A Subjective and Objective Integration Approach of Determining Weights for Trustworthy Measurement

Abstract

The trustworthiness of the software is crucial to some safety-critical areas. Often, it is measured through the trustworthy degree of software attributes. Weights play key roles in obtaining accurate measurement. How to determine weights of different attributes is a challenge. In this paper, we propose an approach for determining weights. First, we construct positive reciprocal matrices by the evaluations of different experts, and the subjective weights are obtained by aggregating the matrices. In addition, we prove that the aggregation matrix preserves some properties, which further shows that our method is reasonable. Then, we acquire objective weights based on the trustworthy degrees of the attributes and the subjective weights. Finally, we propose an approach for determining weights based on the subjective and objective integration. A case study is presented, which shows that the proposed method is effective.