About me

I graduated from SHENYUAN Honors College of Beihang University(北京航空航天大学沈元荣誉学院)and am currently a second-year Ph.D. student in the School of Automation Science and Electrical Engineering, Beihang University (北京航空航天大学自动化科学与电气工程学院). My research focuses on software testing and deep learning, which is becoming increasingly urgent and necessary due to the wide application of deep learning software. If you have any ideas on this topic, please feel free to contact me. My research interests include partition testing, metamorphic testing, and their application to reinforcement learning. I have published more than 3 papers in top international journals such as TR, SQJ, which have received numerous citations.

To promote communication and collaboration among those working on reinforcement learning testing methods, I would like to found a community. If you are interested in joining us, please do not hesitate to contact me!

News

• 2021.09:  🎉🎉 An automatic health-reporting program for Beihang students has been created on Github, which supports the deployment of Tencent scf.
• 2023.01:  🥳🥳 With the pandemic over, there is no longer a need for the automatic health-reporting program, but it is still available for research purposes.

Publications

SQJ 2020

Cross-project bug type prediction based on transfer learning, Xiaoting Du, Zenghui Zhou, Beibei Yin, Guanping Xiao.

Project

• Proposed a framework of cross-project bug type prediction based on a transfer learning method, TrAdaBoost.
• Conduct experimrents on four projects Linux, MySQL, HTTPD, and AXIS which show that the framework can improve the results of cross-project bug type prediction.
• The impact factors of the prediction results were investigated, including the pair of source and target projects, and the data size of the source project.

TR 2021

DEEPSIM: Deep Semantic Information-Based Automatic Mandelbug Classification, Xiaoting Du , Zheng Zheng, Guanping Xiao, Zenghui Zhou, Kishor S. Trivedi.

Project

• Proposed DEEPSIM, an automatic Mandel- bug classification method that combines a semantic model with a deep learning classifier.
• We evaluated DEEPSIM on a total of 6557 bug reports and compared the results with those of existing studies.The experimental results showed that our method was effective for the classification of Mandel- bugs and performed better than the existing methods for this purpose.
• We investigated the impacts of the semantic model and word embedding parameters on the classification results.t.

• KBS 2024 TraceNet: Tracing and locating the key elements in sentiment analysis, Qinghua Zhao, Junfeng Liu, Zhongfeng Kang, Zenghui Zhou. Project

• ESA 2023 Is word order considered by foundation models? A comparative task-oriented analysis, Qinghua Zhao, Jiaang Li, Junfeng Liu, Zhongfeng Kang, Zenghui Zhou. Project

• MET 2021 Follow-up Test Cases are Better Than Source Test Cases in Metamorphic Testing: A Preliminary Study, Zenghui Zhou, Zheng Zheng, Tsong Yueh Chen, Jinyi Zhou,Kun Qiu. Project

Honors and Awards

• Outstanding Graduate Part-time Cadre(2023)
• Outstanding Student Cadre of Beihang University(2022, 2019)
• Outstanding Graduate Student of Beihang University(2022)
• Merit student of SASEE(2021)
• First-class Academic Excellence Scholarship(2019, 2020)
• First prize scholarship of discipline competition(2017)
• Outstanding Freshman of Beihang University(2015)

Others

Educations

• 2015.09 - 2019.06, Undergraduate, SHENYUAN Honors College of Beihang University.
• 2020.09 - 2022.03 (now), Ph.D, School of Automation Science and Electrical Engineering, Beihang University.