MACHINE LEARNING APPROACHES FOR ANALYZING WOMEN'S SAFETY USING TWITTER

Women's safety in urban India remains a critical issue, with reports of harassment and violence affecting their daily lives.This study leverages the rich, real-time data available on Twitter to analyze the public's perception of women's safety in various Indian cities.By employing machine learning techniques on tweets, we aim to provide a comprehensive analysis of safety-related sentiments and identify patterns indicative of unsafe conditions.The approach involves collecting tweets related to women's safety, preprocessing the text data, and applying sentiment analysis and topic modeling.Key machine learning algorithms, including Support Vector Machines (SVM), Random Forest, and Long Short-Term Memory (LSTM) networks, are utilized to classify sentiments and uncover trends.Geospatial analysis is also integrated to map the geographical distribution of sentiments, enabling the identification of high-risk areas.The findings reveal significant variations in safety perceptions across different cities and times, with social events and policy changes impacting public sentiment.This study not only highlights the potential of social media data in assessing women's safety but also provides actionable insights for policymakers and law enforcement to enhance urban safety measures.