A Data Mining Approach to Identify Human Behavior on Different Activities of Student

Human behavior involves a broad range of behaviors cultural elements, arrogances, feelings, values, integrity, consultant, rapport, hypnosis, influence, coercion, and heredity. The use of social media is quickly increasing at the moment. The situation is much more concerning for those users who share personal information on multiple social media platforms. Their actions can be seen on social media. We explored several human behavior activities on social media in this our research paper. To ensure the validity of the acquired data, we applied 10-fold cross-validation and presented the results in the form of a performance matrix. Finally, we have used three classifier's algorithms to identify the best result for this research. The final result reflects the validity and quality of the obtained data, as well as the accuracy and efficiency with which a model to study the causes of human behavior changes have been built. We have applied 3 different algorithms (Multilayer Perceptron, Random Forest, and AdaBoostM1), where we have gotten best accuracy from 99.20% by using AdaBoostM1 algorithm and 99.70% from Multilayer Perceptron and 99.60% by using Random Forest algorithm.