Knowledge and Language Computing Laboratory
KaLC Lab, since April 2025, is located in the Institute of Library, Information and Media Science at the University of Tsukuba, Japan.
Latest News RSS
- 2026-01-01: Two New Staff Members Join KaLC Lab
- 2025-09-19: Assoc Prof Yada (PI) presented a study "Attempt to extract emotion-cause events from patient narrative blogs by using LLM" (in Japanese) at a Japan domestic NLP conference, YANS 2025.
- 2025-07-26: Assoc Prof Yada (PI) has been selected as a JST FOREST researcher. We will begin recruiting staff for our lab soon — details will be posted on this website.
- 2025-07-23: Two new master's students joined our lab.
- 2025-06-09: The laboratory website has launched.
Aims
Computationally Modelling Human Knowledge and Language
Knowledge and Language Computing Laboratory (KaLC Lab) aims at developing computational models of human knowledge and language and apply them to various real-world problems.
Target: Domain Specific Knowledge
Human knowledge spans across various domains such as medicine, law, and art. We aim to model this knowledge into more than a graph.
Method: Natural Language Processing (NLP)
Fully making use of the cutting-edge NLP techniques, we process a wide range of linguistic data that encode human knowledge.
Data: Multi-modal Text
Linguistic data can be found in various forms such as text, speech, and images. Also, the unit of analysis can be a word, sentence, or document. We handle them all.
Outcome: Real-world Applications
We apply our research to real-world problems and develop human-oriented applications that benefit society.
Research
Research Projects
We are working on various research projects related to knowledge and language computing.
Develop a dictionary construction system that integrates knowledge across all academic fields.
Extract the cause of patient emotions from patient narrative.
Enhance collaboration between school librarians and teachers.
PI Message
Message from the PI
Shuntaro Yada, PhD
Thank you for e-visiting the Knowledge and Language Computing Laboratory (KaLC Lab) at the University of Tsukuba. I am Shuntaro Yada, the Principal Investigator (PI) of the KaLC Lab. I warmly welcome students interested in tackling real-world problems. This challenge always involves understanding human knowledge, and such professional knowledge is often encoded in language. Making use of the latest Natural Language Processing (NLP) techniques, we aim to model human knowledge and apply it to various domains, such as medicine, law, and art.
Students can freely explore their research topics while assisting with the lab's projects. We also welcome collaborative research with other research groups and companies. If you are interested in joining the lab or collaborating with us, please feel free to contact us via the lab's email.
Team
Lab Members
Technical Staff -
Researcher -
M1 - Early detection of depression on social media
M1 - Personalised healthcare chatbot
FAQs
Frequently Asked Questions
How can I join the lab?
University of Tsukuba students who are or will be enrolled in the Master or Doctoral program in Informatics can join the lab. Please refer to the "How to Join" (to be announced) page.
Does the lab accept research students (Kenkyusei)?
No, we do not accept research students. Please directly apply for the Master or Doctoral program if you want to join the lab.
Do you hire postdocs or research assistants?
Currently, we are recruiting those who can work around two days a week. Please contact us via the contact form above.
Does the lab accept collaborations?
We are warmly open to collaborations with other research groups and companies. Feel free to submit your inquiry via the contact form above.
What resources are available in the lab?
Local LLM servers and per-student desks with external displays are available for students. Also, according to the research interests, additional supervision by co-researchers can be arranged.
What are the lab hours?
There are no fixed lab hours. Students can use the lab freely. We hold a meeting to share research progress about once a week.