Conference paper

Inverse Document Frequency in K-Nearest Neighbour (K-NN) for Competition Recommendation based on Activity in Online Learning

Irawan Dwi Wahyono, Khoirudin Asfani, Mohd Murtadha Mohamad, Djoko Saryono, Hari Putranto, Mohd Nihra Haruzuan Bin Mohamad Said

7th International Conference on Electrical, Electronics and Information Engineering: Technological Breakthrough for Greater New Life, ICEEIE 2021 · 2021

1Citations
6Authors

Summary

Summary Inverse Document Frequency in K-Nearest Neighbour (K-NN) for Competition Recommendation based on Activity in Online Learning
Key quote

The recommendation application achieved an accuracy of 71.65% in matching student projects with competition requirements.

Wahyono et al., 2021

Details

Abstract

this research develops an application that can recommend for students to determine kinds of competition based on their activity in online learning. The application uses an algorithm of artificial intelligence that can make classification a requirement of competition and match the competition based on student's ability. This research uses a modification algorithm that text mining with TF-IDF and K-NN. Text mining is used to classification a final project student when the student wants to join in the competitions and classification kinds of competition. The K-NN algorithm is used to find near points between a final project of students and the requirement of the competition. The result of the recommendation gives an average of accuracy that is 71, 65%. © 2021 IEEE.

Affiliations

Faculty Of Engineering, Universitas Negeri Malang, Malang, Indonesia; Faculty Of Letter, Universitas Negeri Malang, Malang, Indonesia

Institute of Electrical and Electronics Engineers Inc.