TeKET: a Tree-Based Unsupervised Keyphrase Extraction Technique

Extraction of time-related expressions using text mining with application to Hebrew

In this research, we extract time-related expressions from a rabbinic text in a semi-automatic manner. These expressions usually appear next to rabbinic references (name / nickname / acronym / book-name). The first step toward our goal is to find all the expressions near references in the corpus. However, not all of the phrases around the references are time-related expressions. Therefore, these phrases are initially considered to be potential time-related expressions. To extract the time-related expressions, we formulate two new statistical functions, and we use screening and heuristic methods. We tested these statistical functions, grammatical screenings, and heuristic methods on a corpus containing responsa documents. In this corpus, many rabbinic citations are known and marked. The statistical functions and the screening methods filtered the potential time-related expressions and reduced 99.88% of the initial expressions (from 484,681 to 575).

Mutual-Attention Net: A Deep Attentional Neural Network for Keyphrase Generation

Neural keyphrase generation (NKG) is a recently proposed approach to automatically extract keyphrase from a document. Unlike the traditional keyphrase extraction, the NKG can generate keyphrases that do not appear in the document. However, as a supervised method, NKG is hindered by noise. In order to solve the problem that the existing NKG model does not consider denoising the source document, in this work, this paper introduces a new denoising architecture mutual-attention network (MA-net). Considering the structure of documents in popular datasets, the multihead attention is applied to dig out the relevance between title and abstract, which aids denoising. To further accurate generation of high-quality keyphrases, we use multihead attention to compute the content vector instead of Bahdanau attention. Finally, we employ a hybrid network that augments the proposed architecture to solve OOV (out-of-vocabulary) problem. It can not only generate words from the decoder but also copy words from the source document. Evaluation using five benchmark datasets shows that our model significantly outperforms the state-of-the-art ones currently in the research field.

Enhancing biofeedback-driven self-guided virtual reality exposure therapy through arousal detection from multimodal data using machine learning

Virtual reality exposure therapy (VRET) is a novel intervention technique that allows individuals to experience anxiety-evoking stimuli in a safe environment, recognise specific triggers and gradually increase their exposure to perceived threats. Public-speaking anxiety (PSA) is a prevalent form of social anxiety, characterised by stressful arousal and anxiety generated when presenting to an audience. In self-guided VRET, participants can gradually increase their tolerance to exposure and reduce anxiety-induced arousal and PSA over time. However, creating such a VR environment and determining physiological indices of anxiety-induced arousal or distress is an open challenge. Environment modelling, character creation and animation, psychological state determination and the use of machine learning (ML) models for anxiety or stress detection are equally important, and multi-disciplinary expertise is required. In this work, we have explored a series of ML models with publicly available data sets (using electroencephalogram and heart rate variability) to predict arousal states. If we can detect anxiety-induced arousal, we can trigger calming activities to allow individuals to cope with and overcome distress. Here, we discuss the means of effective selection of ML models and parameters in arousal detection. We propose a pipeline to overcome the model selection problem with different parameter settings in the context of virtual reality exposure therapy. This pipeline can be extended to other domains of interest where arousal detection is crucial. Finally, we have implemented a biofeedback framework for VRET where we successfully provided feedback as a form of heart rate and brain laterality index from our acquired multimodal data for psychological intervention to overcome anxiety.

Four-way classification of Alzheimer's disease using deep Siamese convolutional neural network with triplet-loss function

Alzheimer's disease (AD) is a neurodegenerative disease that causes irreversible damage to several brain regions, including the hippocampus causing impairment in cognition, function, and behaviour. Early diagnosis of the disease will reduce the suffering of the patients and their family members. Towards this aim, in this paper, we propose a Siamese Convolutional Neural Network (SCNN) architecture that employs the triplet-loss function for the representation of input MRI images as k-dimensional embeddings. We used both pre-trained and non-pretrained CNNs to transform images into the embedding space. These embeddings are subsequently used for the 4-way classification of Alzheimer's disease. The model efficacy was tested using the ADNI and OASIS datasets which produced an accuracy of 91.83% and 93.85%, respectively. Furthermore, obtained results are compared with similar methods proposed in the literature.

Unsupervised query reduction for efficient yet effective news background linking

In this article, we study efficient techniques to tackle the news background linking problem, in which an online reader seeks background knowledge about a given article to better understand its context. Recently, this problem attracted many researchers, especially in the Text Retrieval Conference (TREC) community. Surprisingly, the most effective method to date uses the entire input news article as a search query in an ad-hoc retrieval approach to retrieve the background links. In a scenario where the lookup for background links is performed online, this method becomes inefficient, especially if the search scope is big such as the Web, due to the relatively long generated query, which results in a long response time. In this work, we evaluate different unsupervised approaches for reducing the input news article to a much shorter, hence efficient, search query, while maintaining the retrieval effectiveness. We conducted several experiments using the Washington Post dataset, released specifically for the news background linking problem. Our results show that a simple statistical analysis of the article using a recent keyword extraction technique reaches an average of 6.2× speedup in query response time over the full article approach, with no significant difference in effectiveness. Moreover, we found that further reduction of the search terms can be achieved by eliminating relatively low TF-IDF values from the search queries, yielding even more efficient retrieval of 13.3× speedup, while still maintaining the retrieval effectiveness. This makes our approach more suitable for practical online scenarios. Our study is the first to address the efficiency of news background linking systems. We, therefore, release our source code to promote research in that direction.

Channel-independent recreation of artefactual signals in chronically recorded local field potentials using machine learning

Acquisition of neuronal signals involves a wide range of devices with specific electrical properties. Combined with other physiological sources within the body, the signals sensed by the devices are often distorted. Sometimes these distortions are visually identifiable, other times, they overlay with the signal characteristics making them very difficult to detect. To remove these distortions, the recordings are visually inspected and manually processed. However, this manual annotation process is time-consuming and automatic computational methods are needed to identify and remove these artefacts. Most of the existing artefact removal approaches rely on additional information from other recorded channels and fail when global artefacts are present or the affected channels constitute the majority of the recording system. Addressing this issue, this paper reports a novel channel-independent machine learning model to accurately identify and replace the artefactual segments present in the signals. Discarding these artifactual segments by the existing approaches causes discontinuities in the reproduced signals which may introduce errors in subsequent analyses. To avoid this, the proposed method predicts multiple values of the artefactual region using long–short term memory network to recreate the temporal and spectral properties of the recorded signal. The method has been tested on two open-access data sets and incorporated into the open-access SANTIA (SigMate Advanced: a Novel Tool for Identification of Artefacts in Neuronal Signals) toolbox for community use.

Evaluating keyphrase extraction algorithms for finding similar news articles using lexical similarity calculation and semantic relatedness measurement by word embedding

A textual data processing task that involves the automatic extraction of relevant and salient keyphrases from a document that expresses all the important concepts of the document is called keyphrase extraction. Due to technological advancements, the amount of textual information on the Internet is rapidly increasing as a lot of textual information is processed online in various domains such as offices, news portals, or for research purposes. Given the exponential increase of news articles on the Internet, manually searching for similar news articles by reading the entire news content that matches the user's interests has become a time-consuming and tedious task. Therefore, automatically finding similar news articles can be a significant task in text processing. In this context, keyphrase extraction algorithms can extract information from news articles. However, selecting the most appropriate algorithm is also a problem. Therefore, this study analyzes various supervised and unsupervised keyphrase extraction algorithms, namely KEA, KP-Miner, YAKE, MultipartiteRank, TopicRank, and TeKET, which are used to extract keyphrases from news articles. The extracted keyphrases are used to compute lexical and semantic similarity to find similar news articles. The lexical similarity is calculated using the Cosine and Jaccard similarity techniques. In addition, semantic similarity is calculated using a word embedding technique called Word2Vec in combination with the Cosine similarity measure. The experimental results show that the KP-Miner keyphrase extraction algorithm, together with the Cosine similarity calculation using Word2Vec (Cosine-Word2Vec), outperforms the other combinations of keyphrase extraction algorithms and similarity calculation techniques to find similar news articles. The similar articles identified using KPMiner and the Cosine similarity measure with Word2Vec appear to be relevant to a particular news article and thus show satisfactory performance with a Normalized Discounted Cumulative Gain (NDCG) value of 0.97. This study proposes a method for finding similar news articles that can be used in conjunction with other methods already in use.

ACCU3RATE: A mobile health application rating scale based on user reviews

BACKGROUND

Over the last decade, mobile health applications (mHealth App) have evolved exponentially to assess and support our health and well-being.

OBJECTIVE

This paper presents an Artificial Intelligence (AI)-enabled mHealth app rating tool, called ACCU3RATE, which takes multidimensional measures such as user star rating, user review and features declared by the developer to generate the rating of an app. However, currently, there is very little conceptual understanding on how user reviews affect app rating from a multi-dimensional perspective. This study applies AI-based text mining technique to develop more comprehensive understanding of user feedback based on several important factors, determining the mHealth app ratings.

METHOD

Based on the literature, six variables were identified that influence the mHealth app rating scale. These factors are user star rating, user text review, user interface (UI) design, functionality, security and privacy, and clinical approval. Natural Language Toolkit package is used for interpreting text and to identify the App users' sentiment. Additional considerations were accessibility, protection and privacy, UI design for people living with physical disability. Moreover, the details of clinical approval, if exists, were taken from the developer's statement. Finally, we fused all the inputs using fuzzy logic to calculate the new app rating score.

RESULTS AND CONCLUSIONS

ACCU3RATE concentrates on heart related Apps found in the play store and App gallery. The findings indicate the efficacy of the proposed method as opposed to the current device scale. This study has implications for both App developers and consumers who are using mHealth Apps to monitor and track their health. The performance evaluation shows that the proposed mHealth scale has shown excellent reliability as well as internal consistency of the scale, and high inter-rater reliability index. It has also been noticed that the fuzzy based rating scale, as in ACCU3RATE, matches more closely to the rating performed by experts.

SANTIA: a Matlab-based open-source toolbox for artifact detection and removal from extracellular neuronal signals

Neuronal signals generally represent activation of the neuronal networks and give insights into brain functionalities. They are considered as fingerprints of actions and their processing across different structures of the brain. These recordings generate a large volume of data that are susceptible to noise and artifacts. Therefore, the review of these data to ensure high quality by automatically detecting and removing the artifacts is imperative. Toward this aim, this work proposes a custom-developed automatic artifact removal toolbox named, SANTIA (SigMate Advanced: a Novel Tool for Identification of Artifacts in Neuronal Signals). Developed in Matlab, SANTIA is an open-source toolbox that applies neural network-based machine learning techniques to label and train models to detect artifacts from the invasive neuronal signals known as local field potentials.

COVID-19 Infection Detection from Chest X-Ray Images Using Hybrid Social Group Optimization and Support Vector Classifier

A novel strain of Coronavirus, identified as the Severe Acute Respiratory Syndrome-2 (SARS-CoV-2), outbroke in December 2019 causing the novel Corona Virus Disease (COVID-19). Since its emergence, the virus has spread rapidly and has been declared a global pandemic. As of the end of January 2021, there are almost 100 million cases worldwide with over 2 million confirmed deaths. Widespread testing is essential to reduce further spread of the disease, but due to a shortage of testing kits and limited supply, alternative testing methods are being evaluated. Recently researchers have found that chest X-Ray (CXR) images provide salient information about COVID-19. An intelligent system can help the radiologists to detect COVID-19 from these CXR images which can come in handy at remote locations in many developing nations. In this work, we propose a pipeline that uses CXR images to detect COVID-19 infection. The features from the CXR images were extracted and the relevant features were then selected using Hybrid Social Group Optimization algorithm. The selected features were then used to classify the CXR images using a number of classifiers. The proposed pipeline achieves a classification accuracy of 99.65% using support vector classifier, which outperforms other state-of-the-art deep learning algorithms for binary and multi-class classification.

One-shot Cluster-Based Approach for the Detection of COVID-19 from Chest X-ray Images

Coronavirus disease (COVID-19) has infected over more than 28.3 million people around the globe and killed 913K people worldwide as on 11 September 2020. With this pandemic, to combat the spreading of COVID-19, effective testing methodologies and immediate medical treatments are much required. Chest X-rays are the widely available modalities for immediate diagnosis of COVID-19. Hence, automation of detection of COVID-19 from chest X-ray images using machine learning approaches is of greater demand. A model for detecting COVID-19 from chest X-ray images is proposed in this paper. A novel concept of cluster-based one-shot learning is introduced in this work. The introduced concept has an advantage of learning from a few samples against learning from many samples in case of deep leaning architectures. The proposed model is a multi-class classification model as it classifies images of four classes, viz., pneumonia bacterial, pneumonia virus, normal, and COVID-19. The proposed model is based on ensemble of Generalized Regression Neural Network (GRNN) and Probabilistic Neural Network (PNN) classifiers at decision level. The effectiveness of the proposed model has been demonstrated through extensive experimentation on a publicly available dataset consisting of 306 images. The proposed cluster-based one-shot learning has been found to be more effective on GRNN and PNN ensembled model to distinguish COVID-19 images from that of the other three classes. It has also been experimentally observed that the model has a superior performance over contemporary deep learning architectures. The concept of one-shot cluster-based learning is being first of its kind in literature, expected to open up several new dimensions in the field of machine learning which require further researching for various applications.

Deep Learning in Mining Biological Data

Recent technological advancements in data acquisition tools allowed life scientists to acquire multimodal data from different biological application domains. Categorized in three broad types (i.e. images, signals, and sequences), these data are huge in amount and complex in nature. Mining such enormous amount of data for pattern recognition is a big challenge and requires sophisticated data-intensive machine learning techniques. Artificial neural network-based learning systems are well known for their pattern recognition capabilities, and lately their deep architectures-known as deep learning (DL)-have been successfully applied to solve many complex pattern recognition problems. To investigate how DL-especially its different architectures-has contributed and been utilized in the mining of biological data pertaining to those three types, a meta-analysis has been performed and the resulting resources have been critically analysed. Focusing on the use of DL to analyse patterns in data from diverse biological domains, this work investigates different DL architectures' applications to these data. This is followed by an exploration of available open access data sources pertaining to the three data types along with popular open-source DL tools applicable to these data. Also, comparative investigations of these tools from qualitative, quantitative, and benchmarking perspectives are provided. Finally, some open research challenges in using DL to mine biological data are outlined and a number of possible future perspectives are put forward.

Application of deep learning in detecting neurological disorders from magnetic resonance images: a survey on the detection of Alzheimer's disease, Parkinson's disease and schizophrenia

Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders-focusing on Alzheimer's disease, Parkinson's disease and schizophrenia-from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided.