Professor Hisashi Miyamori

Area and Subject Taught Information Retrieval
Research Theme(s) Next-generation Search technology Based on Analysis of the Semantics, Quality, and Sociality of Media Information
Academic Degrees Doctor of Engineering, Waseda University
Keywords for Research Field Information Retrieval, Multimedia, Pattern Recognition, Machine Learning, Natural Language Processing, Image and Video Processing, Web Computing, Information Extraction, Data Mining
Office Phone Number 81-75-705-1533

Research Overview

With the huge amount of information from the Web, videos, and natural language texts in our society, it has been increasingly important to efficiently obtain and organize useful information of good quality, and to provide them in an appropriate manner according to the user's needs.
In our laboratory, we are investigating the following questions:

  1. How can we tell credible information from the mixture of diff erent information, and how should we organize and utilize them as valuable sources of knowledge?
  2. How can we effectively obtain and provide information appropriate to a given circumstance from the mass of information that exists?
  3. How can we produce computers that can understand the meaning of information as humans do and make accurate judgments about the value the information has?

To answer these questions, we are carrying out research and development from various angles by examining huge amount of multimedia data such as Web contents, television programs, and information from sensors while constructing an application system such as question-answering and dialogue systems, that utilizes technologies for information retrieval, pattern recognition, machine learning, natural language processing, image and video processing, and Web computing.

Notable Publications and Works in the Last Three Years

  1. Tasuku Kimura, Ryosuke Nakata, Hisashi Miyamori: KSU Team's Multiple Choice QA System at the NTCIR-12 QA Lab-2 Task, The 12th NTCIR Conference, 07-NTCIR-QALab, pp.437-444, 2016
  2. Tasuku Kimura, Hisashi Miyamori: A Method of Classifying Relationships between Hashtags using Co-occurrence and Latent Topics, IEICE Trans.Inf.&Syst.(D), J98-D, 8, pp.1151-1161, 2015
  3. Tasuku Kimura, Hisashi Miyamori: KSU Team's System and Experience at the NTCIR-11 RITE-VAL Task, The 11th NTCIR Conference, RITE-VAL-06, pp.255-261, 2014