Yuto Yamaguchi, Ph.D. (Engineering)
- Software Engineering Manager @ Indeed

Research: published 10+ papers in top-tier conferences and got 600+ citations. Worked on Data Mining, Social Network Analysis, Machine Learning, Graph Data Machine Learning.

Software Engineering: Built a large scale recommendation product from scratch, launched it in the US, and grew it to one of the most important products in the company. Work involved pretty much everything: frontend, backend, data engineering, machine learning modeling, etc.

Management: Led a talented team of 10+ software engineers and data scientists. Created the foundation of team's process, laid out the product roadmap, made lots of business impact, and grew people in the team. Now enjoying launching a new team from scratch.

Summary

• Working as an ML engineer / manager at a global scale company for 5+ years.
• Worked as an ML researcher for 4+ years (incl 1 year stay at CMU).
• Published 10+ papers in top-tier conferences.
• Got 600+ citations to my papers.
• Worked as a program comittee member at AAAI conference 2016-2017.

Work Experience (Since 2014)

• Software Engineering Manager @ Indeed, from Jan. 2020 to present
  

  Leading a team of engineers and data scientists developing a system for recommending the best candidates to a given job opening. The team is mainly responsible for building a data / ML pipeline, maintaining and improving ML models on top of it.

• Software Engineer @ Indeed, from Apr. 2017 to Dec. 2019
  

  Mainly working on Machine Learning, developing Data aggregation pipeline, ML model training pipeline, Recommendation engine, etc.

• Tenure-Track Researcher @ AIST, from Apr. 2016 to Mar. 2017
  

  Mainly worked on tensor factorization and network generative models (published 2 IJCAI papers).

• Postdoc Researcher @ U. Tsukuba, from Apr. 2015 to Mar. 2016
  

  Worked with students mainly in the area of social media data mining/analysis.
  

• Visiting Researcher @ CMU, from Apr. 2014 to Mar. 2015
  

  Mainly worked on label propagation algorithms in various settings (published 5 papers in one year).
  Supervisor: Prof. Christos Faloutsos

Education

• Ph.D. in Engineering @ U. Tsukuba, Mar. 2014
  Supervisor: Prof. Hiroyuki Kitagawa
• Master of Engineering @ U. Tsukuba, Mar. 2012
  Supervisor: Prof. Hiroyuki Kitagawa
• Bachelor of Engineering @ U. Tsukuba, Mar. 2010
  Supervisor: Prof. Hiroyuki Kitagawa

Selected Publication (Since 2014)

Full list -> DBLP, Google Scholar

1. FORank: Fast ObjectRank for Large Heterogeneous Graphs
   Tomoki Sato, Hiroaki Shiokawa, Yuto Yamaguchi, Hiroyuki Kitagawa
   The 27th International Conference on World Wide Web (WWW2018), Lyon, France, Apr 23-27, 2018 (poster).
   
2. Collecting Non-Geotagged Local Tweets via Bandit Algorithms
   Saki Ueda, Yuto Yamaguchi, Hiroyuki Kitagawa
   The 26th ACM International Conference on Information and Knowledge Management (CIKM2017), Singapore, Nov 6-10, 2017 (short).
   
3. Tensor Decomposition with Missing Indices
   Yuto Yamaguchi, Kohei Hayashi
   International Joint Conference on Artificial Intelligence (IJCAI2017), Melbourne, Australia, Aug 19-25, 2017 (26%).
   [PDF] [Slides] [Code]
4. When Does Label Propagation Fail? A View from a Network Generative Model
   Yuto Yamaguchi, Kohei Hayashi
   International Joint Conference on Artificial Intelligence (IJCAI2017), Melbourne, Australia, Aug 19-25, 2017 (26%).
   [PDF] [Slides] [Code]
5. Modeling Temporal Activity to Detect Anomalous Behavior in Social Media
   Alceu Ferraz Costa, Yuto Yamaguchi, Agma Juci Machado Traina, Caetano Traina Jr., and Christos Faloutsos
   ACM Transactions on Knowledge Discovery from Data (TKDD), 11(4), 49.
   
6. CAMLP: Confidence-Aware Modulated Label Propagation
   Yuto Yamaguchi, Christos Faloutsos, Hiroyuki Kitagawa
   SIAM International Conference on Data Mining (SDM2016), pp. 513-521, Miami, Florida, USA, May 5-7, 2016 (26%).
   [Code]
7. RSC: Mining and Modeling Temporal Activity in Social Media
   Alceu Ferraz Costa, Yuto Yamaguchi, Agma Juci Machado Traina, Caetano Traina Jr., and Christos Faloutsos
   The 21st SIGKDD Conference on Knowledge Discovery and Data Mining (KDD2015), pp. 269-278, Sydney, Australia, August 10-13, 2015 (19%).
   [Code and Data]
8. Patterns in Interactive Tagging Networks
   Yuto Yamaguchi, Mitsuo Yoshida, Christos Faloutsos, and Hiroyuki Kitagawa
   The 9th International AAAI Conference on Web and Social Media (ICWSM2015), pp. 513-522, Oxford, UK, May 26-29, 2015 (19%).
   [Code] [Data]
9. Why Do You Follow Him? Multilinear Analysis on Twitter
   Yuto Yamaguchi, Mitsuo Yoshida, Christos Faloutsos, and Hiroyuki Kitagawa
   The 24th International World Wide Web Conference (WWW2015), pp. 137-138 (companion volume), Florence, Italy, May 18-22, 2015 (poster) (30%).
   [Code] [Data]
10. SocNL: Bayesian Label Propagation with Confidence
    Yuto Yamaguchi, Christos Faloutsos, and Hiroyuki Kitagawa
    The 19th Pacific-Asia Conference on Knowledge Discovery and Data mining (PAKDD 2015), pp. 633-645, Ho Chi Min City, Viet Nam, May 19-22, 2015 (6.7% as regular paper).
    [Code]
11. OMNI-Prop: Seamless Node Classification on Arbitrary Label Correlation
    Yuto Yamaguchi, Christos Faloutsos, and Hiroyuki Kitagawa
    The Twenty-Ninth AAAI Conference on Artificial Intelligence (AAAI 2015), pp. 3122-3128, Austin, USA, January 25-29, 2015 (27%).
    [Code]
12. Online User Location Inference Exploiting Spatiotemporal Correlations in Social Streams
    Yuto Yamaguchi, Toshiyuki Amagasa, Hiroyuki Kitagawa, and Yohei Ikawa
    The 23rd ACM International Conference on Information and Knowledge Management (CIKM 2014), pp. 1139-1148, Shanghai, China, November 3-7, 2014 (21%).
    [Code] [Data]

Patents

1. METHOD, COMPUTER PROGRAM AND COMPUTER FOR ESTIMATING LOCATION BASED ON SOCIAL MEDIA
   Miki Enoki, Yohei Ikawa, Michiaki Tatsubori, and Yuto Yamaguchi
   US Patent App. 14/067,178
2. SYSTEM AND METHOD FOR EXTRACTION OF OFF-TOPIC PART FROM CONVERSATION
   Nobuyasu Itoh, Masafumi Nishimura, and Yuto Yamaguchi
   US Patent App. 13/740,478

Professional Activities

• Members: AAAI (2016), DBSJ (2010-)
• PC member: AAAI (2016, 2017)
• Ad-hoc reviewer: ACM TWEB (2015), WWW (2016), DASFAA (2016)

Funding

• Grant-in-Aid for JSPS Fellows, from Apr. 2014 to Mar. 2015
  Japan Society for the Promotion of Science, Japan
  approx. 5,200,000 yen / year.
• Grant-in-Aid for JSPS Fellows, from Apr. 2012 to Mar. 2014
Japan Society for the Promotion of Science, Japan
approx. 3,300,000 yen / year.
• Student Scholarship, from Apr. 2010 to Mar. 2012
Japan Student Services Organization, Japan
approx. 1,050,000 yen / year.

Talks

• User Profiling in Social Media: Locations and More,
  ERATO Kawarabayashi Large Graph Project -- Graph Mining & Web & AI Seminar, May. 16, 2016.
• Node Classification: Homophily or Heterophily,
  IBM Research -- Tokyo, Apr. 20, 2016.
• Online User Location Inference Exploiting Spatiotemporal Correlations in Social Streams,
  Carnegie Mellon University -- DB Seminar, Oct. 27, 2014.
• A Study on User Location Inference in Social Media,
  IBM Research -- Tokyo, Mar. 24, 2014.

Honors and Awards

• Research Fellowship for Young Scientists, from Apr. 2012 to Mar. 2015
  Japan Society for the Promotion of Science, Japan
• Outstanding Student Award from the Provost of SIE (Top 1.5%), Mar. 2012
  University of Tsukuba, Japan