Sara, Muqsitoh Aldila Andela (2025) Data Mining Untuk Klasifikasi Tingkat Burnout Mahasiswa/i Dalam Menyelesaikan Tugas Akhir Di Universitas Islam Riau Menggunakan Metode Random Forest. Other thesis, Universitas Islam Riau.

Text 203510276.pdf - Submitted Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Completing the final project is a mandatory requirement for students to get a Bachelor's degree. The challenges faced such as high workload, time pressure, and significant responsibilities, can have an impact on the mental and emotional wellbeing of students. In some cases, this condition can lead to burnout, which can be recognized by three main dimensions, namely, emotional saturation, cynicism, and decreased personal achievement. This research aims to categorize the burnout level of students in completing the final project and develop a system that helps assess the burnout level. The method used is the Random Forest method, one of the decision tree-based classification methods. This method works by building a number of trees through bootstrap sampling, selecting a random subset of predictors, and calculating entropy and information gain to build a decision tree. The classification result is determined based on the most votes from all trees in the model. The random forest method was used to identify the optimal value of n_estimators. The test results show that with n_estimators = 11, the classification system achieved accuracy 88.33%, precision 89.52%, recall 72.16%. These results show that the level of burnout can be accurately predicted by the random forest method. thus the classification of the level of burnout in students in completing the final project can be used as a good alternative in identifying the level of burnout.

Actions (login required)

View Item