文献类型：会议论文

英文题名：Reports Aggregation of Crowdsourcing Test Based on Feature Fusion

作者：Cai, Lizhi[1,2];Wang, Naiqi[1,2];Chen, Mingang[1];Wang, Jin[1,2];Wang, Jilong[1,2];Gong, Jiayu[1]

机构：[1]Shanghai Dev Ctr Comp Software Technol, Shanghai Key Lab Comp Software Testing & Evaluati, Shanghai, Peoples R China;[2]East China Univ Sci & Technol, Sch Informat Sci & Engn, Shanghai, Peoples R China

会议论文集：21st IEEE International Conference on Software Quality, Reliability and Security (QRS)

会议日期：DEC 06-10, 2021

会议地点：Hainan, PEOPLES R CHINA

语种：英文

外文关键词：crowdsourcing testing; similarity calculation; feature fusion

摘要：In recent years, a new testing method based on the concept of crowdsourcing has made great progress. Developers upload the project to the crowdsourcing test platform and recruit a large number of crowdsourcing workers for testing, so that the testing process has higher test adequacy, faster testing speed and lower testing cost. However, the test reports submitted after the test have serious problems such as large quantity and high similarity, resulting in the failure to achieve the expected results. Based on the method of feature fusion, this paper integrates the text description information, bug type information and screenshot information of crowdsourcing test reports, clusters crowdsourcing test reports through the calculation of similarity between reports, and finally achieves better results.