Implementation of K-Nearest Neighbor for Classification Teacher Engagement Profiling and Interventions

In this study we used the K-Nearest Neighbor approach to classify teacher profiles based on the Index values obtained previously through the Indonesian Teacher Engagement Index application. We found that the accuracy of the K-Nearest Neighbor is quite good if it is tested using a small data set that is equal to 88.94% and RMSE 0.300. Whereas if tested again into greater data the resulting accuracy decreases to 79.75% and RMSE 0.373. Some intervention designs related to meaningfulness to increase teacher engagement are through Meaningful and Fulfilling Education programs, Meaningful and Fulfilling Work Life, Meaningful and Fulfilling Character, and Meaningful and Fulfilling Leaders.