Link: https://suicide-analysis.cyclic.app/
Please click on the videos below:
Suicides are increasing at an alarming rate all over the world and it is important to reduce the number of Suicides taking place. In this project we have used Machine Learning and other ideas could provide a better solution to this problem. We have taken various dataset’s such Suicide Rates over a period of time, Suicide Reasons, Emotions of a person. These factors are trained using various ML Algorithms such as Linear Regression, K-Nearest Neighbors, Random Forest , Decision Tree to visualize and as well as predict the Suicidal Rates.
As part of novelty in this project we have used Speech to Text to find the Emotions of the person and gives as the output using Emoji’s Reactions this is done using Nodejs , the second Part of novelty in this project is using Depression Test to find whether the person is Depression or not this uses Flask Framework in back-end to integrate with the front-end i.e., Web Application for implementation ML algorithm.
Thus we came to conclusion that Suicidal Rates are increasing rate and the Various factors can be used Determine the reasons of suicides. These Results would be helpful to reduce the Number of Suicidal Rates in the World and create Awareness among people.
The process of building an integrated online system for handling suicide is a novel idea. There are various types of resources existing online, but the challenge undertaken was to integrate many parts along with developing new modules including the Homepage, Helpline, Analysis of Suicide based various factors. The most important modules of our project are:
The ‘Speak Out’ part is a unique approach to understand and analyse the user’s state of mind. This novel idea can be used as an online self-diagnosis available anywhere, anytime. After the user speaks his thoughts out loud, its net sentiment will be calculated.
Loneliness is a major factor for being depressed. This led to the implementation of text to speech of the user input whose voice will act as a companion for the user during tough times. Rather than displaying the sentiment directly, a novel approach of using emoji’s was implemented. In times where texting is dominant and emoji’s are fast replacing words to convey emotions, this feature will serve as a testament to gain the comfort and trust of the user.
A novel machine learning model runs in the back-end to predict the user’s state of mind based on his input. In this User Interface, We have Depression Test i.e., Psychometric Test in which there are 10 general questions that the user has to enter based on the options provided such as 0- More Than Half the Days, 1- Nearly Everyday, 2-Not at all, 3-Several Days.
After the completion of the Depression Test it shows whether the person is being Depressed or not and the Output will be displayed as either “Yes” or “No” based on the options entered by the User. Planning for psychometric testing through design and reducing non random error in measurement will add to the reliability and validity of instruments and increase the strength of study findings.
Although a single factor cannot be pinpointed, among the common factors leading towards taking the extreme step, some of them can be analyzed through this project. A visualization of these could give us a better understanding of the causes of their suicide. A prediction plot of the overall rate of suicides in the future years is found using regression algorithms. A culminated model was created to predict the total count of suicide based on various factors including state, age group, and Year. These results would be helpful to reduce the number of suicide rates in the world and create awareness among people.
This website will act as one-stop solution to help and inform users about suicide data and to analyze their state of mind. The features implemented would be useful in the long run and are both scalable and adaptable. With the suicide death toll on the rise yearly, this research and website will act as a step in the right direction towards preventing future decisions to end one’s own life.