Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : Journal of Comprehensive Science

Evaluasi Aplikasi M-Pajak Sebagai Layanan Digital Pajak dengan Pendekatan Information System Success Model dan Sentiment Analysis Hutama, Ferdian Al Ghifari Ganda
Journal of Comprehensive Science Vol. 4 No. 12 (2025): Journal of Comprehensive Science
Publisher : Green Publisher Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59188/jcs.v4i12.3799

Abstract

The M-Pajak application was introduced as part of Indonesia’s digital taxation initiative to streamline tax administration and expand service accessibility. Nevertheless, user feedback on the Google Play Store consistently indicates system instability, recurrent errors, and limited service responsiveness, suggesting that the application has not met its functional expectations. This study aims to evaluate the service performance of M-Pajak through the Information System Success Model (ISSM) integrated with sentiment analysis. A mixed-methods design was employed using secondary data obtained via web scraping, followed by text preprocessing and sentiment classification using a BERT-based model. From a total of 5,886 processed reviews, 73% were classified as negative, reflecting substantial user dissatisfaction and persistent technical deficiencies. The qualitative mapping of sentiment outputs onto ISSM dimensions reveals that system quality and service quality constitute the primary sources of failure, characterized by unstable features, error-prone operations, and unresponsive customer support. Information quality is also inadequate due to limited clarity and insufficient guidance embedded within the application. Consequently, user satisfaction and net benefits remain low, indicating that M-Pajak has yet to deliver meaningful improvements in digital tax service efficiency. These findings underscore the need for comprehensive technical refinement, enhancement of informational clarity, and strengthening of service support mechanisms. Future research may incorporate user interviews or system performance evaluation to enrich the assessment.