PENERAPAN METODE K-NEAREST NEIGHBOR DALAM MEMPREDIKSI WAKTU KELULUSAN MAHASISWA SARJANA YANG BERMAIN GAME

Graduating from college is something that students really want. By graduating from college, it has become a sign for a student to become a worthy scholar to continue and enter the next level. Graduation time is influenced by the academic value obtained from a student. If a student gets a high score, the student's graduation will be faster or on time. On the other hand, if a student gets a score below the average, the student's graduation time will be longer. At this time, one of the causes of students getting low grades is because students who are so busy playing games neglect their lectures and lose concentration while studying. So this can affect the time of their graduation. Students should be able to control themselves to manage their time playing online games and lectures in order to complete their obligations as a student, and students who get low grades for playing games should also be aware that getting low grades continuously will result in the student being threatened with dropping out (DO).Therefore, an information system program was designed that can be used by students who like to play games to be able to predict their graduation time so that they can find out their graduation time. The design of this program applies the K-Nearest Neighbor method which is a classification technique for objects based on learning data that is closest to the object.The final result of the application of the K-Nearest Neighbor method in the program has its advantages and disadvantages. The classification process is strongly influenced by the large amount of training data, and the determined value of 'K' (neighbors). The more the amount of training data, the level of accuracy can be reduced. The level of accuracy in testing using training data is 230 data, and using test data as much as 30 data with several specified 'K' values, namely, 2, 3, 4, 10. Accuracy results with 4 K values used can reach an accuracy rate of 90% .