Efficient Estimation of Plausible Abductive Hypotheses Using A<sup>*</sup> Search

Bibliographic Information

Other Title
  • A<sup>*</sup>探索に基づく仮説推論の効率化

Search this article

Description

Abduction is inference to the best explanation. While abduction has long been considered as a promising framework for natural language processing (NLP), its computational complexity hinders applying abduction to practical NLP problems. In this paper, we adapt an A* search-based technique for abduction in order to prune its search space efficiently. More specifically, we present an algorithm to effectively eliminate literals which do not introduce any unification from the search space. Our experiment reveals that our search-based technique considerably reduces the size of the search space and improves the efficiency of the state-of-the-art inference algorithm for abduction on first-order logic.

Journal

  • IPSJ SIG Notes

    IPSJ SIG Notes 2014 (10), 1-9, 2014-06-26

    Information Processing Society of Japan (IPSJ)

Keywords

Details 詳細情報について

  • CRID
    1573387452738608128
  • NII Article ID
    110009799604
  • NII Book ID
    AN10115061
  • Text Lang
    ja
  • Data Source
    • CiNii Articles

Report a problem

Back to top