Association mining from large document collection indexed by grammar-based compression
Finished by March 31, 2013
![English](../../../image/english-2.gif)
![Japanese](../../../image/japanese.gif)
![Researcher](../image/title.gif)
![Photo](../../../researcher/2ki/image/sakamoto.jpg)
![Hiroshi Sakamoto](../../researcher/2ki/image/sakamoto.gif)
Kyushu Institute of Technology, Faculty of Computer Science and Systems Engineering, Professor
![Research Site](../image/site.gif)
![Theme](../image/theme.gif)
![Abstract](../image/abstract.gif)
This research aims to discover hidden knowledge from large document collection by the technique of grammar-based compression, which is proposed to find important portions of text data by reducing redundancy. Practically, the researcher would challenge to the difficult problem of identifying important associations included in real-world texts like news articles, patent data, and gene sequences.
![Copyright © Japan Science and Technology ,All Right Reserved.](../../../image/copy.gif)