A Contributor-Based Segmentation Model for Open Source Software Source Code Trustworthiness Measurement

A Contributor-Based Segmentation Model for Open Source Software Source Code Trustworthiness Measurement

연구 분야: Software Development

학회: International Conference on AI Logic and Applications

This paper presents a trustworthiness measurement model for open source software (OSS) trustworthiness assessment. The model decomposes trustworthiness attributes at the source code level and evaluates contributor contribution values. By quantifying attributes such as security, maintainability, reliability, testability, and compatibility, and introducing a novel method for measuring contributor impact based on Abstract Syntax Trees (ASTs), our model provides a holistic view of OSS trustworthiness. Through contributor segmentation and weight assignment, it reflects the varying influence of contributors. Experimental validation using Huawei’s OpenEuler OSS demonstrates the model’s effectiveness, bridging theory and practice in OSS quality assurance and empowering stakeholders in critical system decisions.