Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Comparison of K-N earest Neighbor (K-NN) and Naïve Bayes Algorithm for Sentiment Analysis on Google Play Store Textual Reviews

Rektorat UDINUS 2021 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2021
Goals
Record
da03d984-a2d0-4854-801d-2eb29159fcf8
Published
2026-10-30

UDINUS conducted a research activity involving the analysis of user sentiment toward the IGROW platform on the Google Play Store. The activity included UDINUS researchers Mussalimun, Elvien Hastatomo Khasby, Gempita Icky Dzikrillah, and Muljono as participants. The study focused on a fintech lending or peer-to-peer lending application in the context of agricultural capital. IGROW was described as part of a crowdfunding company registered with the Indonesian Financial Services Authority and as a platform available on the Google Play Store.

The activity examined textual reviews and opinions submitted by users. The source describes these reviews as mixed, containing positive and negative opinions that may affect business development. To provide sentiment analysis of the IGROW platform, the study applied the K-Nearest Neighbor algorithm and the Naive Bayes classification algorithm. The comparison used accuracy, precision, and recall values for both classification approaches.

The reported results show that K-Nearest Neighbor achieved accuracy of 73.85 percent, precision of 76.60 percent, and recall of 85.71 percent. Naive Bayes achieved accuracy of 75.38 percent, precision of 80.95 percent, and recall of 80.95 percent. Naive Bayes produced slightly better predictions than K-Nearest Neighbor because it obtained slightly higher accuracy and precision values, while K-Nearest Neighbor obtained the higher recall value. The activity therefore provides a comparative assessment of two classification algorithms for reviews of a fintech platform connected with agricultural capital. The documented research presents an institutional activity of UDINUS focused on computational analysis of user opinions and comparison of classification results.

Reference https://www.scopus.com/pages/publications/85123644639?origin=resultslist