Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Journal Of Information system and Computer science

MENENTUKAN PREDIKSI REKOMENDASI BIMBINGAN KONSELING SISWA SEKOLAH MENENGAH KEJURUAN MENGGUNAKAN K-NEAREST NEIGHBOR Wildani Eko Nugroho; Heru Saputro
Journal of Information System and Computer Vol. 2 No. 2 (2022): Desember 2022
Publisher : Universitas Islam Nahdlatul Ulama Jepara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34001/jister.v2i2.409

Abstract

The problem of counseling guidance is one of the problems found in schools which is one thing that is difficult to predict. Accurate predictions are needed to support decision making for those who have policy in making decisions. One of the things that can be done to make predictions in providing guidance and counseling recommendations is to use the K-Nearest Neighbor method. Of the 388 students in the school, tests or experiments can be carried out using the K-Nearest Neighbor algorithm method. The results obtained from the experiment were 96.40% with 362 students who did not have to do counseling guidance, and 62 students who had to do counseling guidance. From the experimental results, it is hoped that the quality of the education system being carried out is expected to be further improved because it will affect the quality of the school.
PEMILIHAN FITUR DENGAN FORWARD SELECTION PADA METODE K-NEAREST NEIGHBOR UNTUK PREDIKSI BIMBINGAN KONSELING SISWA Wildani Eko Nugroho
Journal of Information System and Computer Vol. 3 No. 1 (2023): Juli 2023
Publisher : Universitas Islam Nahdlatul Ulama Jepara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34001/jister.v3i1.651

Abstract

One of the most difficult issues to predict in schools is that of guidance and counseling. For individuals who have the authority to make decisions, accurate predictions are needed to support decision-making. KNN is one of the methods that can make predictions, especially for counseling guidance predictions. The KNN algorithm approach used to conduct tests and experiments on 388 student data obtained an accuracy of 96.40%. After optimization using feature selection with forward selection, the accuracy increased by 0.51%. The final result of the accuracy obtained after experiencing an increase in the level of accuracy value of 96.91% with the number of students who do not have to do counseling guidance is as many as 362 students, and 62 students must do counseling guidance.