Classification Using Deep Learning Based on Selection Optimizing Counseling Data
Alexius Endy Budianto, Mohd Fairuz Iskandar Othman, Yahaya Abd. Rahim

Universitas Kanjuruhan Malang, Universiti Teknikal Malaysia Malaka

Faculty of Science and Technology, Kanjuruhan University Malang, Indonesia,
endybudianto[at]unikama.ac.id

Faculty of Information and Communication Technology at Universiti Teknikal Malaysia Melaka,
P031810018[at]utem.edu.my, mohdfairuz[at]utem.edu.my, yahaya[at]utem.edu.my

Abstract

Abstract. Counseling is professional assistance given by counselors to clients directly, to solve problems and improve their behavior. in the future. The discussion of the problem in question is in-depth which concerns important matters about the client, broad covering various aspects of the clients problem, and specifically to alleviate the clients pressing problems. The Support Vector Machine and Multi Layer Perceptron algorithm is applied to classify counseling data provided by professionals to clients with an accuracy of 86.39%. Data from Vocational Research in Malang City in 2018. To improve the optimal accuracy of the type of counseling data classification, this study proposes an algorithm optimization feature for the Deep Learning algorithm. The results showed a significant increase in accuracy. The value of classification accuracy in counseling data generated by the Deep Learning algorithm has increased to 97.37% with a better level of accuracy.

Keywords: Deep Learning, Optimize Selection, Data Counseling

Topic: Computer Science