Image-based computer identification device and method for mobile crowdsourcing test report

Abstract

The embodiment of the invention discloses an image-based computer recognition device and equipment for a mobile crowdsourcing test report, a computer readable storage medium and a method. The device comprises a to-be-identified test report acquisition module used for acquiring a to-be-identified test report of a crowdsourcing test of a mobile application program; a feature extraction module used for extracting text features and image features of each test report in the sample set; a similarity calculation module used for respectively calculating text similarity and image similarity among the test reports in the sample set; a distance matrix generation module used for generating a distance matrix of the to-be-identified test report according to the text similarity and the image similarity;and an error identification module used for performing grouping identification on the to-be-identified test reports based on the distance matrix so as to cluster the to-be-identified test reports intothe test report groups describing the same defects. The crowdsourcing test report of the mobile application program is efficiently and accurately identified, and the audit efficiency of the crowdsourcing test report is improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of software testing, and in particular to a computer identification device, equipment, computer-readable storage medium and method for an image-based mobile crowdsourcing test report. Background technique [0002] Crowdsourcing testing, as a new testing method, has received extensive attention from the industry. By employing a large number of crowdsourced testers in a short period of time, crowdsourced testing can simulate various actual usage environments and provide feedback from real users. With its unique advantages, it is widely used in the testing of mobile applications with fast development cycles and diverse usage environments. [0003] Usually, in crowdsourcing testing, crowdsourcing testers participate in the microtasks released by the crowdsourcing initiator on the crowdsourcing platform, and the testers' feedback needs to be provided in the form of test reports. The pu...

AI Technical Summary

Problems solved by technology

However, the design of existing related technologies is based on the analysis of text similarity of test reports, and the particularity of crowdsourcing mobile testing makes it difficult to apply these technologies

Images