RULE-BASED EXPERT SYSTEM FOR SOFTWARE VERSION CLASSIFICATION WITH DECISION RULES

Humberto Cuteso Matumueni 1; Lamento Emilio Adão2; Joana Bartolomeu3; Maria Afonsina Álvaro4; Domingos Pascoal Sebastião5

ISSN 2320-9186 Open Access

Article Details

RULE-BASED EXPERT SYSTEM FOR SOFTWARE VERSION CLASSIFICATION WITH DECISION RULES

Journal Global Scientific Journal (GSJ)

Published June 2026

Vol / Issue Vol. 14 · Issue 6

Edition Volume 14 Issue 6 June 2026 Edition

Authors

Humberto Cuteso Matumueni 1 Lamento Emilio Adão2 Joana Bartolomeu3 Maria Afonsina Álvaro4 Domingos Pascoal Sebastião5

Abstract

Software version control is an essential activity in software engineering, enabling the management, tracking, and maintenance of software products throughout their lifecycle. Traditional version control systems such as Concurrent Versions System (CVS), Apache Subversion (SVN), and Git provide mechanisms for storing, controlling, and recovering software versions; however, they do not perform intelligent classification of software releases. This study proposes the development of an expert system for software version classification using artificial intelligence techniques. The methodology involved knowledge acquisition through questionnaires applied to experts, the construction of IF...THEN production rules, the use of decision tables, and implementation using the Exsys Corvid platform. The expert system classifies software versions into categories such as Alpha, Beta, Stable, Outdated, and Critical based on predefined criteria. To evaluate the proposed approach, a comparative analysis was conducted between the expert system and traditional evaluation methods using Cohen's Kappa coefficient. The results demonstrated a fair level of agreement (K = 0.333) between evaluators, indicating that the expert system is capable of supporting decision-making processes related to software version management. The study concludes that integrating expert system techniques into software version control environments can improve consistency, automation, and decision support in software release management.

Cite This Article

Humberto Cuteso Matumueni 1, Lamento Emilio Adão2, Joana Bartolomeu3 et al. (June 2026). RULE-BASED EXPERT SYSTEM FOR SOFTWARE VERSION CLASSIFICATION WITH DECISION RULES. Global Scientific Journal (GSJ). Vol. 14, Issue 6.