Sungjoon Park


sungjoon.park (at) softly.ai
CEO / co-founder
Softly.ai
Seoul, Repulic of Korea

Google Scholar
LinkedIn

Bio

I am co-founder and CEO of SoftlyAI, developing AI-based client engagement specialist agent solution for business. I received Ph.D. from School of Computing at KAIST, advised by Alice Oh. My research was supported by Google PhD Fellowship program. Before that, I obtained BA and MA in Department of Psychology in Seoul National University.


Education


Work Experience


Awards


Research Interest

My research aims to (1) develop machine learning models which can understand complicated psychological characteristics of people from text and (2) apply them to computational psychotherapy applications, in order to lead people live in better mental state. I focus on natural language since is a good indicator of latent psychological states of humans, and this is basically because our daily usage of language encodes important personality characteristics even in a single word, and further it could represent psychological states during conversation.

However, the relationship between language and mental state would be highly complex and difficult to be inferred precisely, thus learning the regularities through carefully designed machine learning models over fine-grained data are natural to be preferred than applying rule-based techniques. Furthermore, once a model can learn the relationship between them, then various psychotherapy programs would be applicable based on the detected mental state, leads to positive psychological changes. Thus my research agenda has developed to three step-by-step subproblems: (a) Learning Representation of Text, (b) Measuring Psychological Characteristics from Text, and (c) Developing Psychotherapy Applications.

Each of the problems is closely related to each other that one could evoke another question: What characteristic should be measured to develop an effective psychotherapy applications? or, what representation learning method should be needed to measure the extent of mental disorders from text? Throughout conducting research to answer these problems, I would like to bridge the gap between two fields: natural language processing and computational psychotherapy.


Publications

  1. Park, S.*, Moon, J.*, Kim, S.*, Cho, W.*, Han, J., Park, J., Song, C., Kim, J., Song, Y., Oh, T., Lee, J., Oh, J., Lyu, S., Jeong, Y., Lee, I., Seo, S., Lee, D., Kim, H., Lee, M., Jang, S., Do, S., Kim, S., Lim, K., Lee, J., Park, K., Shin, J., Kim, S., Park, L., Oh, A., Ha, J., Cho, K. (2021) KLUE: Korean Language Understanding Evaluation In NeurIPS 2021 Datasets and Benchmarks Track (Neurips 2021) *equal contribution
    Neurips Arxiv Github Leaderboard
  2. Park, S., Kim, J., Ye, S., Jeon, J., Park, H., Oh, A. (2021) Dimensional Emotion Detection in Categorical Emotion In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021)
    PDF Data
  3. Park, S.*, Park, K.*, Ahn, J. Oh, A. (2020) Suicidal Risk Detection for Military Personnel In Proceedings of the in the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020) *equal contribution
    PDF Data
  4. Park, C., Shin, J., Park, S., Lim, J., Lee, C. (2019) Fast End-to-end Coreference Resolution for Korean In Proceedings of the Findings of 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020 - Findings)
    PDF
  5. Seonwoo, Y., Park, S., Kim, D., Oh, A. (2019) Additive Compositionality of Word Vectors In Workshop on Noisy User-generated Text (W-NUT) @EMNLP 2019
    PDF
  6. Park, S., Park, H., Kim, C. (2019) A Comparison between Factor Structure and Semantic Representation of Personality Test Items Using Latent Semantic Analysis Korean Journal of Cognitive Science, 30(3), pp.133-156.
    PDF
  7. Park, S., Kim, D., Oh, A. (2019) Conversation Model Fine-Tuning for Classifying Client Utterances in Counseling Dialogues In Proceedings of the 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2019)
    PDF Poster
  8. Park, S., Seonwoo, Y., Kim, J., Kim, J., Oh, A. (2019) Denoising Recurrent Neural Networks for Classifying Crash-related Events IEEE Transactions of Intelligent Transportation Systems.
    PDF
  9. Seonwoo, Y., Park, S., Oh, A. (2018) Hierarchical Dirichlet Gaussian Marked Hawkes Process for Narrative Reconstruction in Continuous Time Domain In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018)
    PDF
  10. Park, S., Byun, J., Baek, S., Cho, Y., Oh, A. (2018) Subword-level Word Vector Representations for Korean In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018)
    PDF Poster Data
  11. Park, S., Bak, J., Oh, A. (2017) Rotated Word Vector Representations and their Interpretability. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2017)
    PDF Poster Source
  12. Kim, S., Park, S., Hale, S. A., Kim, S., Byun, J., & Oh, A. H. (2016). Understanding editing behaviors in multilingual Wikipedia. PLOS ONE, 11(5), e0155305.
    Article
  13. Kim, J., Keegan, B. C., Park, S., & Oh, A. (2016). The Proficiency-Congruency Dilemma: Virtual Team Design and Performance in Multiplayer Online Games. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI 2016)
    PDF
  14. Park, S., Kim, S., Hale, S. A., Kim, S., Byun, J., & Oh, A. (2015). Multilingual Wikipedia: Editors of Primary Language Contribute to More Complex Articles. In Ninth International AAAI Conference on Web and Social Media.(Wiki-ICWSM Workshop in ICWCM 2015)
    PDF

Talks / Presentations


Teaching Experiences


Academic Services


References


Contact

Sungjoon Park
sungjoon.park (at) softly.ai
CEO / co-founder
SoftlyAI
Seoul, Republic of Korea

(Last Update: 08/02/2024)