AI and IoT based Mobile Application for Women Safety
M. Amina Begum, Badadapure Pravinkumar Rajkumar, Neelam Labhade-Kumar, Veeramalai Sankaradass, C. Sathya, Bandi Bhaskar
Abstract
As the number of offenses committed against women continues to rise, women's security has become an incredible thing. Unacceptable levels of physical violence towards women can be found in all regions of the globe. Due to ineffective monitoring, this has picked up speed. There is currently no viable option to address this issue. The current crop of apps and gadgets is largely ineffective because they require so much user input. Our work is an attempt to address this issue. We create a mobile application to keep women safe. There are two phases to this study. In the first phase, sensor data under both high-and no-risk scenarios are used to train the Artificial Intelligence (AI) model and then tested. K-Nearest Neighbour (kNN) and the Support Vector Machine (SVM) are selected as the AI model. An evaluation will be conducted to determine which of the two models performs better. Successful models are then deployed to the cloud. In the second stage, we employed temperature and pulse sensors to identify the woman's activity, and the sensor data was transferred to the cloud, in which an AI model was deployed to analyse the collected data. Once that's done, the cloud will transfer the information to the mobile app. The woman's temperature and heart rate are displayed on the mobile app. The mobile app will automatically call and communicate location information to the registered individual through SMS if the cloud-based AI model determines that the woman is in a dangerous situation.