Research

Client-side Information Retrieval

In this research project, we develop technologies that employ client-side AI running on researchers' PCs to continuously monitor research activities and information-seeking behaviors, autonomously retrieving and providing information necessary for advancing research. This approach reduces privacy and security risks while enabling the timely delivery of information that researchers intend to search for, information they did not explicitly intend to search for but is still useful, and information that contributes to the progress of their research.

Related Publications

1. SIGIR 2026 Koji Nishikawa, Makoto P. Kato: H-MAPS: Hierarchical Memory-Augmented Proactive Search Assistant for Scientific Literature. To appear in Proc. of ACM SIGIR 2026.

Low-Resource Information Retrieval

This project develops technologies for building information retrieval systems in low-resource settings, based on generalist retrieval models that can be adapted to new domains and tasks without large amounts of training data.

Related Publications

1. SIGIR 2025 Kenya Abe, Kunihiro Takeoka, Makoto P. Kato, Masafumi Oyamada: LLM-based Query Expansion Fails for Unfamiliar and Ambiguous Queries. In Proc. of ACM SIGIR 2025, pp. 3035–3039.
2. SIGIR-AP 2025 Yuto Nakachi, Makoto P. Kato: Impact of LLM-Modified Queries and Documents in Training Data on Neural Retrieval Models. In Proc. of ACM SIGIR-AP 2025, pp. 364–373.
3. ECIR 2025 Haruki Fujimaki, Makoto P. Kato: Investigating the Performance of Dense Retrievers for Queries with Numerical Conditions. In Proc. of ECIR 2025, pp. 210–218.
4. IEICE 2025 Huu-Long Pham, Ryota Mibayashi, Takehiro Yamamoto, Makoto P. Kato, Yusuke Yamamoto, Yoshiyuki Shoji, Hiroaki Ohshima: Pre-trained BERT Model Retrieval: Inference-Based No-Learning Approach using k-Nearest Neighbour Algorithm. IEICE Trans. Inf. Syst. 108(8), pp. 872–882, 2025.
5. WISE 2024 Takumi Ito, Atsuki Maruta, Makoto P. Kato, Sumio Fujita: PR-Rank: A Parameter Regression Approach for Learning-to-Rank Model Adaptation Without Target Domain Data. In Proc. of WISE 2024, pp. 3–18.
6. CIKM 2024 Kota Usuha, Makoto P. Kato, Sumio Fujita: Over-penalization for Extra Information in Neural IR Models. In Proc. of ACM CIKM 2024, pp. 4096–4100.

Information Retrieval Evaluation

Through the NTCIR evaluation campaigns and theoretical studies, we develop evaluation methodologies for information access systems, including online evaluation based on interleaving and multileaving. We have organized NTCIR tasks such as INTENT, 1CLICK, MobileClick, OpenLiveQ, and Data Search.

Related Publications

1. ECIR 2023 Kojiro Iizuka, Hajime Morita, Makoto P. Kato: Theoretical Analysis on the Efficiency of Interleaved Comparisons. In Proc. of ECIR 2023, pp. 459–473.
2. ICTIR 2021 Kojiro Iizuka, Yoshifumi Seki, Makoto P. Kato: Decomposition and Interleaving for Variance Reduction of Post-click Metrics. In Proc. of ACM ICTIR 2021, pp. 221–230.
3. CIKM 2020 Makoto P. Kato, Akiomi Nishida, Tomohiro Manabe, Sumio Fujita, Takehiro Yamamoto: What Rankers Can be Statistically Distinguished in Multileaved Comparisons? In Proc. of ACM CIKM 2020, pp. 2081–2084.
4. CIKM 2018 Makoto P. Kato, Tomohiro Manabe, Sumio Fujita, Akiomi Nishida, Takehiro Yamamoto: Challenges of Multileaved Comparison in Practice: Lessons from NTCIR-13. In Proc. of ACM CIKM 2018, pp. 1515–1518.
5. SIGIR 2017 Tomohiro Manabe, Akiomi Nishida, Makoto P. Kato, Takehiro Yamamoto, Sumio Fujita: A Comparative Live Evaluation of Multileaving Methods on a Commercial cQA Search. In Proc. of ACM SIGIR 2017, pp. 949–952.

Related Resources

1. interleaving : a Python library for interleaving-based online evaluation
2. openliveq and mobileclick-eval : evaluation tools for NTCIR tasks

Data Search

This line aims to develop a data search engine for organizing the world's open data and making it universally accessible and useful, providing flexible access to open data through an index built by data analysis. It also includes evidence-based information retrieval — matching natural-language texts with data and knowledge bases to verify the credibility of information and to rank entities by their properties — supporting open science and data-driven decision making.

Related Publications

1. HCII 2025 Wiradee Imrattanatrai, Makoto P. Kato, Ken Fukuda: Enhancing User Understanding of Entity Relationships with Knowledge Graphs: A Dataset for Multi-entity Relationship Explanation. In Proc. of HCI International 2025, pp. 197–205.
2. DSE 2022 Atsuki Maruta, Makoto P. Kato: Intent-Aware Data Visualization Recommendation. Data Science and Engineering 7(4), pp. 301–315, 2022.
3. NTCIR-16 Makoto P. Kato, Hiroaki Ohshima, Ying-Hsang Liu, Hsin-Liang Chen, Yu Nakano: Overview of the NTCIR-16 Data Search 2 Task. In Proc. of NTCIR 2022.
4. SIGIR 2021 Makoto P. Kato, Hiroaki Ohshima, Ying-Hsang Liu, Hsin-liang Oliver Chen: A Test Collection for Ad-hoc Dataset Retrieval. In Proc. of ACM SIGIR 2021, pp. 2450–2456.
5. WISE 2021 Atsuki Maruta, Makoto P. Kato: Intent-Aware Visualization Recommendation for Tabular Data. In Proc. of WISE 2021, pp. 252–266.
6. ASIS&T 2021 Ying-Hsang Liu, Hsin-liang Oliver Chen, Makoto P. Kato, Mingfang Wu, Kathleen Gregory: Data Discovery and Reuse in Data Service Practices: A Global Perspective. In Proc. of ASIS&T Annual Meeting 2021, pp. 610–612.
7. ASIS&T 2020 Ying-Hsang Liu, Hsin-Liang Chen, Makoto P. Kato, Mingfang Wu, Isto Huvila: Supporting open research data practice through data curation and discovery: A global perspective. In Proc. of ASIS&T Annual Meeting 2020.
8. ECIR 2020 Makoto P. Kato, Wiradee Imrattanatrai, Takehiro Yamamoto, Hiroaki Ohshima, Katsumi Tanaka: Context-Guided Learning to Rank Entities. In Proc. of ECIR 2020, pp. 83–96.
9. SIGIR 2019 Wiradee Imrattanatrai, Makoto P. Kato, Masatoshi Yoshikawa: Identifying Entity Properties from Text with Zero-shot Learning. In Proc. of ACM SIGIR 2019, pp. 195–204.
10. IEICE 2018 Wiradee Imrattanatrai, Makoto P. Kato, Katsumi Tanaka, Masatoshi Yoshikawa: Entity Ranking for Queries with Modifiers Based on Knowledge Bases and Web Search Results. IEICE Trans. Inf. Syst. 101-D(9), pp. 2279–2290, 2018.
11. WI 2017 Wiradee Imrattanatrai, Makoto P. Kato, Katsumi Tanaka: Entity search by leveraging attributive terms in sentential queries over RDF data. In Proc. of IEEE/WIC/ACM WI 2017, pp. 769–776.

Related Resources

1. NTCIR-15 Data Search and NTCIR-16 Data Search 2 tasks (task organizer): shared evaluation infrastructure for ad-hoc dataset retrieval
2. Keynote ROCLING 2022: Matching Texts with Data for Evidence-based Information Retrieval

Information Need Elicitation & Search Behavior

We studied how to elicit information needs from users — using sensors, brain signals, and search behaviors such as clicks to estimate desired information actively and effectively — together with broader analyses of user search behavior, from query suggestion to online purchase behavior.

Related Publications

1. WebSci 2023 Yuki Yanagida, Makoto P. Kato, Yuka Kawada, Takehiro Yamamoto, Hiroaki Ohshima, Sumio Fujita: What Web Search Behaviors Lead to Online Purchase Satisfaction? In Proc. of ACM WebSci 2023, pp. 324–334.
2. WSDM 2016 Makoto P. Kato, Katsumi Tanaka: To Suggest, or Not to Suggest for Queries with Diverse Intents: Optimizing Search Result Presentation. In Proc. of ACM WSDM 2016, pp. 133–142.
3. SIGIR 2014 Makoto P. Kato, Takehiro Yamamoto, Hiroaki Ohshima, Katsumi Tanaka: Investigating users’ query formulations for cognitive search intents. In Proc. of ACM SIGIR 2014, pp. 577–586.
4. CIKM 2014 Shuya Ochiai, Makoto P. Kato, Katsumi Tanaka: Re-call and Re-cognition in Episode Re-retrieval: A User Study on News Re-finding a Fortnight Later. In Proc. of ACM CIKM 2014, pp. 579–588.
5. WSDM 2013 Shinya Tanaka, Adam Jatowt, Makoto P. Kato, Katsumi Tanaka: Estimating content concreteness for finding comprehensible documents. In Proc. of ACM WSDM 2013, pp. 475–484.
6. IR Journal 2013 Makoto P. Kato, Tetsuya Sakai, Katsumi Tanaka: When do people use query suggestion? A query suggestion log analysis. Information Retrieval 16(6), pp. 725–746, 2013.
7. CHI 2013 Makoto P. Kato, Ryen W. White, Jaime Teevan, Susan T. Dumais: Clarifications and question specificity in synchronous social Q&A. In Proc. of ACM CHI 2013, pp. 913–918.
8. WWW 2012 Makoto P. Kato, Tetsuya Sakai, Katsumi Tanaka: Structured query suggestion for specialization and parallel movement: effect on search behaviors. In Proc. of The Web Conference (WWW) 2012, pp. 389–398.

Related Resources

1. Cognitive Relevance Dataset and search behavior data used in these studies

Analogy-based Information Retrieval

We studied search by examples and analogies, which retrieves information in unfamiliar domains using what users already know: Query by Analogical Example, a relational search method using Web search engine indices; a geographic search that lets users search unfamiliar places as if they were in their hometown; RhythMiXearch, which finds unknown music by mixing known music; and a Web image search method for abstract query terms based on social tags.

Related Publications

1. SIGIR 2012 Makoto P. Kato, Hiroaki Ohshima, Katsumi Tanaka: Content-based retrieval for heterogeneous domains: domain adaptation by relative aggregation points. In Proc. of ACM SIGIR 2012, pp. 811–820.
2. ICME 2012 Yuki Sugiyama, Makoto P. Kato, Hiroaki Ohshima, Katsumi Tanaka: Relative Relevance Feedback in Image Retrieval. In Proc. of IEEE ICME 2012, pp. 272–277.
3. CIKM 2010 Makoto P. Kato, Hiroaki Ohshima, Satoshi Oyama, Katsumi Tanaka: Search as if you were in your home town: geographic search by regional context and dynamic feature-space selection. In Proc. of ACM CIKM 2010, pp. 1541–1544.
4. ICUIMC 2010 Makoto P. Kato, Satoshi Oyama, Hiroaki Ohshima, Katsumi Tanaka: Query by example for geographic entity search with implicit negative feedback. In Proc. of ICUIMC 2010, p. 45.
5. CIKM 2009 Makoto P. Kato, Hiroaki Ohshima, Satoshi Oyama, Katsumi Tanaka: Query by analogical example: relational search using web search engine indices. In Proc. of ACM CIKM 2009, pp. 27–36.
6. ISMIR 2009 Makoto P. Kato: RhythMiXearch: Searching for Unknown Music by Mixing Known Music. In Proc. of ISMIR 2009, pp. 477–482.
7. WISE 2008 Makoto P. Kato, Hiroaki Ohshima, Satoshi Oyama, Katsumi Tanaka: Can Social Tagging Improve Web Image Search? In Proc. of WISE 2008, pp. 235–249. (Kambayashi Best Paper Award)

Related Resources

1. QAE and GORD datasets created in these studies