Using Principal Component Analysis to support students' performance prediction and data analysis

We propose a method based on Principal Component Analysis (PCA) for predicting students’ performances and for identifying relevant patterns concerning their characteristics. The proposed method allowed us to study the predictive capability of students’ performances and the effectiveness of PCA for interpreting patterns in educational data. The proposed method was validated using two public datasets describing students achievements, as well as their social and personal characteristics. Experiments were conducted by comparing the predictive performances between the datasets presenting high and reduced dimensions. The results reported that PCA retained relevant information of data and was useful for identifying implicit knowledge in students’ data.