IMPLEMENTATION OF MACHINE LEARNING APPROACH TO PREDICT LECTURER ACADEMIC PERFORMANCE INDEX Sucianna Ghadati Rabiha, Yulyanty Chandra, Willy Johan Widjaja Saputra
Universitas Bina Nusantara
Abstract
To achieve high student involvement in the education process, a high level of teacher involvement is needed. The main contribution of this research is to implement Machine Learning technique in predicting Lecturer Academic Performance Index. The techniques we use in designing the prediction model are Naive Bayes, K-Nearest Neighbor (KNN), and Neural Network (NN). this study uses a Neural Network approach to further analyze the effect of each variable on the accuracy of the IKADQ value. the accuracy value of the index prediction generated when using variables on the Teaching and Learning Method dimension is quite high, namely 70.35%, with a precision value of 16.45% and recall of 19.33%. And when viewed from the accuracy value generated for each variable, the best value that greatly influences the accuracy in predicting the index value is the Commitment variable, where the accuracy reaches 64.27%.
