2009 Fiscal Year Final Research Report

Visualization of structured data based on graph representation

Research Project

Project/Area Number	20700081
Research Category	Grant-in-Aid for Young Scientists (B)
Allocation Type	Single-year Grants
Research Field	Media informatics/Database
Research Institution	Chiba University
Principal Investigator	MORI Yasukuni Chiba University, 大学院・融合科学研究科, 助教 (40361414)
Project Period (FY)	2008 – 2009
Keywords	データ可視化 / パターン認識 / データマイニング / 特徴選択
Research Abstract	A methodology which extract an effective information from data with complicated structure or distribution was proposed in this study. The proposed method is based on graph-representation which express a subset of data as a node and a relationship between subsets as a edge. The proposed method was applied to some real data sets. As a result, the method can represent the characteristics of data effectively. It is expected that the learning performance or recognition rate of classification problem is improved and that we find a clue to elucidate the network structure of data with dependencies among them.