Presentation 2019-03-15
Proposal of home context recognition method using feature values of cognitive API
Sinan Chen, Sachio Saiki, Masahide Nakamura,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, with the development of deep learning technology, context recognition technology utilizing multimedia data is expected to be promising. We consider using this technology for home context recognition in smart homes. Generally, in home context recognition, the layout of the room and the environment for each household are different from the requirements of the context to be recognized, it is necessary to construct a unique recognition model for each household. At this time, in the approach using deep learning, a great deal of training data is required, and it is practically impossible to conduct it at home. The purpose of this research is to develop a context recognition method using images that can be introduced at home. In the proposed method, we utilize a cognitive API that performs general image recognition and retrieve the information recognized in the image as text. By using this as a feature quantity, we classify the context by applying usual supervised machine learning. Compared with the approach using deep learning, the proposed method uses generic image recognition and lightweight machine learning provided by the cognitive API. As a result, context recognition specialized for each household can be realized with much less effort.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Context recognition / Image / Cognitive API / Machine learning
Paper # SC2018-38
Date of Issue 2019-03-08 (SC)

Conference Information
Committee SC
Conference Date 2019/3/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English) National Institute of Informatics
Topics (in Japanese) (See Japanese page)
Topics (in English) Services and Machine Learning, etc.
Chair Masahide Nakamura(Kobe Univ.)
Vice Chair Shinji Kikuchi(Univ. of Aizu) / Yoji Yamato(NTT)
Secretary Shinji Kikuchi(NEC) / Yoji Yamato(Fujitsu Lab.)

Paper Information
Registration To Technical Committee on Service Computing
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Proposal of home context recognition method using feature values of cognitive API
Sub Title (in English)
Keyword(1) Context recognition
Keyword(2) Image
Keyword(3) Cognitive API
Keyword(4) Machine learning
1st Author's Name Sinan Chen
1st Author's Affiliation Kobe University(Kobe Univ.)
2nd Author's Name Sachio Saiki
2nd Author's Affiliation Kobe University(Kobe Univ.)
3rd Author's Name Masahide Nakamura
3rd Author's Affiliation Kobe University(Kobe Univ.)
Date 2019-03-15
Paper # SC2018-38
Volume (vol) vol.118
Number (no) SC-511
Page pp.pp.7-12(SC),
#Pages 6
Date of Issue 2019-03-08 (SC)