Short Bio

I am a PhD candidate of software engineering at Sharif University of Technology, supervised by Dr. Heydarnoori. My PhD research focuses on mining information from version control systems to automatically summarize code or generate reports for developers and software teams. I also hold a master degree in IT Engineering from SUT, supervised by Dr. Jalili (currently faculty at RMIT University, Melbourne, Australia). My research focus was on improving evaluation methods of recommender systems and resulted in various publications.


I am a PhD Candidate of software engineering at Sharif University of Technology (SUT). SUT is the top technical university in Iran where I was a graduate research assistant, teaching assistant, and the ASE lab manager.

I also hold a master degree in Information Technology (IT) Engineering from SUT, supervised by Dr. Mahdi Jalili (currently faculty at RMIT University, Melbourne, Australia). I worked at NIP lab as a graduate research assistant from 2012 to 2014. I studied recommender systems and specifically my focus was on improving evaluation methods of these algorithms. Our research resulted in 3 papers.

I was a bachelor student of IT engineering at University of Isfahan, Iran, from 2008 to 2012. As for my bachelor thesis, it was about “Automation Reading of Meter Devices and Electronic Bill Issuance”. My research was supervised by Dr. Neda Moghim.

Research Interests

My research interests include (but are not limited to):

  • Mining software repositories
  • Applied machine learning and deep learning
  • Empirical software engineering
  • Natural language processing
  • Social network analysis
  • Recommender systems

Research History and Publication

I am a software engineer with a passion for data science and applied machine learning techniques.  I am an engineer, practitioner, and researcher. I like to direct my research toward fields that are practical and are good not just on paper but will help people in real life. Therefore, I chose the mining software repositories’ field for my Ph.D. thesis to make developers and software managers’ lives easier.

Currently we are working on the vast amount of data available on Github, to find patterns to automatically generate reports and facilitate developers’ and other software team members’ tasks. Our focus is on generating automatic commit messages using deep learning techniques. As a result of our work so far, we have submitted a paper titled “Generating Summaries for Methods of Event-Driven Programs: an Android Case Study“ [1]. In this paper, we have proposed using SOTA deep learning techniques for generating explanatory comments for event-driven methods, exploiting the relationship between these methods.  This paper is under review; however, a copy is available on arXiv.

In another work, we have worked on automatically recommending tags for software repositories using both classical and state-of-the-art transformer-based multi-label classification techniques [2]. Also, We have worked on improving the quality of users’ answers in the developers Question-Answering (Q&A) website, Stack Overflow, to help other developers program more efficiently. We presented our work at Euromicro SEAA Conference [3]. Furthermore, I have worked on mining users’ reviews in mobile app stores (opinion mining) to help developers and app owners manage the huge amount of data on these stores. 

For my Msc thesis, I worked on designing novel and useful metrics for evaluation of recommender systems (RS). We published/presented three papers as the product of my master thesis in different journals and conferences. I presented our conference paper on unifying inconsistent metrics for the specialized workshop of RS evaluation in RecSys ACM Conference, Silicon Valley, 2014 [6]. In another paper, we worked on the relationship between two main evaluation metrics of RSs, namely precision vs diversity. This study was published as a book chapter [5]. Finally, due to our experience with RSs, we conducted a conclusive survey, deployed various collaborative filtering algorithms on various datasets with different contexts, and evaluated them using a wide range of metrics. We reported our findings in this paper [4].

I also have worked on other papers and researches, for more information please refer to my Google Scholar profile.

For a copy of my CV, feel free to send me an email.

[1] Aghamohammadi, A.*, Izadi, M.*, & Heydarnoori, A., Generating Summaries for Methods of Event-Driven Programs: an Android Case Study, Journal of Systems and Software (JSS), 2020, *co-first authors.

[2] Izadi, M., Ganji S., Heydarnoori, A., & Gousios G., Tag Recommendation for Software Repositories using Multi-label Multi-class Classification, (Submitted), 2020.

[3] Tavakoli M., Izadi M., & Heydarnoori A., Improving Quality of a Post's Set of Answers in Stack Overflow, Euromicro Conference on Software Engineering and Advanced Applications (SEAA), 2020.

[4] Jalili, M., Ahmadian, S., Izadi, M., Moradi, P., & Salehi, M., Evaluating Collaborative Filtering Recommender Algorithms: A Survey, IEEE Access, 6, 74003-74024. 2018.

[5] Javari, A., Izadi, M., & Jalili, M., Recommender Systems for Social Networks Analysis and Mining: Precision versus Diversity, In Complex Systems and Networks (pp. 423-438). Springer, Berlin, Heidelberg, 2016.

[6] Izadi, M., Javari, A., & Jalili, M., Unifying Inconsistent Evaluation Metrics in Recommender Systems, In Proceedings of ACM RecSys Conference, REDD Workshop, Silicon Valley, USA (pp. 1-7), 2014.

[7] Izadi, M., Izadi, M., & Azarsa, B., The Intonation Patterns of English and Persian Sentences: A Contrastive Study, Research Journal of Education, 3(9), 97-101, 2017.