TL;DR
This paper introduces TSDetector, a novel approach that combines screenshot and textual analysis to accurately identify duplicate crowdtesting reports, significantly improving over existing methods.
Contribution
The paper presents TSDetector, the first approach to integrate image and text features for replicate detection in crowdtesting reports, enhancing accuracy and practical applicability.
Findings
TSDetector outperforms existing methods significantly.
Combining image and text features improves detection accuracy.
Real-world case studies confirm practical value.
Abstract
Crowdtesting is effective especially when it comes to the feedback on GUI systems, or subjective opinions about features. Despite of this, we find crowdtesting reports are highly replicated, i.e., 82% of them are replicates of others. Hence automatically detecting replicate reports could help reduce triaging efforts. Most of the existing approaches mainly adopted textual information for replicate detection, and suffered from low accuracy because of the expression gap. Our observation on real industrial crowdtesting data found that when dealing with crowdtesting reports of GUI systems, the reports would accompanied with images, i.e., the screenshots of the app. We assume the screenshot to be valuable for replicate crowdtesting report detection because it reflects the real scenario of the failure and is not affected by the variety of natural languages. In this work, we propose a…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
