The Media is stated as the “fourth pillar of democracy" that influences day to day life of each individual. The falling rank of India in the World Press Freedom Index (WPFI) portrays biasness in news articles, especially for political news articles. 1 The biased headlines morph the perception of the people regarding the topic of interest and fill them up with baseless prejudice. Currently, thousands of Hindi newspaper publishers in India have more than 100 million subscribers. 2 Political bias detection from news is a fresh research area, researchers did not considers contextual analysis yet. This is the first political news bias detection model that considers neutrosophic sets for contextual word analysis to address inconsistency and indeterminacy, newly enhanced Hindi SentiWordNet and MuRil language model for encoded output and multiclass classification. This proposed novel model transforms input words into a neutrosophic domain and is characterized by degree of truth, indeterminacy, and false membership components. This set applies the concept of indeterminacy that improves the biasness detection accuracy for each news headline. The experimentation and evaluation of proposed model shows the better accuracy than the state-of-the-art methods, MuRil language model combined with neutrosophic sets improves 37.5% to 50% accuracy.

Civil unrest, a major trouble in the country's progress, requires timely detection and prevention. It causes numerous major issues, including loss of life and injury, resource depletion, political instability, and violations of human rights. Automating the early warning civil unrest event prediction with social media data becomes critically important. Existing baseline methods through text datasets obtained from social media have shown promising results. However, most existing baseline methods are domain-specific and lacking in robustness and generalization. As of now, there has been less work done addressing these issues in civil unrest event prediction. To overcome these limitations, this article presents a novel method as CUP_CDLSTM by combining the convolutional neural network-long short-term memory (CNN-LSTM) model and the pretrained distilBERT model to predict civil unrest event prediction using social media data. First, the proposed CUP_CDLSTM model utilizes the LSTM model to learn temporal features followed by CNN utilized to learn spatial features from the correlation matrix of different features. DistilBERT is utilized to generate weighted word embedding to advance the contextual features. The proposed CUP_CDLSTM model is trained with two datasets of different geographical locations for predicting civil unrest events which include spatial, temporal, and event weights as input features. The proposed CUP_CDLSTM model outperforms the baseline methods by up to 5% on both the Hong Kong protest dataset and the black lives matter (BLM) protest dataset in terms of accuracy. It has shown significantly faster training and inference times than existing baseline models.

Abstract Interaction via social media involves frequent code‐mixed text, spelling errors and noisy elements, which creates a bottleneck in the performance of natural language processing applications. This proposed work is the first approach for code‐mixed Hindi‐English social media text that comprises language identification, detection and correction of non‐word (Out of Vocabulary) errors as well as real‐word errors occurring simultaneously. Each identified language (Devanagari Hindi, Roman Hindi, and English) has its own complexities and challenges. Errors are detected individually for each language and a suggestive list of the erroneous words is created. After this, a fuzzy graph between different words of the suggestive lists is generated using various semantic relations in Hindi WordNet. Word embeddings and Fuzzy graph‐based centrality measures are used to find the correct word. Several experiments are performed on different social media datasets taken from Instagram, Twitter, YouTube comments, Blogs, and WhatsApp. The experimental results demonstrate that the proposed system corrects out‐of‐vocabulary words as well as real‐word errors with a maximum recall of 0.90 and 0.67, respectively for Dev_Hindi and 0.87 and 0.66, respectively for Rom_Hindi. The proposed method is also applied for state‐of‐art sentiment analysis approaches where the F1‐score has been visibly improved.

Depression, a prevalent mental health concern, requires timely identification and intervention. Automating the early stage identification of depression cues within social media text has become critically important. Existing methods for depression identification through text obtained from social media have shown promising results; however, these methods do not address the graphlike nature of social media data. To date, there has been little work addressing this problem by appropriately modeling social media data. This article aims to effectively harness the power of multiembedding techniques and graph attention networks (GATs) for depression identification. To overcome these limitations, the current study presents a novel methodology that combines multiple token-level embeddings, including BERT, RoBERTa, and DeBERTa, with GATs to leverage the graphlike nature of social media data and detect depression cues from such text. A dataset comprising 100 000 tweets has been curated using data from publicly available annotated tweet datasets. This dataset maintains a balanced distribution between depressive and nondepressive text samples. This was followed by a rigorous cleaning and preprocessing pipeline. Next, these data were transformed into a numeric feature matrix using multiple-word embeddings, which enabled the treatment of tweets as nodes in a graphlike structure. This graph was used for training a GAT model with multiple self-attentional layers, culminating in a linear layer for mapping traits to binary classes. The model presented in the current study achieves a precision of 97.2%, a recall of 96.4%, and an F1 score of 96.7% on a benchmark dataset.

Civil unrest is among the important hurdles in the countries’ progress as it deteriorates gross domestic product (GDP), international relations, foreign direct investment (FDI), globalization, public opinion, tourism, and businesses. Due to civil unrest a variety of serious problems, viz. loss of life/injury, resources, political stability, and human rights occur. Recently, few researchers have given insights on the prediction of occurrences of civil unrest events by using hypothesis testing and some basic machine/deep learning models. Important factors such as people’s emotions/sentiments, contextual information, and civil unrest events feature’ importance are ignored presently. For the first time, the proposed work overcomes all these research gaps by hybridizing the neutrosophic set, aspect-based sentiment analysis, particle swarm optimization (PSO), and hierarchical long short-term memory (hierarchical LSTM). Neutrosophic set along with aspect-based sentiment analysis has been used to get the sentiment and features’ importance. The resulting features’ weights have been optimized using PSO. For a more comprehensive understanding of the input sequence and feature weights, hierarchical LSTM has been used. Doing so obtained results that are more accurately improved for civil unrest events prediction. The performance of the proposed model has been evaluated and compared with state of art methods. Experimentation and evaluation show the proposed model outperforms the baseline methods by 3% to 15%on the standard datasets in terms of accuracy.

The explosive growth of social media has fueled an extensive increase in online freedom of speech. The worldwide platform of human voice creates possibilities to assail other users without facing any consequences, and flout social etiquettes, resulting in an inevitable increase of hate speech. Nowadays, English hate speech detection is a popular research area, but the prevalence of implicit hate content in regional languages desire effective language-independent models. The proposed research is the first unsupervised Hindi and Bengali hate content detection framework consisting of three significant concepts: HateCircle, hate tweet classification, and code-switch data preparation algorithms. The novel HateCircle method is proposed to detect hate orientation for each term by co-occurrence patterns of words, contextual semantics, and emotion analysis. The efficient multiclass hate tweet classification algorithm is proposed with parts of speech tagging, Euclidean distance, and the Geometric median methods. The detection of hate content is more efficient in the native script compared to the Roman script, so the transliteration algorithm is also proposed for code-switch data preparation. The experimentation evaluates the combination of various lexicons with our enriched hate lexicon that achieves a maximum of 0.74 F1-score for the Hindi and 0.88 F1-score for the Bengali datasets. The novel HateCircle and hate tweet detection framework evaluates with our proposed parts of speech tagging and Geometric median detection methods. Results reveal that HateCircle and hate tweet detection framework also achieves a maximum of 0.73 accuracy for the Hindi and 0.78 accuracy for the Bengali dataset. The experiment results signify that contextual semantic hate speech detection research with a language-independency feature offsets the growth of implicit abusive text in social media.

Abstract Keyword extraction is one of the most important aspects of text mining. Keywords help in identifying the document context. Many researchers have contributed their work to keyword extraction. They proposed approaches based on the frequency of occurrence, the position of words or the similarity between two terms. However, these approaches have shown shortcomings. In this paper, we propose a method that tries to overcome some of these shortcomings and present a new algorithm whose efficiency has been evaluated against widely used benchmarks. It is found from the analysis of standard datasets that the position of word in the document plays an important role in the identification of keywords. In this paper, a fuzzy logic-based automatic keyword extraction (FLAKE) method is proposed. FLAKE assigns weights to the keywords by considering the relative position of each word in the entire document as well as in the sentence coupled with the total occurrences of that word in the document. Based on the above data, candidate keywords are selected. Using WordNet, a fuzzy graph is constructed whose nodes represent candidate keywords. At this point, the most important nodes (based on fuzzy graph centrality measures) are identified. Those important nodes are selected as final keywords. The experiments conducted on various datasets show that proposed approach outperforms other keyword extraction methodologies by enhancing precision and recall.

In this article, we propose a system called “UTTAM,” for correcting spelling errors in Hindi language text using supervised learning. Unlike other languages, Hindi contains a large set of characters, words with inflections and complex characters, phonetically similar sets of characters, and so on. The complexity increases the possibility of confusion and occasionally leads to entering a wrong character in a word. The existence of spelling errors in text significantly decreases the accuracy of the available resources, like search engine, text editor, and so on. The proposed work is the first approach to correct non-word (Out of Vocabulary) errors as well as real-word errors simultaneously in a sentence of Hindi language. The proposed method investigates the human behavior, i.e., the type and frequency of spelling errors done by humans in Hindi text. Based on the type and frequency of spelling errors, the heterogeneous data is collected in matrices. This data in matrices is used to generate the suitable candidate words for an input word. After generating candidate words, the Viterbi algorithm is applied to perform the word correction. The Viterbi algorithm finds the best sequence of candidate words to correct the input sentence. For Hindi, this work is the first attempt for real-word error correction. For non-word errors, the experiments show that “UTTAM” performs better than the existing systems SpellGuru and Saksham.

In this article, we propose Fuzzy Hindi WordNet, which is an extended version of Hindi WordNet. The proposed idea of fuzzy relations and their role in modeling Fuzzy Hindi WordNet is explained. We mathematically define fuzzy relations and the composition of these fuzzy relations for this extended version. We show that the concept of composition of fuzzy relations can be used to infer a relation between two words that otherwise are not directly related in Hindi WordNet. Then we propose fuzzy graph connectivity measures that include both local and global measures. These measures are used in determining the significance of a concept (which is represented as a vertex in the fuzzy graph) in a specific context. Finally, we show how these extended measures solve the problem of word sense disambiguation (WSD) effectively, which is useful in many natural language processing applications to improve their performance. Experiments on standard sense tagged corpus for WSD show better results when Fuzzy Hindi WordNet is used in place of Hindi WordNet.