Data Mining - Concepts and Techniques.

Objective Electroencephalography (EEG) analysis has been an important tool in neuroscience with applications in neuroscience, neural engineering (e.g. Brain-computer interfaces, BCI's), and even commercial applications. Many of the analytical tools used in EEG studies have used machine learning to uncover relevant information for neural classification and neuroimaging. Recently, the availability of large EEG data sets and advances in machine learning have both led to the deployment of deep learning architectures, especially in the analysis of EEG signals and in understanding the information it may contain for brain functionality. The robust automatic classification of these signals is an important step towards making the use of EEG more practical in many applications and less reliant on trained professionals. Towards this goal, a systematic review of the literature on deep learning applications to EEG classification was performed to address the following critical questions: (1) Which EEG classification tasks have been explored with deep learning? (2) What input formulations have been used for training the deep networks? (3) Are there specific deep learning network structures suitable for specific types of tasks? Approach A systematic literature review of EEG classification using deep learning was performed on Web of Science and PubMed databases, resulting in 90 identified studies. Those studies were analyzed based on type of task, EEG preprocessing methods, input type, and deep learning architecture. Main results For EEG classification tasks, convolutional neural networks, recurrent neural networks, deep belief networks outperform stacked auto-encoders and multi-layer perceptron neural networks in classification accuracy. The tasks that used deep learning fell into five general groups: emotion recognition, motor imagery, mental workload, seizure detection, event related potential detection, and sleep scoring. For each type of task, we describe the specific input formulation, major characteristics, and end classifier recommendations found through this review. Significance This review summarizes the current practices and performance outcomes in the use of deep learning for EEG classification. Practical suggestions on the selection of many hyperparameters are provided in the hope that they will promote or guide the deployment of deep learning to EEG datasets in future research.

https://iopscience.iop.org/article/10.1088/1741-2552/ab0ab5/pdf

Deep learning for electroencephalogram (EEG) classification tasks: a review.

Deep learning in spiking neural networks

This paper discusses the fourth year of the ”Sentiment Analysis in Twitter Task”. SemEval-2016 Task 4 comprises five subtasks, three of which represent a significant departure from previous editions. The first two subtasks are reruns from prior years and ask to predict the overall sentiment, and the sentiment towards a topic in a tweet. The three new subtasks focus on two variants of the basic “sentiment classification in Twitter” task. The first variant adopts a five-point scale, which confers an ordinal character to the classification task. The second variant focuses on the correct estimation of the prevalence of each class of interest, a task which has been called quantification in the supervised learning literature. The task continues to be very popular, attracting a total of 43 teams.

SemEval-2016 Task 4: Sentiment Analysis in Twitter

Context Electroencephalography (EEG) is a complex signal and can require several years of training, as well as advanced signal processing and feature extraction methodologies to be correctly interpreted. Recently, deep learning (DL) has shown great promise in helping make sense of EEG signals due to its capacity to learn good feature representations from raw data. Whether DL truly presents advantages as compared to more traditional EEG processing approaches, however, remains an open question. Objective In this work, we review 154 papers that apply DL to EEG, published between January 2010 and July 2018, and spanning different application domains such as epilepsy, sleep, brain-computer interfacing, and cognitive and affective monitoring. We extract trends and highlight interesting approaches from this large body of literature in order to inform future research and formulate recommendations. Methods Major databases spanning the fields of science and engineering were queried to identify relevant studies published in scientific journals, conferences, and electronic preprint repositories. Various data items were extracted for each study pertaining to (1) the data, (2) the preprocessing methodology, (3) the DL design choices, (4) the results, and (5) the reproducibility of the experiments. These items were then analyzed one by one to uncover trends. Results Our analysis reveals that the amount of EEG data used across studies varies from less than ten minutes to thousands of hours, while the number of samples seen during training by a network varies from a few dozens to several millions, depending on how epochs are extracted. Interestingly, we saw that more than half the studies used publicly available data and that there has also been a clear shift from intra-subject to inter-subject approaches over the last few years. About [Formula: see text] of the studies used convolutional neural networks (CNNs), while [Formula: see text] used recurrent neural networks (RNNs), most often with a total of 3-10 layers. Moreover, almost one-half of the studies trained their models on raw or preprocessed EEG time series. Finally, the median gain in accuracy of DL approaches over traditional baselines was [Formula: see text] across all relevant studies. More importantly, however, we noticed studies often suffer from poor reproducibility: a majority of papers would be hard or impossible to reproduce given the unavailability of their data and code. Significance To help the community progress and share work more effectively, we provide a list of recommendations for future studies and emphasize the need for more reproducible research. We also make our summary table of DL and EEG papers available and invite authors of published work to contribute to it directly. A planned follow-up to this work will be an online public benchmarking portal listing reproducible results.

https://iopscience.iop.org/article/10.1088/1741-2552/ab260c/pdf

Deep learning-based electroencephalography analysis: a systematic review.

Nowadays, machine learning (ML) is one of the significant technology in the new era. It almost applies to all the fields in our lives. ML has a notable impact on medical care when dealing with medical images. Medical images are used to investigate the internal-parts of the body of human. The type and size of medical images vary from one scan to another. Besides, medical images are not like natural images. While natural images can perform object detection and classification easily, even if the images are re-sized to smaller images. Resizing medical images is not an efficient method because they interact with pixel-level. In this chapter, medical imaging modalities and histopathology are explained. Furthermore, the best medical image type and size for classification and detection of medical diagnoses are explained. Moreover, specific methods are considered in medical images such as image compression, image format, image resize, and other essential aspects. Finally, we also give a brief summary of deep learning algorithms that are used with medical images.

Machine Learning Cancer Diagnosis Based on Medical Image Size and Modalities

In Cross Language Information Retrieval (CLIR), queries in one language retrieve documents in other language(s). This can be done through Query Translation that comes up against Translation/Transliteration challenges like ambiguity as the main problems. In this paper, a comprehensive solution has been introduced for these challenges. 1 st , 4 powerful English/Arabic Machine Readable Dictionaries (MRD) from English to Arabic, including a dictionary for collocations and 3 dictionaries for single English words that have been introduced from different perspectives, aiming to examine the effect of each perspective on the final query result. 2 nd , A modern Arabic Corpus has been built. 3 rd , a comprehensive model for Query Translation from English to Arabic that detects and translates collocations, single words translation and transliteration, and solves the replacement ambiguity, has been introduced. The experiments' results proved that the proposed models are very effective overcoming the Query Translation and CLIR challenges.

Enhancing English/Arabic CLIR Using Word Collocations and Statistical Translation and Transliteration Resources

A reasonable amount of annotated data is required for fine-tuning pre-trained language models (PLM) on down-stream tasks. However, obtaining labeled examples for different language varieties can be costly. In this paper, we investigate the zero-shot performance on Dialectal Arabic (DA) when fine-tuning a PLM on modern standard Arabic (MSA) data only— identifying a significant performance drop when evaluating such models on DA. To remedy such performance drop, we propose self-training with unlabeled DA data and apply it in the context of named entity recognition (NER), part-of-speech (POS) tagging, and sarcasm detection (SRD) on several DA varieties. Our results demonstrate the effectiveness of self-training with unlabeled DA data: improving zero-shot MSA-to-DA transfer by as large as ~10% F₁ (NER), 2% accuracy (POS tagging), and 4.5% F₁ (SRD). We conduct an ablation experiment and show that the performance boost observed directly results from the unlabeled DA examples used for self-training. Our work opens up opportunities for leveraging the relatively abundant labeled MSA datasets to develop DA models for zero and low-resource dialects. We also report new state-of-the-art performance on all three tasks and open-source our fine-tuned models for the research community.

/pdf/zero-resource-multi-dialectal-arabic-natural-language-51mkvgfp6j.pdf

Zero-resource Multi-dialectal Arabic Natural Language Understanding

Recently, heart sound signals have been used in the detection of the heart valve status and the identification of the heart valve disease. Due to these characteristics, therefore, two feature reduction techniques have been proposed prior applying data classifications in this paper. The first technique is the chi-Square which measures the lack of independence between each heart sound feature and the target class, while the second technique is the deep believe network that uses to generate a new data set of a reduced number of features according the partition of the heart signals. The importance of feature reduction prior applying data classification is not only to improve the classification accuracy and to enhance the training and testing performance, but also it is important to detect which of the stages of heart sound is important for the detection of sick people among normal set of people, and which period important for the classification of heart murmur. Different classification algorithms including naive bayesian tree classifier and sequential minimal optimization was applied on three different data sets of 100 extracted features of the heart sound. The extensive experimental results on the heat sound signals data set demonstrate that the proposed approach outperforms other classifiers and providing the highest classification accuracy with minimized number of features.

Aly A. Fahmy

Papers

Machine Learning Cancer Diagnosis Based on Medical Image Size and Modalities

Enhancing English/Arabic CLIR Using Word Collocations and Statistical Translation and Transliteration Resources

Zero-resource Multi-dialectal Arabic Natural Language Understanding

Heart Sound Feature Reduction Approach for Improving the Heart Valve Diseases Identification

Genetic Adaptive Routing Algorithms (GARA) for Computer Networks