Abstract: The differences of users and use environments are mostly ignored in current trustworthiness evaluation of Web services, leading to few guarantees for the evaluation accuracy. This paper proposes a trustworthiness evaluation mechanism based on trustworthiness evidence. As for evidence collection, it introduces three mechanisms: invoking based on Web service, automatic feedback through container monitoring and collecting user feedbacks. In addition, it proposes a novel context-sensitive mechanism to evaluate trustworthiness based on user feedbacks, takes account of context related to the quality of service and further obviously improves the evaluation accuracy.

Key words: Web service, trustworthiness evaluation, trustworthiness evidence, user feedback

摘要： 现有Web 服务的可信评估机制大多忽视Web 服务使用者及应用场景的差异性, 其评估结果的准确性无法得到保证。提出了一种基于可信证据的Web 服务可信评估机制, 实现了可信证据的Web 服务调用主动探测、服务容器自动反馈和用户反馈收集, 同时提出了一种基于用户反馈的上下文敏感可信评估方法,全面考虑了Web 服务质量属性的上下文相关性, 进一步提高了Web 服务可信评估的准确性。

关键词: Web 服务, 可信评估, 可信证据, 用户反馈