Predicting and Preventing Drastic Actions by Processing Heart Rate and Behaviors and Utilizing Innovative AI Methods
Created by Impana Chimmalagi in 2020 for the Synopsys Science & Engineering Fair https://docs.google.com/presentation/d/1mGRdw_YxRVHCbopEZuzNwCRYzYZxGEBbrf-1gePUgmo/edit?usp=sharing
In 2019, 419 shootings have occurred. That means that there were more shootings than days. A new approach is necessary for the betterment of society: we need something that can really get into the assailant’s minds, focus on their mindset, and urge them to rethink their decisions about inflicting harm to others, or themselves, in the near future.
To create a straightforward, adaptive mobile program that accurately records the mobile user’s heart rate and compares it to the user’s previously recorded heart rates, using artificial intelligence, if needed, to stimulate a thought-provoking conversation with the user, causing the user to rethink, question, or think again about his/her decision about an action he/she will commit in the near future.
The hypothesis was correct; A mobile program designed to record the user’s heart rate and compare it to previously recorded heart rates will be able to start a thought-provoking conversation with the user that can emphasize his/her importance as well as everyone else’s, potentially preventing the user from taking a drastic action. As those who are scared or anxious tend to experience an increase in heart rate, the measurement of heart rate proved to be a good indication of the user’s levels of fear or stress. Additionally, the program took 15 significant increases in heart rate and compared the user’s instantaneous heart rate to his/her smoothed average, which further customized the user experience before questioning the user, which would help prevent unjustified accusations of harmful behavior towards the user. For the “Before a Test” scenario, the chatbot launched for 40% of the trials, and for the “After Shuttle Run” scenario, the chatbot launched for 60% of the trials. For the “Relaxation App”, the chatbot launched for 40% of the trials, and for the “Horror Movie” scenario, the chatbot did not launch at all. SafeHavenBot is the first working prototype that accurately detects significant increases in the user’s heart rate and initiates a thought-provoking conversation with them, potentially preventing them from taking a drastic action.