Abstract The integration of different Mobile Edge Computing (MEC) applications has significantly enhanced the realm of security and surveillance, with Human Activity Recognition (HAR) standing out as a crucial application. The diverse sensors found in smartphones have made it convenient for monitoring applications to gather and analyze data, rendering them valuable for HAR purposes. Moreover, MEC can be employed to automate surveillance, allowing intelligent monitoring of restricted areas to identify and respond to unwanted or suspicious activities. This research develops a system using motion sensors in smartphones to identify unusual human activities. People's smartphones were employed to monitor both suspicious and regular activities. Information was collected for various actions categorized as either suspicious or regular. When a person performs a certain action, the smartphone records a series of sensory data, analyse important patterns from the basic data, and then determines what the person is doing by combining information from different sensors. To prepare the data, information from different sensors was aligned to a shared timeline. In this study, we used a sliding window approach on synchronized data to feed sequences into LSTM and CNN models. These models, which include initial layers of LSTM and CNN, automatically find important patterns in the order of human activities. We combined SVM with the features extracted by the shallow Neural Network to make a mixed model that predicts suspicious activities. Lastly, we compared LSTM, CNN, and our new shallow mixed neural network using a new real‐time dataset. The mixed model of CNN and SVM achieved an accuracy of 94.43%. Additionally, the sliding window method's effectiveness was confirmed with a 4.28% improvement in accuracy.

Abstract Wearable sensor‐based devices like actigraphs collect motor activity data which provide objective measures of physical activity. This research puts forward a novel methodology for assessment of objective sleep quality using actigraph recordings of motor activity. High level features of sequential motor activity data are extracted using Long‐Short Term Memory (LSTM) model which are then paired with a significant statistical feature namely, zero percent which describes the percentage of events with zero activity over a series. Overlapping sliding window is used to input sequences into LSTM to capture superior features in activity recordings. The predictive ability of the combined feature vector is evaluated using support vector machine (SVM) classifier. This hybrid LSTM‐SVM framework is validated on a benchmark dataset namely, the MESA Actigraphy dataset and achieves an accuracy of 85.62% for sleep quality prediction. Effectiveness of overlapping sliding window and statistical feature are evaluated, and their significance is validated. It is validated that the concept of overlapping sliding window improves the performance accuracy by 3.51% and the use of discriminative statistical feature improves sleep quality prediction task by 2.95%. Comparison with state of the art validates that this is the first study using objective sleep quality indicator for assessment of sleep quality via actigraph‐based motor activity data.

The automated evaluation of pain is critical for developing effective pain management approaches that seek to alleviate pain while preserving patients’ functioning. Transformer-based models can aid in detecting pain from Hindi text data gathered from social media by leveraging their ability to capture complex language patterns and contextual information. By understanding the nuances and context of Hindi text, transformer models can effectively identify linguistic cues and sentiments and expressions associated with pain, enabling the detection and analysis of pain-related content present in social media posts. The purpose of this research is to analyze the feasibility of utilizing NLP techniques to automatically identify pain within Hindi textual data, providing a valuable tool for pain assessment in Hindi-speaking populations. The research showcases the HindiPainNet model, a deep neural network that employs the IndicBERT model, classifying the dataset into two class labels {pain, no_pain} for detecting pain in Hindi textual data. The model is trained and tested using a novel dataset, दर्द-ए-शायरी (pronounced as Dard-e-Shayari ), curated using posts from social media platforms. The results demonstrate the model's effectiveness, achieving an accuracy of 70.5%. This pioneer research highlights the potential of utilizing textual data from diverse sources to identify and understand pain experiences based on psychosocial factors. This research could pave the path for the development of automated pain assessment tools that help medical professionals comprehend and treat pain in Hindi-speaking populations. Additionally, it opens avenues to conduct further NLP-based multilingual pain detection research, addressing the needs of diverse language communities.

Abstract The internet of things (IoT) and the smart city crusade has seen an exponential increase in the number of data points we collect. Cities can use this data for an exceedingly heterogeneous range of purposes such as traffic, parking and public safety. Quite evidently, public safety is an essential pillar of these interconnected, instrumented and intelligent cities. The big data from cyber‐physical systems and social media platforms can be used for predictive policing to identify potential criminal activities, abuse, offenders, and victims of abuse. This study offers a systematic literature review on the use of soft computing techniques for abuse detection in the complex cyber–physical–social big data systems in cognitive smart cities. The objective is to define and identify the diverse concept of abuse and systematize techniques for automatic abuse detection for cyber abuse detection on social media and real‐time abuse detection using IoT. The cyber abuse studies on social media platforms have further been categorized as cyber‐hate and cyberbullying whereas the real‐time abuse includes studies using Internet of cyber‐physical systems. As in a cognitive smart city, citizens expect more from their urban environments with minimal intervention, this study helps to establish the need to capture situational context and awareness in real‐time and foster the need to develop a proactive as well as reactive safety mechanism to mitigate the risks of online abuse. The need of the hour is to entrench self‐learning, thinking and understanding capabilities into the physical and social world for reinforcing intelligent mechanisms which can detect assaults and disorderly conduct.

scale up the online disinformation operation, unsubstantiated pieces of information on social media platforms can cause significant havoc by misleading people. It is essential to develop models that can detect rumours and curtail its cascading effect and virality. Undeniably, quick rumour detection during the initial propagation phase is desirable for subsequent veracity and stance assessment. Linguistic features are easily available and act as important attributes during the initial propagation phase. At the same time, the choice of features is crucial for both interpretability and performance of the classifier. Motivated by the need to build a model for automatic rumour detection, this research proffers a hybrid model for rumour classification using deep learning (Convolution neural network) and a filter-wrapper (Information gain-Ant colony) optimized Naive Bayes classifier, trained and tested on the PHEME rumour dataset. The textual features are learnt using the CNN which are combined with the optimized feature vector generated using the filter-wrapper technique, IG-ACO. The resultant optimized vector is then used to train the Naïve Bayes classifier for rumour classification at the output layer of CNN. The proposed classifier shows improved performance to the existing works.

Cyberspace has been recognized as a conducive environment for use of various hostile, direct, and indirect behavioural tactics to target individuals or groups. Denigration is one of the most frequently used cyberbullying ploys to actively damage, humiliate, and disparage the online reputation of target by sending, posting, or publishing cruel rumours, gossip, and untrue statements. Previous pertinent studies report detecting profane, vulgar, and offensive words primarily in the English language. This research puts forward a model to detect online denigration bullying in low-resource Hindi language using attention residual networks. The proposed model Hindi Denigrate Comment–Attention Residual Network (HDC-ARN) intends to uncover defamatory posts (denigrate comments) written in Hindi language which stake and vilify a person or an entity in public. Data with 942 denigrate comments and 1499 non-denigrate comments is scraped using certain hashtags from two recent trending events in India: Tablighi Jamaat spiked Covid-19 (April 2020, Event 1) and Sushant Singh Rajput Death (June 2020: Event 2). Only text-based features, that is, the actual content of the post, are considered. The pre-trained word embedding for Hindi language from fastText is used. The model has three ResNet blocks with an attention layer that generates a post vector for a single input, which is passed through a sigmoid activation function to get the final output as either denigrate (positive class) or non-denigrate (negative class). An F-1 score of 0.642 is achieved on the dataset.

Cyberbullying is the online fight between individuals or groups, and it can be viewed like harassment, rumor, denigration, exclusion, etc. Social networks are the main source of cyberbullying as various types of users interact with each other through text, audio, video, and images. One set of users uses the social media for the benefit of the whole society and the other set of users uses the social media for destructive purpose in the form of spreading rumors, harassment or to threaten others, etc., which is also called anomalous behavior. This article worked to detect the anomalous patterns using an exponential function and then proceeds to find the category of cyberbullying to which user belongs using subtractive clustering and fuzzy c-means clustering. The identification of category helps to find the extent to which these messages are harmful and based on which the culprit is apprehended or entrapped. State-of-the-art studies are focused on the detection of cyberbullying but this article captured different categories of cyberbullying.

Medical imaging has evolved to become an essential tool for screening and diagnosing diseases, but they have certain limitations just like every other technology. X-rays, which is one of the most common radiological examinations, is not immune to imperfections. In this paper, we aim to tackle one such imperfection in X-rays, which is the presence of undesirable noises which causes aberrations in the output projections. This makes diagnosis and analysis difficult since such noises shroud the intricate details that these images contain. Distinctive denoising algorithms have been proposed in the past for a spectrum of vision datasets, but a very few of them are for X-rays. We introduce a new denoising network called SkiDNet, a deep learning approach using an encoder-decoder architecture with skip connections of varying length. The network has been trained on the NIH Chest X-Ray Dataset. With the unique properties injected by different types of connections, SkiDNet is able to surpass the performance of existing models. Furthermore, adopting a different approach to weight initialization and batch normalization makes the network more robust. Denoised X-rays obtained from the network were objectively evaluated using different metrics namely mean squared error, peak signal-to-noise ratio, and the structural similarity index.

Social media is the collection of different social networks containing different type of information. The information may be in the form of text, video, audio and image. Also various categories of users, various types of communities are available on social network. This research reports on the extraction of new keywords from messages posted on social media which will be helpful in the identification of various communities, category of user and hidden pattern present in the social media. In this paper, we applied Probalistic approach to recognize the new keywords and assign the group accordingly. State-of-the-art studies performed detection on the basis of existing keywords but the proposed approach take decision based on the existing keywords and also on new keywords extracted from social media.

The Online Social Network (OSN) has evolved as a popular platform enabling rich topic-centric interactions and serving as a medium to facilitate online radicalization (Behr et al. 2013 Behr, I., A. Reding, C. Edwards, and L. Gribbon. 2013. Radicalization in the digital era. The use of the Internet in 15 cases of terrorism and extremism. http://www.rand.org/content/dam/rand/pubs/research_reports/RR400/RR453/RAND_RR453.pdf (accessed November 5, 2013). [Google Scholar]). Keeping in view the growing need of uncovering online radicalization, we focus on the information network of Twitter and present an approach for identifying dynamic communities, which arise due to “radical” topic-centric user interactions. The approach at successive timestamps deploys evolving topic-entity maps along with evolving interaction graphs. We propose “Rate of Overlap (ROAct)” to determine the similarity among successive community timestamps. We further validate our approach using an open dataset of criminal offences in the city of Denver, Colorado. The approach presented is simple, fast, and effective for dynamic identification of topic-centric communities and, thus, will enable law enforcement agencies to identify hidden radicalization.

Grid environment is a virtual organization with varied resources from different administrative domains; it raises the requirement of a secure and reliable protocol for secure communication among various users and servers. The protocol should guarantee that an attacker or an unidentified resource will not breach or forward the information. For secure communication among members of a grid group, an authenticated message transferring system should be implemented. The key objective of this system is to provide a secure transferring path between a sender and its authenticated group members. In recent times, many researchers have proposed various steganographic techniques for secure message communications. This paper proposes a new secure message broadcasting system to hide the messages in such a way that an attacker cannot sense the existence of messages. In the proposed system, the authors use steganography and image encryption to hide group keys and secret messages using group keys in images for secure message broadcasting. The proposed system can withstand against conspiracy attack, message modification attack and various other security attacks. Thus, the proposed system is secure and reliable for message broadcasting.

This paper involves examining the key issues and processes involved in requirement engineering for localisation. Analyzing the localisation method for development of localisation tool to support Indian languages. Reviewing various initiatives taken up Department Of Information Technology (DIT), IBM India and Microsoft India with respect to localisation.