Prediction of Human Suicidal Tendency based on Social Media using Recurrent Neural Networks through LSTM

Suicida1 Tendency Prediction utilizing data from Social Media is a recent factor that influences suicide-related event behavior. Suicide is the world’s eleventh leading cause of death, according to the World Health organization (WHO). Suicide identification is influenced by societal and environmental factors as well as individual factors. Suicide has recently become a ‘social’ menace. Twitter is rapidly being explored as a source that shows and detects a person’s psychological status. The goal of this work is to check if tweets about suicides can be categorized based on their content. This is accomplished through the repetitive application of deep learning and machine learning algorithms, as well as the involvement of coders. Twitter data has been scrutinized for a series of suicide words and remarks using the public API. Suicide related tweets are gathered using keywords connected to suicide and saved in a CSV file, which serves as the dataset. A recursive neural network is then used to classify these tweets in order to determine whether or not a suicide can take place. The proposed model identifies users who have a proclivity for suicide, allowing them to be approached and referred for psychiatric care, thereby benefiting them and potentially saving their lives.