Published Paper Details:

The AI-Powered Safety Analysis of Cities for Women project aims to address the critical issue of women's safety in urban environments through the use of artificial intelligence and data-driven insights. With increasing incidents of crimes against women in public spaces, there is an urgent need for smart technologies that can assess safety levels based on real-world data. This system uses machine learning techniques to analyze multiple factors that influence the safety of a location, such as the city, area type, time of day, frequency of people, presence of police stations and bars, tier classification of the city, and the residential profile of the location. The dataset is pre-processed and cleaned, and features are transformed using One Hot Encoding to handle categorical variables effectively. A Logistic Regression model is trained using a robust pipeline to classify locations as either "Safe" or "Unsafe" with a strong focus on precision and recall for the "Safe" class. The model is evaluated using standard metrics such as accuracy, F1-score, and a confusion matrix to ensure reliability. Once trained, the model is saved and integrated into a web-based application built using Flask. The application provides an intuitive interface where users can select a city and time to predict the safety status of that location. If a prediction is made, the app displays not only the safety result but also a dynamic map view of the city using Google Maps embedding. Furthermore, it offers detailed emergency contact information, including the names and contact numbers of district officials like the Deputy Commissioner, emergency helpline numbers, emails, and official websites, sourced from a dedicated CSV file. By providing real-time, location-based safety predictions, the project empowers women to make informed decisions about their movements in urban areas. It also offers a potential tool for policymakers and law enforcement agencies to identify unsafe zones and allocate resources effectively. The combination of data science, geographic visualization, and public service contact integration makes this project a meaningful contribution toward enhancing women's safety in smart cities.