A Cascaded Unsupervised-Supervised NLP Pipeline for Detecting Accusatory Language in Public Procurement

RegNLI: Detecting Online Product Misbranding through Regulatory and Linguistic Alignment

Multilingual Fake News Detection Using Machine Learning with Contextual-Based Feature Extraction

The proposed approach provides a simple and efficient solution for multilingual fake news detection in data-scarce environments with ensemble-based classifiers such as Random Forest and Gradient Boosting achieving reliable performance across both languages.

A Multi-Algorithm Model Selector for Sentiment Classification in Public Evaluations of Government Performance Using GovBERTic

Social media has become a dynamic arena through which citizens express evaluations of government performance, yet the scale, heterogeneity, and informality of online discourse complicate systematic issue detection and sentiment interpretation. This study aims to develop GovBERTic (Government BERTopic), a reproducible computational pipeline that operationalizes public opinion analysis by linking semantic topic discovery, multi-algorithm sentiment model selection, and topic-level sentiment aggregation within a single workflow. The analysis used 7,877 tweets collected from X concerning the Prabowo–Gibran administration, which were cleaned and filtered to produce 6,569 valid tweets. BERTopic, supported by multilingual sentence embeddings, UMAP, HDBSCAN, and class-based TF-IDF, was applied to identify dominant discourse themes, while CatBoost, Multi-Layer Perceptron, and Complement Naïve Bayes were compared through a model selector for three-class sentiment classification. BERTopic identified eight main topics with a Topic Diversity score of 0.9125, covering government policy, development, food security, education, drug eradication, Papua, and Palestine. Complement Naïve Bayes achieved the best relative performance, with 64.61% accuracy and a 59.21% F1-score, indicating a moderate but computationally efficient baseline rather than definitive high-accuracy sentiment monitoring. Topic-level sentiment analysis showed that negative sentiment dominated four topics, neutral sentiment dominated three topics, and positive sentiment appeared in one topic. These findings suggest that GovBERTic can support exploratory, data-informed government communication analysis by connecting salient policy issues with their associated public sentiment patterns.

AN NLP-BASED MACHINE LEARNING FRAMEWORK FOR CYBERBULLINGDETECTION THROUGH SENTIMENT ANALYSIS

Cyberbullying has become a major problem in the digital world, with negative consequences for both individuals and the general well-being of society. Accurately identifying cyberbullying on social media platforms—which account for a sizable portion of digital communication—is a workable answer to this pervasive problem. While machine learning algorithms and pre-trained language models have been the mainstay of traditional techniques, these frequently encounter issues including excessive computational complexity and poor adaptation to subtle linguistic patterns. In order to enhance cyberbullying detection in online communication, this study suggests a sophisticated framework that combines Long Short-Term Memory (LSTM) networks with Natural Language Processing (NLP) techniques. To guarantee high-quality and noisefree input data, the system uses sophisticated text preparation techniques as tokenisation, stop word removal, stemming, and lemmatisation. Embedding techniques are used to extract contextual patterns and sentiment features while maintaining semantic information. An LSTM model, which successfully captures the sequential and temporal dependencies in textual data, is then fed these processed inputs. This model is ideal for comprehending the dynamic nature of cyberbullying language. Additionally, resampling approaches are used to improve the robustness of the model without introducing bias in order to solve class imbalance in the multi-class context. The suggested solution shows how integrating deep learning with thorough NLP improves the precision and contextual awareness needed for successful cyberbullying detection. KEYWORDS: Natural Language Processing (NLP), Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNNs),

Bharat Sum: OCR-Enabled Multilingual News Summarization and Bias Analysis Framework

India is home to an incredible number of languages which leads to the production of substantial amounts of news articles provided in their regional languages including Telugu, Tamil, Hindi, Bengali, Kannada, and Malayalam. Unfortunately, current systems for processing this data do so independently, rather than as part of a comprehensive framework utilizing all the necessary components in one queue of processing pipelines; OCR extraction, Translation, Summarization, and Bias Detection must all be completed one at a time, and do not allow for seamless data passing between functions. In this paper, we will present Bharat Sum, a multilingual news summarization and bias detection system that incorporates OCR capabilities into five different types of processing stages; OCR Text Extraction using Tesseract, Automatic Language Detection using LangDetect, Topic Segmented Abstractive Summarisation using mT5, Translation to English using mBART, and Sentiment based Bias Classification using DistilBERT - all accessible through a single scalable architecture running on commodity hardware and implemented via Streamlit. Our testing involved 150 news articles covering each of the five languages named above. The results achieved were as follows; OCR extraction accuracy of 89.7%, Language Detection Accuracy of 94.2%, Summarisation Quality (ROUGE-L F1) of 92.1% Translation consistency of 91.4% and Sentiment Classification Accuracy of 88.6%. The average end-to-end processing time was between 10 and 16 seconds. Our analysis of Bharat Sum has revealed that it significantly outperforms previous single function systems by providing an Integrated, Real-Time Multilingual Processing capability which currently does not exist in this context. Bharat Sum has the potential to address significant gaps in the research literature regarding the analysis of Integrated Multilingual Media, and will likely serve as an economically viable solution for organisations conducting Digital Journalism, Media Monitoring, or Accessing Multilingual Information.

CVI-Validated Indo-Transformer Framework for Intelligent Cooperative Supervision

Cooperative supervision reports contain complex narrative structures and overlapping administrative terminology, complicating automatic classification into governance, risk profile, financial performance, and capital adequacy. Reliable automation is particularly important for accelerating the analysis of supervisory findings while addressing limited labeled data and imbalanced categories. This study aimed to develop and externally evaluate a text-classification framework combining quantitatively validated Generative Artificial Intelligence (GenAI) labeling with conventional and Transformer-based models. Data comprised 294 preprocessed sentences collected from the Department of Cooperatives, Small and Medium Enterprises, Industry, and Trade of Semarang Regency during 2023–2025. Few-shot annotations were generated using ChatGPT, Gemini, Perplexity, and DeepSeek, and three-model combinations were evaluated using the Content Validity Index (CVI); majority voting from the best combination established ground truth. TF-IDF with Logistic Regression and Support Vector Machine served as baselines, whereas IndoBERT and IndoRoBERTa represented contextual models. Performance was assessed through stratified five-fold cross-validation and external testing on 58 unseen sentences. ChatGPT–Gemini–Perplexity achieved the highest Scale-Level CVI of 0.898. IndoBERT obtained the best cross-validated F1-score of 0.9099, exceeding IndoRoBERTa (0.8217), Logistic Regression (0.8004), and SVM (0.7863). On unseen data, IndoBERT retained an F1-score of 0.862, compared with 0.759 for IndoRoBERTa. These findings demonstrate that CVI-validated ensemble GenAI can construct consistent labels for low-resource administrative texts and that IndoBERT provides the strongest and most stable generalization for cooperative supervision classification. The framework offers a practical basis for scalable annotation and reliable automated support for evidence-based supervisory decision-making.