Women Safety Analytics- Protecting Women from Safety Threats

In today’s world, incidents of harassment and threats against women are increasingly frequent across various settings. Social media platforms, in particular, play a significant Role in facilitating these global issues. This paper aims to develop a novel methodology that provides a robust safety mechanism for women, helping them avoid dangerous situations and individuals. The proposed solution, referred to as the Machine Learning based Social Threat Filter (MLSTF), introduces an innovative approach to threat detection. By focusing on threats originating from social media platforms primarily Facebook, Twitter, and similar sites—this paper demonstrates how to identify and process social media posts, analysing tweets and other posts to detect harmful content. This is achieved by calculating word intensities to assess the nature of the content. Additionally, this paper describes a method that captures real time data about women’s safety through a wearable device (SWD) designed to monitor individual users continuously. This SWD incorporates intelligent sensors, including a body position identifier, GPS, an accident detection sensor, a small microphone to capture distress signals like a black box, and a pen hole camera for location based imaging. Together, these sensors gather real time data on a user’s safety, transmitting it to a server for processing and analysis, ensuring responsive action where needed.