A Data Mining Approach to Identify the Stress Level Based on Different Activities of Human

Abstract

Stress is one of the biggest realities in our modern lives because of the rapid variations in human lives and it induces depression. Depression is an illness characterized by anxiety and gloominess felt over a phase of time. Some signs of depression matched with other physical illnesses implying huge trouble in diagnosing it. In this analysis, we have tried to identify the reason for depression among students, based on their nature. We have collected data and generated a dataset that contains 539 instances containing 23 unique attributes individually. By using this data, we created a system that helps to identify the reason for depression. In this paper, a dataset has been analyzed to identify the rate of depression among students using Multilayer Perceptron (MLP), Multi-objective Evolutionary Algorithm and Fuzzy Unordered Rule Induction Algorithm. With the assistance of 100-fold-cross validation, we measure the validity of data that is collected by us, and the performance matrix helps us to report the evaluation of data. This evaluation report has shown us the accuracy and effectiveness of constructing a model to predict the reason for depression. We have got 90.90% accuracy by using Multilayer Perceptron, 92.95% accuracy by using the Fuzzy Unordered Rule Induction Algorithm and 92.76% accuracy by using Multi-objective Evolutionary Algorithm. Our main goal is to identify the rate of depression among students based on human nature.