Concept Drift in documents and Formal Concept Analysis

  • Authors:
  • Yutaka Miyazaki;Yuzuru Tanaka

  • Affiliations:
  • Meme Media Laboratory, Hokkaido University, Sapporo, Hokkaido, 060-8628, JAPAN;Meme Media Laboratory, Hokkaido University, Sapporo, Hokkaido, 060-8628, JAPAN

  • Venue:
  • Proceedings of the 2010 conference on Information Modelling and Knowledge Bases XXI
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Formal Concept Analysis (FCA) enables us to extract conceptual information from data and to visualize its structure in a lattice form. Here, we apply FCA to the analysis of a series of documents published periodically, in order to capture drifts of formal concepts occurred in the series by shifting the corresponding formal context as a viewpoint by some constant amount of time little by little. The resulting concept drifts can reflect the change of topics frequently discussed at that time.