Thanks to the recent advances of IoT technology, we can easily obtain a wide variety of data.
However, in many cases, it is necessary to screen out unnecessary information in order to make effective use of the collected data.
In addition, it is necessary to accurately annotate the data by hand for utilizing them in a machine learning domain.
On the other hand, unsupervised knowledge extraction is a promising approach to directly utilize a wide variety of data. Especially, unsupervised knowledge extraction is useful in unknown or dynamic environments because it can adaptively extract knowledge from continually given data.
The purpose of this special session is to discuss unsupervised knowledge extraction techniques in intelligent systems for extracting, storing, and exploiting multidimensional information from environmental data with multiple modalities. This session welcomes fundamental research on information extraction techniques and knowledge-based relation learning approaches. Moreover, the session also welcomes research on applied topics relating to exploiting knowledge with intelligent systems.
Continual Learning, Lifelong Learning, Online Learning, Deep Learning, Knowledge Extraction
Yuichiro Toda, Okayama University
Vincenzo Lomonaco, University of Pisa
Naoki Masuyama, Osaka Prefecture University
Andrea Cossu, Scuola Normale Superiore
Seiki Ubukata, Osaka Prefecture University
Chin Wei Hong, Tokyo Metropolitan University
Paper submission: January 31, 2022 (11:59 PM AoE)
Notification of acceptance: April 26, 2022
Final paper submission: May 23, 2022