Analisis Komparatif Metode MABAC, TOPSIS, dan SAW pada Sistem Pendukung Keputusan Pemilihan Tools AI

Engineering notes

Engineering notes will be added by the aipentium editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为大语言模型、生成式AI、ChatGPT相关技术、计算机视觉、深度学习等高价值论文补充中文说明。

Original abstract

Selecting appropriate artificial intelligence (AI) tools poses a significant challenge for individuals and organizations given the numerous alternatives available with varying characteristics. This study aims to compare three Multi-Criteria Decision Making (MCDM) methods, namely MABAC (Multi-Attributive Border Approximation area Comparison), TOPSIS (Technique for Order Preference by Similarity to Ideal Solution), and SAW (Simple Additive Weighting), in a decision support system for AI tools selection. Criteria weighting is performed objectively using the Entropy method. Four alternatives are evaluated: ChatGPT, Gemini, Claude, and Copilot, based on ten criteria: price, ease of use, features, integration, support, security, speed, accuracy, templates, and learning curve. The results show that all three methods produce consistent rankings, with ChatGPT ranked first, followed by Copilot second, Claude third, and Gemini fourth. Consistency between methods is measured using Spearman correlation, yielding rs values of MABAC-TOPSIS = 1.0000, MABAC-SAW = 1.0000, and TOPSIS-SAW = 1.0000. These results indicate that all three methods provide highly consistent outcomes and can complement each other. This research contributes methodologically to the application of MCDM in the AI tools selection domain and serves as a practical reference for users in determining the most suitable AI tools for their needs.

Links and sources

Comments