Loopo: Integrated Text Miner for FACT-Graph-Based Trend Analysis

  • Authors:
  • Ryosuke Saga;Hiroshi Tsuji;Kuniaki Tabata

  • Affiliations:
  • Kanagawa Institute of Technology, School of Information Technology, Atsugi, Japan 243-0292;Graduate School of Engineering, Osaka Prefecture University, Nakaku, Japan 559-8531;Kanagawa Institute of Technology, School of Information Technology, Atsugi, Japan 243-0292

  • Venue:
  • Proceedings of the Symposium on Human Interface 2009 on Human Interface and the Management of Information. Information and Interaction. Part II: Held as part of HCI International 2009
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes an integrated tool to analyze trend visualization graph called "FACT-Graph". FACT-Graph is generated from text data with time stamp and is useful for trend analysis. However, it faces three key problems: First, it is difficult to configure parameters (such as analysis span, exceptive keywords and thresholds) to generate FACT-Graph; Second, a FACT-Graph does not provide the required information and interface for trend analysis because the process of generating the FACT-Graph eliminates that information; and third, it cannot reflect a user's awareness in a FACT-Graph. In order to solve these problems, the authors have developed a tool called "Loopo". Loopo integrates a term database, analysis components, and a graph-drawing function and provides users (i.e., analyzers) with information for trend analysis. Loopo also provides an interactive GUI for configuring parameters at ease and to reflect a user's awareness in a FACT-Graph instantly.