasets were used to train and test the proposed systems. Our collected dataset (A-dataset) includes a total of 1,513 comments and commits manually labeled. Additionally, a dataset, consisting of 4,071 labeled comments, used in previous studies (M-dataset) was also used in this study. The RF classifier achieved an accuracy of 0.822 with A-dataset and 0.820 with the M-dataset. CNN with A-dataset achieved an accuracy of 0.838 using BERT features. With M-dataset, the CNN achieves an accuracy of 0.809 and 0.812 with BERT and Word2Vec, respectively.</p>
13
Deng, Lei, Hui Wu, Xuejun Liu e Hui Liu. "DeepD2V: A Novel Deep Learning-Based Framework for Predicting Transcription Factor Binding Sites from Combined DNA Sequence". International Journal of Molecular Sciences 22, n.º 11 (24 de maio de 2021): 5521. http://dx.doi.org/10.3390/ijms22115521.
Texto completo da fonteResumo:
Predicting in vivo protein–DNA binding sites is a challenging but pressing task in a variety of fields like drug design and development. Most promoters contain a number of transcription factor (TF) binding sites, but only a small minority has been identified by biochemical experiments that are time-consuming and laborious. To tackle this challenge, many computational methods have been proposed to predict TF binding sites from DNA sequence. Although previous methods have achieved remarkable performance in the prediction of protein–DNA interactions, there is still considerable room for improvement. In this paper, we present a hybrid deep learning framework, termed DeepD2V, for transcription factor binding sites prediction. First, we construct the input matrix with an original DNA sequence and its three kinds of variant sequences, including its inverse, complementary, and complementary inverse sequence. A sliding window of size k with a specific stride is used to obtain its k-mer representation of input sequences. Next, we use word2vec to obtain a pre-trained k-mer word distributed representation model. Finally, the probability of protein–DNA binding is predicted by using the recurrent and convolutional neural network. The experiment results on 50 public ChIP-seq benchmark datasets demonstrate the superior performance and robustness of DeepD2V. Moreover, we verify that the performance of DeepD2V using word2vec-based k-mer distributed representation is better than one-hot encoding, and the integrated framework of both convolutional neural network (CNN) and bidirectional LSTM (bi-LSTM) outperforms CNN or the bi-LSTM model when used alone. The source code of DeepD2V is available at the github repository.
14
Peng, Min, Chongyang Wang, Yu Shi e Xiang-Dong Zhou. "Efficient End-to-End Video Question Answering with Pyramidal Multimodal Transformer". Proceedings of the AAAI Conference on Artificial Intelligence 37, n.º 2 (26 de junho de 2023): 2038–46. http://dx.doi.org/10.1609/aaai.v37i2.25296.
Texto completo da fonteResumo:
This paper presents a new method for end-to-end Video Question Answering (VideoQA), aside from the current popularity of using large-scale pre-training with huge feature extractors. We achieve this with a pyramidal multimodal transformer (PMT) model, which simply incorporates a learnable word embedding layer, a few convolutional and transformer layers. We use the anisotropic pyramid to fulfill video-language interactions across different spatio-temporal scales. In addition to the canonical pyramid, which includes both bottom-up and top-down pathways with lateral connections, novel strategies are proposed to decompose the visual feature stream into spatial and temporal sub-streams at different scales and implement their interactions with the linguistic semantics while preserving the integrity of local and global semantics. We demonstrate better or on-par performances with high computational efficiency against state-of-the-art methods on five VideoQA benchmarks. Our ablation study shows the scalability of our model that achieves competitive results for text-to-video retrieval by leveraging feature extractors with reusable pre-trained weights, and also the effectiveness of the pyramid. Code available at: https://github.com/Trunpm/PMT-AAAI23.
15
Ryzhova, Anna Romanivna, e Yurii Oleksiiovych Onykiienko. "Analysis of the Microcontroller Resources Using Specifics for Speech Recognition". Microsystems, Electronics and Acoustics 27, n.º 2 (21 de agosto de 2022): 265406–1. http://dx.doi.org/10.20535/2523-4455.mea.265406.
Texto completo da fonteResumo:
The use of neural networks for information recognition, in particular, voice, expands the functional capabilities of embedded systems on microcontrollers. But it is necessary to take into account the limitations of the microcontroller resources. The purpose of the work is to analyze the impact of voice processing parameters and neural network architecture on the degree of microcontroller resources usage. To do this, a database of samples of the keyword, samples of other words and voices, and samples of noise are created, the probability of recognizing the keyword among other words and noises is evaluated, the dependence of the amount of memory used on the microcontroller and the decision-making time on the number MFC coefficients is established, the dependence of the amount of used memory of the microcontroller and the decision-making time on the type of convolutional neural network is established also. During the experiment, the Arduino Nano 33 BLE Sense development board was used. The neural network model was built and trained on the Edge Impulse software platform. To conduct the experiment, three groups of data with the names "hello", "unknown", "noise" were created. The group "hello" contains 94 examples of the word "hello" in English, spoken by a female voice. The "unknown" group contains 167 examples of other words pronounced by both female and male voices. The "noise" group contains 166 samples of noise and random sounds. According to Edge Impulse's recommendation, 80% of the samples from each of the data groups were used to train the neural network model, and 20% of the samples were used for testing. Analysis of the results shows that with an increase in the number of MFC coefficients and, accordingly, the accuracy of keyword recognition, the amount of program memory occupied by the code increases by 480 bytes (less than 1%). For the nRF52840 microcontroller, this is not a significant increase. The amount of RAM used during the experiment did not change. Although the calculation time of the accuracy of the code word definition increased by only 14 ms (less than 5%) with the increase in the number of MFC coefficients, the calculation procedure is quite long (approximately 0.3 s) compared to the sound sample length of 1 s. This can be a certain limitation when processing a sound signal with 32-bit microcontrollers. To analyze phrases or sentences, it is necessary to use more powerful microcontrollers or microprocessors. Based on the results of experimental research, it can be stated that the computing resources of 32-bit microcontrollers are quite sufficient for recognizing voice commands with the possibility of pre-digital processing of the sound signal, in particular, the use of low-frequency cepstral coefficients. The selection of the number of coefficients does not significantly affect the amount of used FLASH and RAM memory of the nRF52840 microcontroller. The comparison results show the superiority of the 2D network in the accuracy of the keyword definition for both 12 and 13 MFC coefficients. The use of a one-dimensional convolutional neural network for voice sample recognition in the conducted experiment provides memory savings of approximately 5%. The quality of keyword recognition with the number of MFC coefficients of 12 is approximately 0.7. For 17 MFC coefficients, the recognition quality is already 0.97. The amount of RAM used in the case of the 2D network has decreased slightly. Voice sample processing time for both types of networks is practically the same. Thus, 1D convolutional neural networks have certain advantages in microcontroller applications for voice processing and recognition. The limitation of voice recognition on the microcontroller is the sufficiently long processing time of the sound sample (approximately 0.3 s) with the duration of the sample itself being 1 s, this can be explained by a sufficiently low clock frequency of 64 MHz. Increasing the clock frequency will reduce the calculation time.
16
Lin, Yang, Xiaoyong Pan e Hong-Bin Shen. "lncLocator 2.0: a cell-line-specific subcellular localization predictor for long non-coding RNAs with interpretable deep learning". Bioinformatics 37, n.º 16 (25 de fevereiro de 2021): 2308–16. http://dx.doi.org/10.1093/bioinformatics/btab127.
Texto completo da fonteResumo:
Abstract Motivation Long non-coding RNAs (lncRNAs) are generally expressed in a tissue-specific way, and subcellular localizations of lncRNAs depend on the tissues or cell lines that they are expressed. Previous computational methods for predicting subcellular localizations of lncRNAs do not take this characteristic into account, they train a unified machine learning model for pooled lncRNAs from all available cell lines. It is of importance to develop a cell-line-specific computational method to predict lncRNA locations in different cell lines. Results In this study, we present an updated cell-line-specific predictor lncLocator 2.0, which trains an end-to-end deep model per cell line, for predicting lncRNA subcellular localization from sequences. We first construct benchmark datasets of lncRNA subcellular localizations for 15 cell lines. Then we learn word embeddings using natural language models, and these learned embeddings are fed into convolutional neural network, long short-term memory and multilayer perceptron to classify subcellular localizations. lncLocator 2.0 achieves varying effectiveness for different cell lines and demonstrates the necessity of training cell-line-specific models. Furthermore, we adopt Integrated Gradients to explain the proposed model in lncLocator 2.0, and find some potential patterns that determine the subcellular localizations of lncRNAs, suggesting that the subcellular localization of lncRNAs is linked to some specific nucleotides. Availabilityand implementation The lncLocator 2.0 is available at www.csbio.sjtu.edu.cn/bioinf/lncLocator2 and the source code can be found at https://github.com/Yang-J-LIN/lncLocator2.
17
Tang, Xu, Xiangrong Zhang, Fang Liu e Licheng Jiao. "Unsupervised Deep Feature Learning for Remote Sensing Image Retrieval". Remote Sensing 10, n.º 8 (7 de agosto de 2018): 1243. http://dx.doi.org/10.3390/rs10081243.
Texto completo da fonteResumo:
Due to the specific characteristics and complicated contents of remote sensing (RS) images, remote sensing image retrieval (RSIR) is always an open and tough research topic in the RS community. There are two basic blocks in RSIR, including feature learning and similarity matching. In this paper, we focus on developing an effective feature learning method for RSIR. With the help of the deep learning technique, the proposed feature learning method is designed under the bag-of-words (BOW) paradigm. Thus, we name the obtained feature deep BOW (DBOW). The learning process consists of two parts, including image descriptor learning and feature construction. First, to explore the complex contents within the RS image, we extract the image descriptor in the image patch level rather than the whole image. In addition, instead of using the handcrafted feature to describe the patches, we propose the deep convolutional auto-encoder (DCAE) model to deeply learn the discriminative descriptor for the RS image. Second, the k-means algorithm is selected to generate the codebook using the obtained deep descriptors. Then, the final histogrammic DBOW features are acquired by counting the frequency of the single code word. When we get the DBOW features from the RS images, the similarities between RS images are measured using L1-norm distance. Then, the retrieval results can be acquired according to the similarity order. The encouraging experimental results counted on four public RS image archives demonstrate that our DBOW feature is effective for the RSIR task. Compared with the existing RS image features, our DBOW can achieve improved behavior on RSIR.
18
Panda, Binayak, Sudhanshu Shekhar Bisoyi e Sidhanta Panigrahy. "An ensemble approach for imbalanced multiclass malware classification using 1D-CNN". PeerJ Computer Science 9 (14 de novembro de 2023): e1677. http://dx.doi.org/10.7717/peerj-cs.1677.
Texto completo da fonteResumo:
Dependence on the internet and computer programs demonstrates the significance of computer programs in our day-to-day lives. Such demands motivate malware developers to create more malware, both in terms of quantity and variety. Researchers are constantly faced with hurdles while attempting to protect themselves from potential hazards and risks due to malware authors’ usage of code obfuscation techniques. Metamorphic and polymorphic variations are easily able to elude the widely utilized signature-based detection procedures. Researchers are more interested in deep learning approaches than machine learning techniques to analyze the behavior of such a vast number of virus variants. Researchers have been drawn to the categorization of malware within itself in addition to the classification of malware against benign programs to examine the behavioral differences between them. In order to investigate the relationship between the application programming interface (API) calls throughout API sequences and classify them, this work uses the one-dimensional convolutional neural network (1D-CNN) model to solve a multiclass classification problem. On API sequences, feature vectors for distinctive APIs are created using the Word2Vec word embedding approach and the skip-gram model. The one-vs.-rest approach is used to train 1D-CNN models to categorize malware, and all of them are then combined with a suggested ModifiedSoftVoting algorithm to improve classification. On the open benchmark dataset Mal-API-2019, the suggested ensembled 1D-CNN architecture captures improved evaluation scores with an accuracy of 0.90, a weighted average F1-score of 0.90, and an AUC score of more than 0.96 for all classes of malware.
19
Masud, Jakir Hossain Bhuiyan, Chiang Shun, Chen-Cheng Kuo, Md Mohaimenul Islam, Chih-Yang Yeh, Hsuan-Chia Yang e Ming-Chin Lin. "Deep-ADCA: Development and Validation of Deep Learning Model for Automated Diagnosis Code Assignment Using Clinical Notes in Electronic Medical Records". Journal of Personalized Medicine 12, n.º 5 (28 de abril de 2022): 707. http://dx.doi.org/10.3390/jpm12050707.
Texto completo da fonteResumo:
Currently, the International Classification of Diseases (ICD) codes are being used to improve clinical, financial, and administrative performance. Inaccurate ICD coding can lower the quality of care, and delay or prevent reimbursement. However, selecting the appropriate ICD code from a patient’s clinical history is time-consuming and requires expert knowledge. The rapid spread of electronic medical records (EMRs) has generated a large amount of clinical data and provides an opportunity to predict ICD codes using deep learning models. The main objective of this study was to use a deep learning-based natural language processing (NLP) model to accurately predict ICD-10 codes, which could help providers to make better clinical decisions and improve their level of service. We retrospectively collected clinical notes from five outpatient departments (OPD) from one university teaching hospital between January 2016 and December 2016. We applied NLP techniques, including global vectors, word to vectors, and embedding techniques to process the data. The dataset was split into two independent training and testing datasets consisting of 90% and 10% of the entire dataset, respectively. A convolutional neural network (CNN) model was developed, and the performance was measured using the precision, recall, and F-score. A total of 21,953 medical records were collected from 5016 patients. The performance of the CNN model for the five different departments was clinically satisfactory (Precision: 0.50~0.69 and recall: 0.78~0.91). However, the CNN model achieved the best performance for the cardiology department, with a precision of 69%, a recall of 89% and an F-score of 78%. The CNN model for predicting ICD-10 codes provides an opportunity to improve the quality of care. Implementing this model in real-world clinical settings could reduce the manual coding workload, enhance the efficiency of clinical coding, and support physicians in making better clinical decisions.
20
Wang, Zhe, Sen Xiang, Chao Zhou e Qing Xu. "DeepMethylation: a deep learning based framework with GloVe and Transformer encoder for DNA methylation prediction". PeerJ 11 (25 de setembro de 2023): e16125. http://dx.doi.org/10.7717/peerj.16125.
Texto completo da fonteResumo:
DNA methylation is a crucial topic in bioinformatics research. Traditional wet experiments are usually time-consuming and expensive. In contrast, machine learning offers an efficient and novel approach. In this study, we propose DeepMethylation, a novel methylation predictor with deep learning. Specifically, the DNA sequence is encoded with word embedding and GloVe in the first step. After that, dilated convolution and Transformer encoder are utilized to extract the features. Finally, full connection and softmax operators are applied to predict the methylation sites. The proposed model achieves an accuracy of 97.8% on the 5mC dataset, which outperforms state-of-the-art methods. Furthermore, our predictor exhibits good generalization ability as it achieves an accuracy of 95.8% on the m1A dataset. To ease access for other researchers, our code is publicly available at https://github.com/sb111169/tf-5mc.
21
Kutova, О. V., e R. V. Sahaidak-Nikitiuk. "Optimization methods for multi-criteria decisions in pharmacy". Social Pharmacy in Health Care 9, n.º 4 (17 de novembro de 2023): 3–10. http://dx.doi.org/10.24959/sphhcj.23.302.
Texto completo da fonteResumo:
Optimization methods for multi-criteria decisions in pharmacy
In pharmaceutical technological research, the determination of the quantitative composition of granules is considered as a task of multi-criteria selection. Today, to solve this problem, the regression analysis and multi-criteria optimization methods are widely used; they are based on mathematical models obtained for the object under study.
Aim. To identify a decision-making method in a multi-criteria space that is effective for use in pharmaceutical technology research with quantitative factors.
Materials and methods. The study uses tools of the popular computer mathematics system Mathcad (MathSoft Ins., USA) to automate the solution of mathematical problems. To automatically search for the type and coefficients of regression equations, the MS Excel application was used, namely: the data analysis package (regression analysis). The MS Word processor was used to edit the code.
Results. A variety of approaches to the formalization of the multi-criteria optimization task have been studied. The optimal quantitative content of excipients when developing the granule technology has been found using two different optimization criteria, which are formed according to different methodical approaches. The method proposed does not provide for the mandatory introduction of gradation of individual criteria or their weighting factors.
Conclusions. As a result of the comparison of multi-criteria optimization methods, the effectiveness of the decision-making method in the multi-criteria space has been shown; it synthesizes a mathematical procedure related to the vector of criteria and is based on determining the ideal point and introducing the concept of a norm into the space of functionals; it has not been mathematically proven, but it is practically useful decision-making algorithm compared to the mathematical method of convolution of criteria. The optimization method proposed has advantages that are manifested in the possibility of using a relatively simple mathematical apparatus and simplified logic of obtaining a solution.
22
Al-Aidaroos, Ahmed Sheikh, e Sara Mohammed Bamzahem. "The Impact of GloVe and Word2Vec Word-Embedding Technologies on Bug Localization with Convolutional Neural Network". International Journal of Science and Engineering Applications, 16 de janeiro de 2023, 108–11. http://dx.doi.org/10.7753/ijsea1201.1035.
Texto completo da fonteResumo:
In the field of software engineering, software quality assurance faces many challenges, including overcoming the problem of identifying errors in the source code. Finding the location of the error in the source code is a very important process, as is taking advantage of the semantic information available in the bug reports and the source code to find the similarities between them, using modern techniques called word embedding. This study aims to demonstrate how GloVe and Doc2Vec word-embedding technologies affect bug localization accuracy and performance. Therefore, this study proposes to adapt DeepLoc by using GloVe embedding techniques to process the source code instead of Word2vec and using Word2vec embedding techniques to process the bug report instead of Sent2Vec. AspectJ represents the large dataset, which contains many bug reports, while SWT's small dataset contains fewer bug reports. Experimental results show that the improved DeepLoc on SWT achieves 0.60 and 0.72 MAP and MRR, respectively. While the improved DeepLoc on AspectJ achieves 0.17 and 0.27 MAP and MRR, respectively. The results of the improved DeepLoc should be compared using two advanced models from previous studies: DeepLoc, DeepLocator.
23
Kumar, Abhinav, Sunil Saumya e Ashish Singh. "Detecting Dravidian Offensive Posts in MIoT: A Hybrid Deep Learning Framework". ACM Transactions on Asian and Low-Resource Language Information Processing, 24 de abril de 2023. http://dx.doi.org/10.1145/3592602.
Texto completo da fonteResumo:
Hate speech and Offensive Posts (OP) detection on Smart Multimedia Internet of Things (MIoT) have been an active issue for researchers. MIoT media texts in non-native English-speaking countries are often code-mixed or script mixed/switched. This paper proposes an ensemble-based Deep Learning (DL) framework comprised of a Convolutional Neural Network (CNN) and a Dense Neural Network (DNN) for identifying hate and OP in Malayalam Code-Mixed (MCM), Tamil Code-Mixed (TCM), and Malayalam Script-Mixed (MSM) MIoT media postings. Word-level and character-level features are utilized in the convolutional neural network. In contrast, the dense neural network uses character-level Term Frequency-Inverse Document Frequency (TF-IDF) features. The inclusion of character-level features in the proposed ensemble framework resulted in state-of-the-art performance for TCM and MCM datasets, with weighted F 1 -score of 0.91 and 0.78, respectively, and comparable performance for MSM posts, with a weighted F 1 -score of 0.95.
24
Gao, Wei, Jian Wu e Guandong Xu. "Detecting Duplicate Questions in Stack Overflow via Source Code Modeling". International Journal of Software Engineering and Knowledge Engineering, 23 de março de 2022, 1–29. http://dx.doi.org/10.1142/s0218194022500073.
Texto completo da fonteResumo:
Stack Overflow is one of the most popular Question-Answering sites for programmers. However, it faces the problem of question duplication, where newly created questions are identical to previous questions. Existing works on duplicate question detection in Stack Overflow extract a set of textual features on the question pairs and use supervised learning approaches to classify duplicate question pairs. However, they do not consider the source code information in the questions. While in some cases, the intention of a question is mainly represented by the source code. In this paper, we aim to learn the semantics of a question by combining both text features and source code features. We use word embedding and convolutional neural networks to extract textual features from questions to overcome the lexical gap issue. We use tree-based convolutional neural networks to extract structural and semantic features from source code. In addition, we perform multi-task learning by combining the duplication question detection task with a question tag prediction side task. We conduct extensive experiments on the Stack Overflow dataset and show that our approach can detect duplicate questions with higher recall and MRR compared with baseline approaches on Python and Java programming languages.
25
Hannagan, T., A. Agrawal, L. Cohen e S. Dehaene. "Emergence of a compositional neural code for written words: Recycling of a convolutional neural network for reading". Proceedings of the National Academy of Sciences 118, n.º 46 (8 de novembro de 2021). http://dx.doi.org/10.1073/pnas.2104779118.
Texto completo da fonteResumo:
Significance Learning to read results in the formation of a specialized region in the human ventral visual cortex. This region, the visual word form area (VWFA), responds selectively to written words more than to other visual stimuli. However, how neural circuits at this site implement an invariant recognition of written words remains unknown. Here, we show how an artificial neural network initially designed for object recognition can be retrained to recognize words. Once literate, the network develops a sparse neuronal representation of words that replicates several known aspects of the cognitive neuroscience of reading and leads to precise predictions concerning how a small set of neurons implement the orthographic stage of reading acquisition using a compositional neural code.
26
Al-Malla, Muhammad Abdelhadie, Assef Jafar e Nada Ghneim. "Image captioning model using attention and object features to mimic human image understanding". Journal of Big Data 9, n.º 1 (14 de fevereiro de 2022). http://dx.doi.org/10.1186/s40537-022-00571-w.
Texto completo da fonteResumo:
AbstractImage captioning spans the fields of computer vision and natural language processing. The image captioning task generalizes object detection where the descriptions are a single word. Recently, most research on image captioning has focused on deep learning techniques, especially Encoder-Decoder models with Convolutional Neural Network (CNN) feature extraction. However, few works have tried using object detection features to increase the quality of the generated captions. This paper presents an attention-based, Encoder-Decoder deep architecture that makes use of convolutional features extracted from a CNN model pre-trained on ImageNet (Xception), together with object features extracted from the YOLOv4 model, pre-trained on MS COCO. This paper also introduces a new positional encoding scheme for object features, the “importance factor”. Our model was tested on the MS COCO and Flickr30k datasets, and the performance is compared to performance in similar works. Our new feature extraction scheme raises the CIDEr score by 15.04%. The code is available at: https://github.com/abdelhadie-almalla/image_captioning
27
Latha, Yarasu Madhavi, e B. Srinivasa Rao. "Amazon product recommendation system based on a modified convolutional neural network". ETRI Journal, 19 de março de 2024. http://dx.doi.org/10.4218/etrij.2023-0162.
Texto completo da fonteResumo:
AbstractIn e‐commerce platforms, sentiment analysis on an enormous number of user reviews efficiently enhances user satisfaction. In this article, an automated product recommendation system is developed based on machine and deep‐learning models. In the initial step, the text data are acquired from the Amazon Product Reviews dataset, which includes 60 000 customer reviews with 14 806 neutral reviews, 19 567 negative reviews, and 25 627 positive reviews. Further, the text data denoising is carried out using techniques such as stop word removal, stemming, segregation, lemmatization, and tokenization. Removing stop‐words (duplicate and inconsistent text) and other denoising techniques improves the classification performance and decreases the training time of the model. Next, vectorization is accomplished utilizing the term frequency–inverse document frequency technique, which converts denoised text to numerical vectors for faster code execution. The obtained feature vectors are given to the modified convolutional neural network model for sentiment analysis on e‐commerce platforms. The empirical result shows that the proposed model obtained a mean accuracy of 97.40% on the APR dataset.
28
"Hate and Aggression Detection in Social Media over Hindi English language". International Journal of Software Science and Computational Intelligence 14, n.º 1 (janeiro de 2022): 0. http://dx.doi.org/10.4018/ijssci.300357.
Texto completo da fonteResumo:
In today’s time, everyone is familiar with social media platforms. It is quite helpful in connecting people. It has many advantages and some disadvantages too. Currently, in social media, hate and aggression have become a huge problem. On these platforms, many people make inflammatory posts targeting any person or society by using code mixed language, due to which many problems arise in the society. At the current time, much research work is being done on English language-related social media posts. The authors have focused on code mixed language. Authors have also tried to focus on sentences that do not use abusive words but contain hatred-related remarks. In this research, authors have used Natural Language Processing (NLP). Authors have applied Fasttext word embedding to the dataset. Fasttext is a technique of NLP. Deep learning (DL) classification algorithms were applied thereafter. In this research, two classifications have been used i.e. Convolutional Neural Network (CNN) and Bidirectional LSTM (Bi-LSTM).
29
Du, Zhihua, Xiangdong Xiao e Vladimir N. Uversky. "Classification of Chromosomal DNA Sequence Using A Hybrid Deep Learning Architecture". Current Bioinformatics 15 (24 de fevereiro de 2020). http://dx.doi.org/10.2174/1574893615666200224095531.
Texto completo da fonteResumo:
: Chromosomal DNA contains most of the genetic information of eukaryotes and plays an important role in the growth, development and reproduction of living organisms. Most chromosomal DNA sequences are known to wrap around histones, and distinguishing these DNA sequences from ordinary DNA sequences is important for understanding the genetic code of life. The main difficulty behind this problem is the feature selection process. DNA sequences have no explicit features, and the common representation methods, such as one-hot coding, introduced the major drawback of high dimensionality. Recently, deep learning models have been proved to be able to automatically extract useful features from input patterns. In this paper, we present four different deep learning architectures using convolutional neural networks and long short-term memory networks for the purpose of chromosomal DNA sequence classification. Natural language model（Word2vec）was used to generate word embedding of sequence and learn features from it by deep learning. The comparison of these four architectures is carried out on 10 chromosomal DNA datasets. The results show that the architecture of convolutional neural networks combined with long short-term memory networks is superior to other methods in accuracy of chromosomal DNA prediction.
30
M, Ramasamy, Rania Anjum S, V. R. Shree Harini, Sreevidya Bharathan Rajalakshmi e Mr P. Dineshkumar. "Address Recognition System for Postal Service Using Neural Networks". International Journal of Advanced Research in Science, Communication and Technology, 31 de março de 2021, 338–46. http://dx.doi.org/10.48175/ijarsct-918.
Texto completo da fonteResumo:
While most of the Indian industries are in the process of automation, it is a bitter truth that the Indian Postal System is still using manual intervention for its mail sorting and processing. Although for postal automation there are many pieces of work towards street name recognition in non-Indian languages, to the best of our knowledge there is no work on street name recognition in Indian languages. The Automatic Mail Processor (AMP), which we have designed, scans a mail and interprets the imperative fields of the destination address such as the Pin Code, City name, Locality name and the Street name. The interpreted address is subsequently converted into a QR code. The code is reprinted onto the mail which can be read by a low-cost machine. By converting the destination address into a barcode, all of the future sorting processes can be accomplished by using a mechanical machine sorter, which can sort the mails according to the barcode present on them. We used two main approaches to accomplish this task: classifying words directly and character segmentation. For the former, we use Convolutional Neural Network (CNN) with various architectures to train a model that can precisely classify words. We then pass the segmented characters to a R ecurrent Neural Network (RNN) for classification and then reconstruct each word according to the results of classification and segmentation.
31
Chutke, Sravanthi, Nandhitha N.M. e Praveen Kumar Lendale. "Video compression based on zig-zag 3D DCT and run-length encoding for multimedia communication systems". International Journal of Pervasive Computing and Communications, 25 de julho de 2022. http://dx.doi.org/10.1108/ijpcc-01-2022-0012.
Texto completo da fonteResumo:
Purpose With the advent of technology, a huge amount of data is being transmitted and received through the internet. Large bandwidth and storage are required for the exchange of data and storage, respectively. Hence, compression of the data which is to be transmitted over the channel is unavoidable. The main purpose of the proposed system is to use the bandwidth effectively. The videos are compressed at the transmitter’s end and reconstructed at the receiver’s end. Compression techniques even help for smaller storage requirements. Design/methodology/approach The paper proposes a novel compression technique for three-dimensional (3D) videos using a zig-zag 3D discrete cosine transform. The method operates a 3D discrete cosine transform on the videos, followed by a zig-zag scanning process. Finally, to convert the data into a single bit stream for transmission, a run-length encoding technique is used. The videos are reconstructed by using the inverse 3D discrete cosine transform, inverse zig-zag scanning (quantization) and inverse run length coding techniques. The proposed method is simple and reduces the complexity of the convolutional techniques. Findings Coding reduction, code word reduction, peak signal to noise ratio (PSNR), mean square error, compression percent and compression ratio values are calculated, and the dominance of the proposed method over the convolutional methods is seen. Originality/value With zig-zag quantization and run length encoding using 3D discrete cosine transform for 3D video compression, gives compression up to 90% with a PSNR of 41.98 dB. The proposed method can be used in multimedia applications where bandwidth, storage and data expenses are the major issues.
32
Deng, Lei, Youzhi Liu, Yechuan Shi, Wenhao Zhang, Chun Yang e Hui Liu. "Deep neural networks for inferring binding sites of RNA-binding proteins by using distributed representations of RNA primary sequence and secondary structure". BMC Genomics 21, S13 (dezembro de 2020). http://dx.doi.org/10.1186/s12864-020-07239-w.
Texto completo da fonteResumo:
Abstract Background RNA binding proteins (RBPs) play a vital role in post-transcriptional processes in all eukaryotes, such as splicing regulation, mRNA transport, and modulation of mRNA translation and decay. The identification of RBP binding sites is a crucial step in understanding the biological mechanism of post-transcriptional gene regulation. However, the determination of RBP binding sites on a large scale is a challenging task due to high cost of biochemical assays. Quite a number of studies have exploited machine learning methods to predict binding sites. Especially, deep learning is increasingly used in the bioinformatics field by virtue of its ability to learn generalized representations from DNA and protein sequences. Results In this paper, we implemented a novel deep neural network model, DeepRKE, which combines primary RNA sequence and secondary structure information to effectively predict RBP binding sites. Specifically, we used word embedding algorithm to extract features of RNA sequences and secondary structures, i.e., distributed representation of k-mers sequence rather than traditional one-hot encoding. The distributed representations are taken as input of convolutional neural networks (CNN) and bidirectional long-term short-term memory networks (BiLSTM) to identify RBP binding sites. Our results show that deepRKE outperforms existing counterpart methods on two large-scale benchmark datasets. Conclusions Our extensive experimental results show that DeepRKE is an efficacious tool for predicting RBP binding sites. The distributed representations of RNA sequences and secondary structures can effectively detect the latent relationship and similarity between k-mers, and thus improve the predictive performance. The source code of DeepRKE is available at https://github.com/youzhiliu/DeepRKE/.
33
Luo, Hanze, Yingkui Gong, Si Chen, Cheng Yu, Guang Yang, Fengzheng Yu, Ziyue Hu e Xiangwei Tian. "Prediction of Global Ionospheric Total Electron Content (TEC) Based on SAM‐ConvLSTM Model". Space Weather 21, n.º 12 (dezembro de 2023). http://dx.doi.org/10.1029/2023sw003707.
Texto completo da fonteResumo:
AbstractThis paper first applies a prediction model based on self‐attention memory ConvLSTM (SAM‐ConvLSTM) to predict the global ionospheric total electron content (TEC) maps with up to 1 day of lead time. We choose the global ionospheric TEC maps released by the Center for Orbit Determination in Europe (CODE) as the training data set covering the period from 1999 to 2022. Besides that, we put several space environment data as additional multivariate‐features into the framework of the prediction model to enhance its forecasting ability. In order to confirm the efficiency of the proposed model, the other two prediction models based on convolutional long short‐term memory (LSTM) are used for comparison. The three models are trained and evaluated on the same data set. Results show that the proposed SAM‐ConvLSTM prediction model performs more accurately than the other two models, and more stably under space weather events. In order to assess the generalization capabilities of the proposed model amidst severe space weather occurrences, we selected the period of 22–25 April 2023, characterized by a potent geomagnetic storm, for experimental validation. Subsequently, we employed the 1‐day predicted global TEC products from the Center for Operational Products and Services (COPG) and the SAM‐ConvLSTM model to evaluate their respective forecasting prowess. The results show that the SAM‐ConvLSTM prediction model achieves lower prediction error. In one word, the ionospheric TEC prediction model proposed in this paper can establish the ionosphere TEC of spatio‐temporal data association for a long time, and realize high precision of prediction performance.
34
Brekke, P., I. Pilan, H. Husby, T. Gundersen, F. A. Dahl, P. Hurlen, O. E. Nytroe e L. Ovrelid. "Automated identification of patients with syncope in the textual health record – a feasibility study using machine learning and natural language processing". European Heart Journal 41, Supplement_2 (1 de novembro de 2020). http://dx.doi.org/10.1093/ehjci/ehaa946.0723.
Texto completo da fonteResumo:
Abstract Background Syncope is a commonly occurring presenting symptom in emergency departments. While the majority of episodes are benign, syncope is associated with worse prognosis in hypertrophic cardiomyopathy, arrhythmia syndromes, heart failure, aortic stenosis and coronary heart disease. Flagging documented syncope in these patients may be crucial to management decisions. Previous studies show that the International Classification of Diseases (ICD) codes for syncope have a sensitivity of around 0.63, leading to a large number of false negatives if patient identification is based on administrative codes. Thus, in order to provide data-driven, clinical decision support, and to improve identification of patient cohorts for research, better tools are needed. A recent study manually annotated more than 30.000 patient records in order to develop a natural language processing (NLP) tool, which achieved a sensitivity of 92.2%. Since access to medical records and annotation resources is limited, we aimed to investigate whether an unsupervised machine learning and NLP approach with no manual input could achieve similar performance. Methods Our data was admission notes for adult patients admitted between 2005 and 2016 at a large university hospital in Norway. 500 records from patients with, and 500 without a “R55 Syncope” ICD code at discharge were drawn at random. R55 code was considered “ground truth”. Headers containing information about tentative diagnoses were removed from the notes, when present, using regular expressions. The dataset was divided into 70%/15%/15% subsets for training, validation and testing. Baseline identification was calculated by a simple lexical matching using the term “synkope”. We evaluated two linear classifiers, a Support Vector Machine (SVM) and a Linear Regression (LR) model, with a term frequency–inverse document frequency vectorizer, using a bag-of-words approach. In addition, we evaluated a simple convolutional neural network (CNN) consisting of a convolutional layer concatenating filter sizes of 3–5, max pooling and a dropout of 0.5 with randomly initialised word embeddings of 300 dimensions. Results Even a baseline regular expression model achieved a sensitivity of 78% and a specificity of 91% when classifying admission notes as belonging to the syncope class or not. The SVM model and the LR model achieved a sensitivity of 91% and 89%, respectively, and a specificity of 89% and 91%. The CNN model had a sensitivity of 95% and a specificity of 84%. Conclusion With a limited non-English dataset, common NLP and machine learning approaches were able to achieve approximately 90–95% sensitivity for the identification of admission notes related to syncope. Linear classifiers outperformed a CNN model in terms of specificity, as expected in this small dataset. The study demonstrates the feasibility of training document classifiers based on diagnostic codes in order to detect important clinical events. ROC curves for SVM and LR models Funding Acknowledgement Type of funding source: Public grant(s) – National budget only. Main funding source(s): The Research Council of Norway