Paper Title
FORECASTING THE INSPECTION OF SUICIDE ATTEMPTS USING MACHINE LEARNING ALGORITHMS

Abstract
The escalating concern surrounding suicide rates has become a focal point in contemporary society, drawing considerable attention. Although depression is often identified as a primary catalyst, various other factors such as economic hardships (e.g., unemployment), societal issues (e.g., conflicts over dowries), and incurable diseases (e.g., AIDS) exert significant influences. Despite the emergence of AI-driven chatbots intended for suicide prevention, their efficacy presently hovers around 75%. In response, machine learning algorithms are being harnessed to enhance the precision of suicide attempt prediction. Preliminary data analysis offers insights into suicide metrics and the intricate dynamics of contributing elements, with visual representations aiding in comprehending the patterns in suicide endeavors. This paper delves into the existing methodologies employed in crafting suicide prediction models, scrutinizing their merits and demerits while executing a comparative analysis on their efficacy.Moreover, we conducted a comparative evaluation among four distinct machine learning algorithms: XGBoost, Random Forest, Decision Tree, and SVM, aiming to ascertain the most accurate results. The findings revealed that XGBoost attained the highest accuracy at 98.9%, followed by Random Forest with 98.0%. Conversely, SVM achieved an accuracy of 75%, while Decision Tree lagged behind, exhibiting the lowest accuracy. Keywords - XG-Boost, Random Forest, Support Vector Machine (SVM), Decision Tree, Suicide.