Curriculum Vitae


JinYeong Bak (박진영)

jy.bak([email protected])
Ph.D. Candidate
Department of Computer Science





  • KAIST, Ph.D., Computer Science, Mar 2013 –
  • KAIST, M.S., Computer Science, Feb 2011 – Feb 2013
    Thesis : Distributed Online Learning for Topic Models [pdf], [pdf]
    GPA: 3.98/4.3
  • Sungkyunkwan University, B.S., Computer Engineering, Mar 2004 – Feb 2011
    GPA: 4.45/4.5



My research interests are in machine learning and computational social science. In particular, I study probabilistic Bayesian topic models which aims to discover latent patterns such as topics and sentiments in unstructured data.

First, I research the effective methodology of analyzing large data with high complex probabilistic Bayesian topic models. For example, I study posterior inference algorithms on distributed systems such as MapReduce. And also I study online learning which learns with one or small fraction of large data repeatedly. My M.S. thesis is suggesting new algorithm that combine two ideas for topic models. And it shows that suggesting algorithm runs faster than existing one and preserves model fit. From that, I am going to study more efficient way and apply it to more complex models.

Second, I research modeling network with Bayesian topic models. I consider network as relational data which contains not only existence but also its properties such as topics. I apply these topic models to large online social network and social media to find user’s behavior. For example, I found user’s self-disclosure behaviors in Twitter. I analyzed Twitter conversation between two users and found that users with high relationship strength show more self-disclosure behavior such as secrets. I published this results to ACL 2012. From that, I am going to study modeling other user’s behaviors and expand it to nonparametric Bayesian models which can be infered the number of parameters from data.



  • JinYeong Bak, Chin-Yew Lin, and Alice Oh. Self-disclosure topic model for classifying and analyzing twitter conversations. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing , 2014.
    [PDF] [Code and Data] [Slides]
  • JinYeong Bak, Dongwoo Kim, and Alice Oh. Distributed online learning for latent dirichlet allocation. In Proceedings of Workshop on Big Learning : Algorithms, Systems, and Tools at the Neural Information Processing Systems, 2012.
    [PDF] [Supplementary] [Code and Data]
  • JinYeong Bak, Suin Kim, and Alice Oh. Self-disclosure and relationship strength in twitter conversations. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics, 2012.
    [PDF] [Slides] [Bibtex]
  • Suin Kim, JinYeong Bak, and Alice Oh. Do you feel what i feel? social aspects of emotions in twitter conversations. In Proceedings of the AAAI International Conference on Weblogs and Social Media, 2012.
    [PDF] [Visualization] [4-Page Poster] [Poster for 4-page paper]
  • Rae Young Ko, Duk Sun Kim, Jin Yeong Bak, and Sang Gu Lee. Development of mobile sage-math and its use in linear algebra. In J. Korea Soc. Math. Ed. Ser. E: Communications of Mathematical Education, 2009.
  • Duk Sun Kim, Jin Yeong Bak, and Sang Gu Lee. The educational models using enhanced mathematics ict in the korean it environments. In J. Korea Soc. Math. Ed. Ser. E: Communications of Mathematical Education, 2008.





  • Research Intern, Microsoft Research Asia, Beijing, China, September 2013 – February 2014



  • Java, Python, C#, C++, C, R, PHP, JavaScript


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.