Search

Language models such as Bidirectional Encoder Representations from Transformers (BERT) have been very effective in various Natural Language Processing (NLP) and text mining tasks including text classification. However, some tasks still pose challenges for these models, including text classification with limited labels. This can result in a cold-start problem. Although some approaches have attempted to address this problem through single-stage clustering as an intermediate training step coupled with a pre-Trained language model, which generates pseudo-labels to improve classification, these methods are often error-prone due to the limitations of the clustering algorithms. To overcome this, we have developed a novel two-stage intermediate clustering with subsequent fine-Tuning that models the pseudo-labels reliably, resulting in reduced prediction errors. The key novelty in our model, IDoFew, is that the two-stage clustering coupled with two different clustering algorithms helps exploit the advantages of the complementary algorithms that reduce the errors in generating reliable pseudo-labels for fine-Tuning. Our approach has shown significant improvements compared to strong comparative models.

A fascinating challenge in robotics-human interaction is imitating the emotion recognition capability of humans to robots with the aim to make human-robotics interaction natural, genuine and intuitive. To achieve the natural interaction in affective robots, human-machine interfaces, and autonomous vehicles, understanding our attitudes and opinions is very important, and it provides a practical and feasible path to realize the connection between machine and human. Multimodal interface that includes voice along with facial expression can manifest a large range of nuanced emotions compared to purely textual interfaces and provide a great value to improve the intelligence level of effective communication. Interfaces that fail to manifest or ignore user emotions may significantly impact the performance and risk being perceived as cold, socially inept, untrustworthy, and incompetent. To equip a child well for life, we need to help our children identify their feelings, manage them well, and express their needs in healthy, respectful, and direct ways. Early identification of emotional deficits can help to prevent low social functioning in children. In this work, we analyzed the child's spontaneous behavior using multimodal facial expression and voice signal presenting multimodal transformer-based last feature fusion for facial behavior analysis in children to extract contextualized representations from RGB video sequence and Hematoxylin and eosin video sequence and then using these representations followed by pairwise concatenations of contextualized representations using cross-feature fusion technique to predict users emotions. To validate the performance of the proposed framework, we have performed experiments with the different pairwise concatenations of contextualized repre

Skin cancer, primarily resulting from the abnormal growth of skin cells, is among the most common cancer types. In recent decades, the incidence of skin cancer cases worldwide has risen significantly (one in every three newly diagnosed cancer cases is a skin cancer). Such an increase can be attributed to changes in our social and lifestyle habits coupled with devastating man-made alterations to the global ecosystem. Despite such a notable increase, diagnosis of skin cancer is still challenging, which becomes critical as its early detection is crucial for increasing the overall survival rate. This calls for advancements of innovative computer-aided systems to assist medical experts with their decision making. In this context, there has been a recent surge of interest in machine learning (ML), in particular, deep neural networks (DNNs), to provide complementary assistance to expert physicians. While DNNs have a high processing capacity far beyond that of human experts, their outputs are deterministic, i.e., providing estimates without prediction confidence. Therefore, it is of paramount importance to develop DNNs with uncertainty-awareness to provide confidence in their predictions. Monte Carlo dropout (MCD) is vastly used for uncertainty quantification; however, MCD suffers from overconfidence and being miss calibrated. In this paper, we use MCD algorithm to develop an uncertainty-aware DNN that assigns high predictive entropy to erroneous predictions and enable the model to optimize the hyper-parameters during training, which leads to more accurate uncertainty quantification. We use two synthetic (two moons and blobs) and a real dataset (skin cancer) to validate our algorithm. Our experiments on these datasets prove effectiveness of our approach in quantifying reliable uncertainty. Our method achieved 85.65 ± 0.18 prediction accuracy, 83.03 ± 0.25 uncertainty accuracy, and 1.93 ± 0.3 expected calibration error outperforming vanilla MCD and MCD with loss enhanced bas

People across the globe have felt and are still going through the impact of COVID-19. Some of them share their feelings and suffering online via different online social media networks such as Twitter. Due to strict restrictions to reduce the spread of the novel virus, many people are forced to stay at home, which significantly impacts people's mental health. It is mainly because the pandemic has directly affected the lives of the people who were not allowed to leave home due to strict government restrictions. Researchers must mine the related human-generated data and get insights from it to influence government policies and address people's needs. In this paper, we study social media data to understand how COVID-19 has impacted people's depression. We share a large-scale COVID-19 dataset that can be used to analyze depression. We also have modeled the tweets of depressed and non-depressed users before and after the start of the COVID-19 pandemic. To this end, we developed a new approach based on Hierarchical Convolutional Neural Network (HCN) that extracts fine-grained and relevant content on user historical posts. HCN considers the hierarchical structure of user tweets and contains an attention mechanism that can locate the crucial words and tweets in a user document while also considering the context. Our new approach is capable of detecting depressed users occurring within the COVID-19 time frame. Our results on benchmark datasets show that many non-depressed people became depressed during the COVID-19 pandemic.

Since the last years and until now technology has made fast progress for many industries in particularly garment industry which aims to follow consumer desires and demands One of these demands is to fit clothes before purchasing them on line Therefore many research works have been focused on how to develop an intelligent apparel industry to ensure the online shopping experience Image based virtual try on is among the most potential approach of virtual fitting that tries on target clothes into customer s image therefore it has received considerable research efforts in the recent years However there are several challenges involved in development of virtual try on that make it difficult to achieve naturally looking virtual outfit such as shape pose occlusion illumination cloth texture logo and text etc The aim of this study is to provide a comprehensive and structured overview of extensive research on the advancement of virtual try on This review first introduces virtual try on and its challenges followed by its demand in fashion industry We summarize state of the art image based virtual try on for both fashion detection and fashion synthesis as well as their respective advantages drawbacks and guidelines for selection of specific try on model followed by its recent development and successful application Finally we conclude the paper with promising directions for future research 2022 The Author s under exclusive licence to Springer Science Business Media LLC part of Springer Nature

Federated learning has recently emerged as a striking framework for allowing machine and deep learning models with thousands of participants to have distributed training to preserve the privacy of users' data. Federated learning comes with the pros of allowing all participants the possibility of creating robust models even in the absence of sufficient training data. Meanwhile, the participants are allowed to stay anonymous in the process. Recently, Smartphone usage has increased on a huge scale due to its portability and ability to perform many daily life tasks. Typing on a smartphone's soft keyboard generates vibrations that could be abused to distinguish the typed keys, aiding side-channel attacks. This data can be in the form of clinical notes, medical information, username, and passwords. The attackers can steal this data using smartphone hardware sensors. This study proposes a novel framework based on federated learning for side-channel attack detection to secure this information. We collected a dataset from 10 Android smartphone users who were asked to type on the smartphone soft keyboard. We convert this dataset into two windows of five users to make two clients train local models. The federated learning-based framework aggregates model updates contributed by two clients and trains the DNN model individually on the dataset. To reduce the over-fitting factor, each client examines the findings three times. Experiments reveal that the DNN model has a higher accuracy of 80.09\%, showing that the proposed framework can efficiently detect side-channel attacks.

The ability to explain why the model produced results in such a way is an important problem, especially in the medical domain. Model explainability is important for building trust by providing insight into the model prediction. However, most existing machine learning methods provide no explainability, which is worrying. For instance, in the task of automatic depression prediction, most machine learning models lead to predictions that are obscure to humans. In this work, we propose explainable Multi-Aspect Depression Detection with Hierarchical Attention Network MDHAN, for automatic detection of depressed users on social media and explain the model prediction. We have considered user posts augmented with additional features from Twitter. Specifically, we encode user posts using two levels of attention mechanisms applied at the tweet-level and word-level, calculate each tweet and words' importance, and capture semantic sequence features from the user timelines (posts). Our hierarchical attention model is developed in such a way that it can capture patterns that leads to explainable results. Our experiments show that MDHAN outperforms several popular and robust baseline methods, demonstrating the effectiveness of combining deep learning with multi-aspect features. We also show that our model helps improve predictive performance when detecting depression in users who are posting messages publicly on social media. MDHAN achieves excellent performance and ensures adequate evidence to explain the prediction.

Word representation has always been an important research area in the history of natural language processing (NLP). Understanding such complex text data is imperative, given that it is rich in information and can be used widely across various applications. In this survey, we explore different word representation models and its power of expression, from the classical to modern-day state-of-the-art word representation language models (LMS). We describe a variety of text representation methods, and model designs have blossomed in the context of NLP, including SOTA LMs. These models can transform large volumes of text into effective vector representations capturing the same semantic information. Further, such representations can be utilized by various machine learning (ML) algorithms for a variety of NLP-related tasks. In the end, this survey briefly discusses the commonly used ML- and DL-based classifiers, evaluation metrics, and the applications of these word embeddings in different NLP tasks.

In recent times, we can see a massive increase in the number of devices that are being connected to the internet. These devices include but are not limited to smartphones, IoT, and cloud networks. In comparison to other possible cyber-attacks, these days, hackers are targeting these devices with phishing attacks since it exploits human vulnerabilities rather than system vulnerabilities. In a phishing attack, an online user is deceived by a seemingly trusted entity to give their personal data, i.e., login credentials or credit card details. When this private information is leaked to the hackers, this information becomes the source of other sophisticated attacks. In recent times many researchers have proposed the machine learning-based approach to solve phishing attacks; however, they have used a large number of features to develop reliable phishing detection techniques. A large number of features requires large processing powers to detect phishing, which makes it very much unsuitable for resource constrained devices. To address this issue, we have developed a phishing detection approach that only needs nine lexical features for effectively detecting phishing attacks. We used ISCXURL-2016 dataset for our experimental purpose, where 11964 instances of legitimate and phishing URLs are used. We have tested our approach against different machine learning classifiers and have obtained the highest accuracy of 99.57% with the Random forest algorithm.

Understanding customer behaviour (UCB) sequences with the multi-dimensional and temporal of the data is necessary for any competitive and global business aiming to provide interesting insights and to improve business strategies. While existing researchers have applied various data analytics approaches to understand and analyze behaviors of customer, they often failed to allow the analysts including business management, product marketing and development, and decision making, etc. Achieving these goals in collaboration with domain experts, we conducted a design study contributes to address a known problem with a novel solution and to provide data-driven visual decision support in collective policy data. We determine core study demands and then use a Visual Interactive System for Understanding Customer Behavior, named UCBVis that enables decision makers to gain detail insights into customer activities. In this study, we present customer behaviour pattern of multidimensional relationship through the visualisation system based on interweaving the pattern mining and querying with a designed encoding scheme. We use a large number of customer claim records and present visual outcomes to facilitate the exploration of customer behavior. Furthermore, we provide a concise set of insights and challenges associated with the use of UCBVis in the life insurance industry. We show the robustness of UCBVis through a user study with five participants shows that UCBVis is perceived to be more useful and provides actionable insights.

© 2020 Elsevier B.V. Exponential growth of large scale data industrial internet of things is evident due to the enormous deployment of IoT data acquisition devices. Detection of unusual patterns from large scale IoT data is important though challenging task. Recently, one-class support vector machines is extensively being used for anomaly detection. It tries to find an optimal hyperplane in high dimensional data that best separates the data from anomalies with maximum margin. However, the hinge loss of traditional one-class support vector machines is unbounded, which results in larger loss caused by outliers affecting its performance for anomaly detection. Furthermore, existing methods are computationally complex for larger data. In this paper, we present novel anomaly detection for large scale data by using randomized nonlinear features in support vector machines with bounded loss function rather than finding optimized support vectors with unbounded loss function. Extensive experimental evaluation on ten benchmark datasets shows the robustness of the proposed approach against outliers such as 0.8239, 0.7921, 0.7501, 0.6711, 0.6692, 0.4789, 0.6462, 0.6812, 0.7271 and 0.7873 accuracy for Gas Sensor Array, Human Activity Recognition, Parkinson's, Hepatitis, Breast Cancer, Blood Transfusion, Heart, ILPD and Wholesale Customers datasets respectively. In addition to this, introduction of randomized nonlinear feature helps to considerably decrease the computational complexity and space complexity from O(N3) to O(Bkn) and O(N2) to O(Bkn). Thus, very attractive for larger datasets.

© 2019 Elsevier Ltd Since the principal component analysis and its variants are sensitive to outliers that affect their performance and applicability in real world, several variants have been proposed to improve the robustness. However, most of the existing methods are still sensitive to outliers and are unable to select useful features. To overcome the issue of sensitivity of PCA against outliers, in this paper, we introduce two-dimensional outliers-robust principal component analysis (ORPCA) by imposing the joint constraints on the objective function. ORPCA relaxes the orthogonal constraints and penalizes the regression coefficient, thus, it selects important features and ignores the same features that exist in other principal components. It is commonly known that square Frobenius norm is sensitive to outliers. To overcome this issue, we have devised an alternative way to derive objective function. Experimental results on four publicly available benchmark datasets show the effectiveness of joint feature selection and provide better performance as compared to state-of-the-art dimensionality-reduction methods.

© 2020 IEEE. Right from the very beginning, the text has vital importance in human life. As compared to the vision-based applications, preference is always given to the precise and productive information embodied in the text. Considering the importance of text, recognition, and detection of text is also equally important in human life. This paper presents a deep analysis of recent development on scene text and compare their performance and bring into light the real modern applications. Future potential directions of scene text detection and recognition are also discussed.

© 2020 IEEE. Text in natural scene image portrays rich semantic information that plays an important role in content analysis. However, apart from Arabic text in documents, the text in natural scene images exhibit much higher diversity and variability, especially in uncontrolled circumstances. In this paper, a hybrid feature extraction approach is presented to detect extremal region of Arabic scene text. The binary image and image mask are considered as a variant of input image and look for concurrent extremal regions in both images. After determination of conjoined extremal points, the scale invariant technique is applied to consider those invariant points which are common in both images based on their coordinate positions. To evaluate the performance, a multidimensional long short term memory (LSTM) network is adapted and obtained 94.21% accuracy for word recognition on unconstrained Arabic scene text recognition (ASTR) dataset.

© 2020 Elsevier B.V. Along with the emergence of the Internet, the rapid development of handheld devices has democratized content creation due to the extensive use of social media and has resulted in an explosion of short informal texts. Although a sentiment analysis of these texts is valuable for many reasons, this task is often perceived as a challenge given that these texts are often short, informal, noisy, and rich in language ambiguities, such as polysemy. Moreover, most of the existing sentiment analysis methods are based on clean data. In this paper, we present DICET, a transformer-based method for sentiment analysis that encodes representation from a transformer and applies deep intelligent contextual embedding to enhance the quality of tweets by removing noise while taking word sentiments, polysemy, syntax, and semantic knowledge into account. We also use the bidirectional long- and short-term memory network to determine the sentiment of a tweet. To validate the performance of the proposed framework, we perform extensive experiments on three benchmark datasets, and results show that DICET considerably outperforms the state of the art in sentiment classification.

© 2020 IEEE. Humans use tonal stress and gestural cues to reveal negative feelings that are expressed ironically using positive or intensified positive words when communicating vocally. However, in textual data, like posts on social media, cues on sentiment valence are absent, thus making it challenging to identify the true meaning of utterances, even for the human reader. For a given post, an intelligent natural language processing system should be able to identify whether a post is ironic/sarcastic or not. Recent work confirms the difficulty of detecting sarcastic/ironic posts. To overcome challenges involved in the identification of sentiment valence, this paper presents the identification of irony and sarcasm in social media posts through transformer-based deep, intelligent contextual embedding - T-DICE - which improves noise within contexts. It solves the language ambiguities such as polysemy, semantics, syntax, and words sentiments by integrating embeddings. T-DICE is then forwarded to attention-based Bidirectional Long Short Term Memory (BiLSTM) to find out the sentiment of a post. We report the classification performance of the proposed network on benchmark datasets for #irony #sarcasm. Results demonstrate that our approach outperforms existing state-of-the-art methods.

© 2001-2011 IEEE. Accurate classification of Electroencephalogram (EEG) signals plays an important role in diagnoses of different type of mental activities. One of the most important challenges, associated with classification of EEG signals is how to design an efficient classifier consisting of strong generalization capability. Aiming to improve the classification performance, in this paper, we propose a novel multiclass support matrix machine (M-SMM) from the perspective of maximizing the inter-class margins. The objective function is a combination of binary hinge loss that works on C matrices and spectral elastic net penalty as regularization term. This regularization term is a combination of Frobenius and nuclear norm, which promotes structural sparsity and shares similar sparsity patterns across multiple predictors. It also maximizes the inter-class margin that helps to deal with complex high dimensional noisy data. The extensive experiment results supported by theoretical analysis and statistical tests show the effectiveness of the M-SMM for solving the problem of classifying EEG signals associated with motor imagery in brain-computer interface applications.

© 2013 IEEE. Background: EEG signals are extremely complex in comparison to other biomedical signals, thus require an efficient feature selection as well as classification approach. Traditional feature extraction and classification methods require to reshape the data into vectors that results in losing the structural information exist in the original featured matrix. Aim: The aim of this work is to design an efficient approach for robust feature extraction and classification for the classification of EEG signals. Method: In order to extract robust feature matrix and reduce the dimensionality of from original epileptic EEG data, in this paper, we have applied robust joint sparse PCA (RJSPCA), Outliers Robust PCA (ORPCA) and compare their performance with different matrix base feature extraction methods, followed by classification through support matrix machine. The combination of joint sparse PCA with robust support matrix machine showed good generalization performance for classification of EEG data due to their convex optimization. Results: A comprehensive experimental study on the publicly available EEG datasets is carried out to validate the robustness of the proposed approach against outliers. Conclusion: The experiment results, supported by the theoretical analysis and statistical test, show the effectiveness of the proposed framework for solving classification of EEG signals.

Recent works attempt to extend Graph Convolution Networks (GCNs) to point clouds for classification and segmentation tasks. These works tend to sample and group points to create smaller point sets locally and mainly focus on extracting local features through GCNs, while ignoring the relationship between point sets. In this paper, we propose the Dynamic Hop Graph Convolution Network (DHGCN) for explicitly learning the contextual relationships between the voxelized point parts, which are treated as graph nodes. Motivated by the intuition that the contextual information between point parts lies in the pairwise adjacent relationship, which can be depicted by the hop distance of the graph quantitatively, we devise a novel self-supervised part-level hop distance reconstruction task and design a novel loss function accordingly to facilitate training. In addition, we propose the Hop Graph Attention (HGA), which takes the learned hop distance as input for producing attention weights to allow edge features to contribute distinctively in aggregation. Eventually, the proposed DHGCN is a plug-and-play module that is compatible with point-based backbone networks. Comprehensive experiments on different backbones and tasks demonstrate that our self-supervised method achieves state-of-the-art performance. Our source code is available at: https://github.com/Jinec98/DHGCN.

This paper proposes a novel approach for selecting a subset of features in semi-supervised datasets where only some of the patterns are labeled. The whole process is completed in two phases. In the first phase, i.e., Phase-I, the whole dataset is divided into two parts: The first part, which contains labeled patterns, and the second part, which contains unlabeled patterns. In the first part, a small number of features are identified using well-known maximum relevance (from first part) and minimum redundancy (whole dataset) based feature selection approaches using the correlation coefficient. The subset of features from the identified set of features, which produces a high classification accuracy using any supervised classifier from labeled patterns, is selected for later processing. In the second phase, i.e., Phase-II, the patterns belonging to the first and second part are clustered separately into the available number of classes of the dataset. In the clusters of the first part, take the majority of patterns belonging to a cluster as the class for that cluster, which is given already. Form the pairs of cluster centroids made in the first and second part. The centroid of the second part nearest to a centroid of the first part will be paired. As the class of the first centroid is known, the same class can be assigned to the centroid of the cluster of the second part, which is unknown. The actual class of the patterns if known for the second part of the dataset can be used to test the classification accuracy of patterns in the second part. The proposed two-phase approach performs well in terms of classification accuracy and number of features selected on the given benchmarked datasets.

This paper proposes a novel approach for selecting a subset of features in semi-supervised datasets where only some of the patterns are labeled. The whole process is completed in two phases. In the first phase, i.e., Phase-I, the whole dataset is divided into two parts: The first part, which contains labeled patterns, and the second part, which contains unlabeled patterns. In the first part, a small number of features are identified using well-known maximum relevance (from first part) and minimum redundancy (whole dataset) based feature selection approaches using the correlation coefficient. The subset of features from the identified set of features, which produces a high classification accuracy using any supervised classifier from labeled patterns, is selected for later processing. In the second phase, i.e., Phase-II, the patterns belonging to the first and second part are clustered separately into the available number of classes of the dataset. In the clusters of the first part, take the majority of patterns belonging to a cluster as the class for that cluster, which is given already. Form the pairs of cluster centroids made in the first and second part. The centroid of the second part nearest to a centroid of the first part will be paired. As the class of the first centroid is known, the same class can be assigned to the centroid of the cluster of the second part, which is unknown. The actual class of the patterns if known for the second part of the dataset can be used to test the classification accuracy of patterns in the second part. The proposed two-phase approach performs well in terms of classification accuracy and number of features selected on the given benchmarked datasets.