School-Level Based Classification of Indonesian Teaching Materials using KNN

The selection of teaching materials should be adjusted to the student’s ability at each level. It is intended that learning can be done well. However, along with the increasing of existing Indonesian language literature, it is possible that there are differences in the contents that are not well targeted. The school should be more selective choosing the teaching materials that are adjusted the students' abilities at all levels. Therefore an automated system is required that can classify teaching material documents based on the school level. The data used for classification is a text story in the teaching materials. One method of document classification is K-Nearest Neighbor. The K-Nearest Neighbor method is chosen because it has high accuracy and respite on which it is independent. From the research results can be seen that the highest accuracy, precision and recall value is 92.5%, 93.2% and 92%. All three results are achieved at k=5, so in this research used k value k=5 because it is more optimal than other k values.