Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Improvement with Chi Square Selection Feature using Supervised Machine Learning Approach on Covid-19 Data

Rektorat UDINUS 2021 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2021
Goals
Record
75f24828-3765-4e42-a848-88f9ef25e65a
Published
2026-10-30

Universitas Dian Nuswantoro conducted a study on Covid-19 data entitled Improvement with Chi Square Selection Feature using Supervised Machine Learning Approach on Covid-19 Data. The activity involved Rosidin, Sabir; Muljono; Fajar Shidik, Guruh; Zainul Fanani, Ahmad; Al Zami, Farrikh; and Purwanto as participants. The study aimed to improve earlier results whose accuracy was reported as 80.90% by adding a Chi-Square feature selection method to a supervised machine learning approach.

The study used two algorithms, namely the Naïve Bayes algorithm and Support Vector Machine (SVM). The testing process compared results before and after the addition of the Chi-Square selection feature. Before the feature was used, the SVM algorithm produced an accuracy of 85.56%, while the Naïve Bayes algorithm produced an accuracy of 85.19%. After the selection feature was added, the SVM algorithm produced an accuracy of 83.86%, and the Naïve Bayes algorithm produced an accuracy of 87.09%.

Chi-Square feature selection was described as an algorithm for selecting features by discarding irrelevant features. The testing showed different results for the two algorithms. The accuracy of SVM decreased after the selection feature was applied, whereas the accuracy of Naïve Bayes increased. Nevertheless, the study reported improved accuracy for both classification algorithms when compared with the previous studies. The findings therefore distinguish the effects of the feature-selection method according to the algorithm used and identify the Naïve Bayes result after feature selection as higher than its result before feature selection.

In relation to SDG 3, the activity contributes to health research through analysis of Covid-19 data using supervised machine learning and feature selection. This institutional activity demonstrates UDINUS participation in research addressing a health-related data-analysis topic during the 2021 period.

Reference https://www.scopus.com/pages/publications/85118951378?origin=resultslist