Presentation 2002/3/15
Ordering Hypotheses by Prioritized Inductive Learning
Kousuke TSUJITA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Induction is a learning method for drawing a general rule from the background knowledge and observed examples. Existing induction systems do not consider priority relations between acquired knowledge. However, in the real world a person puts priorities over his knowledge and appropriately uses it. In this research, we propose a method for ordering inductive hypotheses based on observations. When a new observation is given to the background knowledge base, the explanation structure of the observation is produced and hypotheses are quantitatively ordered using Bayes theorem. As a result, the system can dynamically change the ranking of hypotheses according to the rate of the instances/exceptions of a concept in observations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Machine Learning / Inductive Learning / Probabilistic Reasoning / Knowledge Information Processing / Bayesian Network
Paper #
Date of Issue

Conference Information
Committee AI
Conference Date 2002/3/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Ordering Hypotheses by Prioritized Inductive Learning
Sub Title (in English)
Keyword(1) Machine Learning
Keyword(2) Inductive Learning
Keyword(3) Probabilistic Reasoning
Keyword(4) Knowledge Information Processing
Keyword(5) Bayesian Network
1st Author's Name Kousuke TSUJITA
1st Author's Affiliation Faculty of Systems Engineering Wakayama University()
Date 2002/3/15
Paper #
Volume (vol) vol.101
Number (no) 742
Page pp.pp.-
#Pages 8
Date of Issue