Development of Meta Level Communication Analysis using Temporal Data Crystallization and Its Application to Multi Modal Human Communication
Abstract
The purpose of this project is to develop methods to analyse human communication using the temporal data crystallization (TDC). The major achievement is the following four. (1) A method of meta-level discussion analysis. This method was shown to increase precision of TDC analysis by comparing the result of the target record to those of other similar records. Its usefulness was shown by applying it to an example case, (2) Another method to increase precision of TDC by using the nonverbal information during talking. Its effectiveness was shown by applying it to TV debate program, (3) Extraction of human relation by analyzing the Tsugo, which is meta level information of human relation such as intension of actions and their constraints, and (4) Behavioral analysis of people by utilizing ubiquitous sensor data such as acceleration sensor and a nearby sensor. TDC method was applied to face-to-face contact information and human relation was extracted.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 22, 2013
- Accession Number
- ADA587634
Entities
People
- Katsumi Nitta
Organizations
- Tokyo Institute of Technology