Salvius Lengkong
Sam Ratulangi University

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Auditable Crowdsourced Digital Dictionary for Low-Resource Toulour with Gamification and Functional Evaluation Kenneth Palilingan; Salaki Joshua; Salvius Lengkong
SISTEMASI Vol 15, No 9 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i9.6936

Abstract

Toulour, also documented as Tondano/Toundano, is a low-resource Minahasan language whose digital lexical resources require accessible retrieval and accountable community maintenance. This study designs and evaluates an auditable crowdsourced and gamified Toulour–Indonesian digital dictionary in which participation incentives are separated from lexical authority. Using Design Science Research Methodology (DSRM), the artifact integrates bidirectional lexical access, contributor proposals, validator-controlled publication, persistent before/after provenance, experience points, quizzes, badges, feedback, and role-based administration. Functional evaluation used 50 predefined black-box cases structured by ISO/IEC 25010:2023 Functional Suitability and informed by ISO/IEC/IEEE 29119-4:2021; 49 cases passed (98.0%), with one oversized-query boundary failure retained. Supplementary evaluation included a pilot user study, automated application-surface security scanning, and staged local load testing. Of 40 questionnaire submissions, 34 eligible adults who had tried the application were analyzed; mean System Usability Scale (SUS) was 77.79, while descriptive perceived usefulness, perceived ease of use, and behavioral intention means were 4.10, 4.07, and 3.90 on a 1–5 scale. Final context-scoped OWASP ZAP verification reported 0 High, 3 Medium, 3 Low, and 4 Informational alert types. k6 testing at 1, 10, 25, and 50 virtual users observed no HTTP request failures, although latency rose substantially at higher concurrency and throughput gains flattened between 25 and 50 virtual users. The results support an operational integration of governed crowdsourcing, reconstructable provenance, participation-linked incentives, and complementary evaluation while retaining explicit pilot, local-deployment, and residual-security limitations.