A domain-independent approach to discourse-level knowledge discovery from texts

  • Authors:
  • John A. Atkinson-Abutridy

  • Affiliations:
  • Departamento de Ingeniería Informática, Universidad de Concepción, Concepción, Chile

  • Venue:
  • IEA/AIE'2005 Proceedings of the 18th international conference on Innovations in Applied Artificial Intelligence
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a new approach for mining novel patterns from textual databases which considers both the mining process itself, the evaluation of this knowledge, and the human assessment. This is achieved by integrating Information Extraction technology and Genetic Algorithms to produce high-level explanatory novel hypotheses. Experimental results using the model are discussed and the assessment by human experts are highlighted. This research is sponsored by the National Council for Scientific and Technological Research (FONDECYT, Chile) under grant number 1040469 "Un Modelo Evolucionario de Descubrimiento de Conocimiento Explicativo desde Textos con Base Semántica con Implicaciones para el Análisis de Inteligencia."