Temporal, Demographic, and Geographical Analysis of Violent Events in Bangla News Media Using NLP Techniques

Temporal, Demographic, and Geographical Analysis of Violent Events in Bangla News Media Using NLP Techniques

Blog Article

Abstract Text classification plays a vital role in natural language processing (NLP), especially in the context of rapidly expanding textual data.This paper presents an automated approach to detecting and classifying violent incidents from a large corpus of Bangla news articles, specifically focusing on incidents such as Murder, Rape, Kidnap, Clash, and Suicide.
Understanding the impact of the COVID-19 pandemic on U.S. older adults: self-reported pandemic-related concerns and consequences in a cross-sectional survey study
.Our methodology employs a semi-supervised learning framework, leveraging a small labeled dataset to train an initial predictive model, which is then expanded to create a comprehensive dataset suitable for deep learning applications.
SHOPPING CENTERS: UMA RELAÇÃO ENTRE OS ATRIBUTOS DE ESCOLHA PELOS CONSUMIDORES VERSUS OS ATRIBUTOS VALORIZADOS PELOS GERENTES
.We explore various feature extraction techniques, including N-gram, TF-IDF, Word2Vec, FastText, and BERT, with the BERT-based classifier achieving the highest accuracy of 92% on the test set.Utilizing this classifier, we conduct a detailed trend and pattern analysis over a 5-year period (2014–2018), revealing significant insights into the frequency and distribution of violent events.

Our findings highlight distinct trends in the occurrence of these incidents, including variations by month and age group, and geographical intensity, ultimately contributing to a better understanding of violent events in the context of Bangla news media