Predicting students’ academic performance using a modified kNN algorithm

Abstract The target (dependent) variable is often influenced not only by ratio scale variables, but also qualitative (nominal scale) variables in classification analysis. Majority of machine learning techniques accept numerical inputs. Hence, it necessary to encode these categorical into values using encoding techniques. If the does have relation or order between its values, assigning numbers will mislead This paper presents a modified k-nearest-neighbors algorithm that calculates distances (nominal) without them. A student’s academic performance dataset used for testing enhanced algorithm. It shows proposed outperforms standard one needs nominal calculate distance variables. results show preforms 14% better than accuracy, and sensitive outliers.