I am a Senior Researcher at Data Mining Technology Group, Data Science Research Laboratories, NEC Corporation. I received my Ph.D. at University of Tsukuba in March 2018 under supervision of Prof. Hiroyuki Kitagawa.

I love coding. Several spin-off stuffs are available on GitHub.

Feel free to drop me an E-mail (stillpedant _at_ gmail.com). I am always welcoming discussions and opportunities for potential collaborations.

Research Interests

My interests are on transforming data into values without requiring humans' hard efforts, experts' domain knowledges, and huge computational resources. Currently, I am actively working on the following topics.

  • Machine Learning
    • Latent Factor Models
    • Bayesian Models (Regularized Bayes, Bayesian Nonparametrics, ...)
    • Machine Learning on Compressed Data
  • Recommender Systems
    • Contextual Recommender Systems
  • Database Systems
  • Information Extraction
    • Relation Extraction (From Unstructured Data to Structured Data)

Publications

  1. Masafumi Oyamada,
    Efficient Machine Learning on Relational Data,
    PhD Thesis, University of Tsukuba, 2018.
    [PDF]
  2. Masafumi Oyamada,
    Adaptive Partial Aggregation Tree,
    The 2nd. cross-disciplinary Workshop on Computing Systems, Infrastructures, and Programming (xSIG 2018) (to appear), 2018.
  3. Masafumi Oyamada, Jianquan Liu, Shinji Ito, Kazuyo Narita, Takuya Araki, Hiroyuki Kitagawa,
    Compressed Vector Set: A Fast and Space-Efficient Data Mining Framework,
    Journal of Information Processing (JIP) vol.26, pp. 416-426, 2018.
    [PDF] [Link]
  4. Masafumi Oyamada and Shinji Nakadai,
    Relational Mixture of Experts: Explainable Demographics Prediction with Behavioral Data,
    IEEE International Conference on Data Mining (ICDM 2017), pp. 357-366, 2017.
    [PDF] (Regular paper, Acceptance rate: 72/778 = 9.25%)
  5. Katsufumi Tomobe, Masafumi Oyamada, and Shinji Nakadai,
    Link Prediction for Isolated Nodes in Heterogeneous Network by Topic-based Co-Clustering,
    21st Pacific Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2017), pp. 147-159, 2017.
  6. Masafumi Oyamada, Jianquan Liu, Kazuyo Narita, and Takuya Araki.
    MOARLE: Matrix Operation Accelerator based on Run-Length Encoding,
    16th Asia-Pacific Web Conference (APWeb 2014), pages 425-436, 2014 (Regular paper, Acceptance rate: 34/134 = 25%) (Best paper runner up).
  7. Masafumi Oyamada , Hideyuki Kawashima, and Hiroyuki Kitagawa,
    Data Stream Processing with Concurrency Control,
    ACM SIGAPP Applied Computing Review (ACM SIGAPP ACR), Vol. 13, No. 2, pp. 54-65, June 2013. [URL]
  8. Masafumi Oyamada , Hideyuki Kawashima, and Hiroyuki Kitagawa,
    Continuous Query Processing with Concurrency Control: Reading Updatable Resources Consistently,
    Proc. 28th ACM Symposium on Applied Computing (SAC 2013), Coimbra, Portugal, pp. 788-794, March 18-22, 2013.
    [PDF] [DBLP] (Acceptance rate: 23.98%)
  9. Masafumi Oyamada , Hideyuki Kawashima, and Hiroyuki Kitagawa,
    Efficient Invocation of Transaction Sequences Triggered by Data Streams"
    The 2nd International Workshop on Streaming Media Delivery and Management Systems (SMDMS 2011), Proceedings of 3PGCIC, Barcelona, Spain, pp. 332-337, October 26-28, 2011.
    [PDF] [DBLP]
  10. Masafumi Oyamada , Hideyuki Kawashima, and Hiroyuki Kitagawa,
    Integration of Data Streams and Relations with Main Memory Database,
    Proc. 8th International Conference on Networked Sensing Systems (INSS 2011), Penghu, Taiwan, June 13-15, 2011.

Honors

  • JSAI Annual Conference Award, 2016
  • Best honorable poster in DEIM, 2015
  • Best paper runner up award in APWeb, 2015
  • Honorable poster in DEIM, 2014
  • Chair of department of computer science award from University of Tsukuba, 2013
  • Honorable student talk in DEIM, 2013

Services

  • JSAI (2018, 2017 PC, 2016 Session Chair)
  • DEIM (2018, 2017 Commentator, 2016 Local Committee)
  • Reviewer: DBSJ Journal (2017), IEICE Transactions (2014, 2016), ADMS (2014, External)

Misc

currentia
A Data Stream Management System (DSMS) with concurrency control
Jothelloc
An othello game written in JavaScript and C++
Letter counter
A crappy letter counter
Real Time Web Analytics