NumberII-2-p9
Title Open Source Affect Analysis System with Extensions
Authors Michal Ptaszynski (Kitami Institute of Technology)
Fumito Masui (Kitami Institute of Technology)
Pawel Dybala (Otaru University of Commerce)
Rafal Rzepka (Hokkaido University)
Kenji Araki (Hokkaido University)
Abstract In research on human-agent interaction (HAI) it is important to perform affect analysis, or emotion recognition of user input for further implementation, such as adaptation of agent behavior to affective states of the user. Several affect analysis systems have been proposed till now. However, none of them has been released yet as an open source software. We present the first open source affect analysis system, ML-Ask, and some of its extensions. The system has been developed for several years and has matured enough to be released to the public. The system can be used for basic affect analysis in HAI research for Japanese, as well as an experimental baseline for specific research in affect analysis.
PDF [Link]