Jun Chen


About Me

I am currently a senior software engineer at Baidu Inc. I got my Ph.D degree and Bachelor of Engineering degree both from School of Software, Tsinghua University in 2017 and 2012, respectively.

Contact

Email: chenjun14 at tsinghua.org.cn, chenjun082 at gmail.com .

Research Interest

My principle research interests generally lie in machine learning and data mining from large scale datasets with the focus on personalization, recommender systems and learning from user behaviors. Besides, I also have broad interests in deep learning, natural language processing, multimedia, information retrieval as well as social networks.

Education

  1. Doctor of Philosophy, Software Engineering.
    Supervised by Prof. Jianmin Wang and Associate Prof. Chaokun Wang
    School of Software, Tsinghua University.
    Sept 2012 - July 2017.

  2. Bachelor of Science, Computer Software.
    School of Software, Tsinghua University.
    Sept 2008 - July 2012.

Working/Teaching Experience

  1. Senior software engineer, full-time, Baidu Inc.

    Jul, 2017 - Now. We dedicate ourselves in advancing the AI technology for big data problems. Currently, I work on a project of automatic question answering (QA) system for healthcare. Previously, I worked on a system for online self-examination of mental disease.

  2. Research Assistant, Tsinghua University.

    Jan, 2016 - Jul, 2017. I worked with Prof. Chaokun Wang on the research of pesonalized recommendation and social networks.

  3. Teaching Assistant for Digital Media (II): Multimedia (44100415), Tsinghua University.

    Sep, 2014 - Jan, 2016. I designed and evaluated the homework and the project (e.g. music resizing challange) for this course.

  4. Teaching Assistant for Architecture of Computer and Network (II) (34100344), Tsinghua University.

    Feb, 2014 - Jul, 2015. I designed and evaluated the homework and the project (e.g. the project of assembly language application and that of languge-to- language translation for compilers) for this course.

  5. Teaching Assistant for An Introduction to Modern Database System (84100162), Tsinghua University.

    Sep, 2013 - Jan, 2015. I designed and evaluated the homework and the project (e.g. similarity join of complex data, query engine for high dimensional data) for this course.

  6. Research Assistant, Tsinghua University.

    Sep, 2012 - Sep, 2013. I worked with Prof. Chaokun Wang on the research of the estimation of music stretching resistance.

  7. Software development engineer, summer intern, Amazon Inc.

    Jul, 2012 - Aug, 2012. I developed a system to enable the book merchants on Amazon.cn to easily manage the stocks of their selling books.

Publication

Journal Papers

  1. Understanding Item Consumption Orders for Right-Order Next-Item Recommendation. [Preprint]
    Jun Chen, Xuecheng Wang, Chaokun Wang.
    Knowledge and Information Systems (KAIS). In press, 2018.

  2. Learning the Personalized Intransitive Preferences of Images. [Link][Data][Codes]
    Jun Chen, Chaokun Wang, Jianmin Wang, Xiang Ying, Xuecheng Wang.
    IEEE Transactions on Image Processing (IEEE TIP). Volume:26, Issue:9, 2017.

  3. Recommendation for Repeat Consumption from User Implicit Feedback. [Link][Codes]
    Jun Chen, Chaokun Wang, Jianmin Wang, Philip S. Yu.
    IEEE Transactions on Knowledge and Data Engineering (IEEE TKDE). Volume:28, Issue:11, 2016.

  4. Inferring Directions of Undirected Social Ties. [Link]
    Jun Zhang, Chaokun Wang, Jianmin Wang, Jeffery X. Yu, Jun Chen, Changping Wang.
    IEEE Transactions on Knowledge and Data Engineering (IEEE TKDE). Volume:28, Issue:12, 2016.

  5. Automatic Music Stretching Resistance Classification Using Audio Features and Genres. [Link][Data]
    Jun Chen, Chaokun Wang.
    IEEE Signal Processing Letters, Volume:20, Number:12, 2013.

  6. Who is answering whom? Finding “Reply-To” relations in group chats with deep bidirectional LSTM networks. [Link]
    Gaoyang Guo, Chaokun Wang, Jun Chen, Pengcheng Ge, Weijun Chen.
    Cluster Computing, preprint, 2018.

  7. On Generalized Bisimilarity Join. [Link]
    Changping Wang, Chaokun Wang, Hao Wang, Meng Wang, Jun Chen.
    Chinese Journal of Software, 2017.

  8. Algorithms for Local Betweenness Centrality of Fully Dynamic Multi-Dimensional Networks. [Link]
    Jianxiang Yang, Chaokun Wang, Meng Wang, Jun Chen.
    Chinese Journal of Computers, Volume:38, Number:9, 2015.

  9. Investigating the role of musical genre in human perception of music stretching resistance. [Link][Data]
    Jun Chen, Chaokun Wang.
    CoRR, 2017.

Conference Papers

  1. Modeling the Intransitive Pairwise Image Preference from Multiple Angles. [Link][AudioSlides][Poster]
    Jun Chen, Chaokun Wang, Jianmin Wang.
    ACM International Conference on Multimedia (MM'17), 2017, Mountain View, California, USA.

  2. Who Is Answering to Whom? Finding “Reply-To” Relations in Group Chats with Long Short-Term Memory Networks. (Best Paper Award) [Link]
    Gaoyang Guo, Chaokun Wang, Jun Chen, and Pengcheng Ge.
    International Conference on Emerging Databases (EDB'17), August 2017, Busan, Korea.

  3. Recommendation for Repeat Consumption from User Implicit Feedback (Extended Abstract). [Link][Codes][Poster]
    Jun Chen, Chaokun Wang, Jianmin Wang, Philip S Yu.
    IEEE International Conference on Database Engineering (ICDE'17), 2017, San Diego, California, USA.

  4. Learning the Structures of Online Asynchronous Conversations. [Link][Slides][Data]
    Jun Chen, Chaokun Wang, Heran Lin, Weiping Wang, Zhipeng Cai, Jianmin Wang.
    Proceedings of the 22nd International Conference on Database Systems for Advanced Applications (DASFAA'17), 2017, Suzhou, China.

  5. Preference Join on Heterogeneous Data. [Link]
    Changping Wang, Chaokun Wang, Hao Wang, Jun Chen, Xiaojun Ye.
    Proceedings of the 18th Asia Pacific Web Conference (APWeb'16), 2016, Suzhou, China.

  6. A Personalized Interest-Forgetting Markov Model for Recommendations. (oral presentation, acceptance rate 11.75%) [Link][Slides]
    Jun Chen, Chaokun Wang, Jianmin Wang.
    Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI'15), 2015, Austin, TX, USA.

  7. Will You "Reconsume" the Near Past? Fast Prediction on Short-term Reconsumption Behaviors. (acceptance rate 26.67%) [Link][Poster][Data]
    Jun Chen, Chaokun Wang, Jianmin Wang.
    Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI'15), 2015, Austin, TX, USA.

  8. CrowdMR: Integrating Crowdsourcing with MapReduce for AI-hard Problems. [Link][Poster][Demo]
    Jun Chen, Chaokun Wang, Yiyuan Bai.
    Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI'15), 2015, Austin, TX, USA.

  9. Modeling the Interest-Forgetting Curve for Music Recommendation. [Link][Poster]
    Jun Chen, Chaokun Wang, Jianmin Wang.
    Proceedings of the 22nd ACM International Conference on Multimedia (MM'14), 2014, Orlando, FL, USA.

  10. MiSCon: A Hot Plugging Tool for Real-time Motion-based System Control. [Link][Poster][Video]
    Jun Chen, Chaokun Wang, Lei Yang, Qingfu Wen and Xu Wang.
    Proceedings of the 22nd ACM International Conference on Multimedia (MM'14), 2014, Orlando, FL, USA.

  11. CoDEM: An Ingenious Tool of Insight into Community Detection in Social Networks. [Link]
    Meng Wang, Chaokun Wang, Jun Chen.
    Proceedings of the 23rd ACM International Conference on Information and Knowledge Management (CIKM'14), 2014, Shanghai, China.

  12. RESIC: A Tool for Music Stretching Resistance Estimation. [Link][Poster][Data]
    Jun Chen, Chaokun Wang.
    Proceedings of the 20th Anniversary International Conference on MultiMedia Modeling (MMM'14), 2014, Dublin, Ireland.

Datasets

  1. Douban Group Conversations: A corpus of online chats on the Chinese web forum Douban Group.
    *Published with my DASFAA'17 paper.

  2. Holidays Image Preference: A dataset of pairwise preference on some images from INRIA Holidays.
    *Published with my IEEE TIP paper.

  3. ManicTime PC APP Usage: The logs of APP usage on personal computers.
    *Published with my AAAI'15 paper.

  4. Music Stretching Resistance: The maximum elongating rate and the minimum compressing rate of 894 songs, the title and genre of the songs as well as the name of the artists.
    *Published with my IEEE Singal Processing Letters 2013 paper.

Honors and Awards

  1. Outstanding Ph.D graduate of Tsinghua University, 2017.

  2. Outstanding Ph.D dissertation of Tsinghua University, 2017.

  3. Outstanding Ph.D graduate of Beijing city, 2017.

  4. Huayu Principle Scholarship, Tsinghua University, 2016.

  5. Graduate National Scholarship of China, Ministry of Education, 2015.

  6. Graduate National Scholarship of China, Ministry of Education, 2014.

  7. Sashixuan Best Student Paper Award, China Computer Federation, 2014.

  8. RIM Scholarship, Tsinghua University, 2013.

  9. Outstanding academic undergraduate thesis, Tsinghua University, 2012.

  10. Huangyicong Scholarship, Tsinghua University, 2010.

  11. AEON Scholarship, Tsinghua University, 2009.

Research Service