Presentation 2003/3/6
Distributed Knowledge Discovery based on Merging Hypotheses
Toshiyuki KIKUCHI, Akihiro YAMAMOTO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper we propose a method for knowledge discovery from distributed data on distributed computers. Each of the computers is assumed to have a knowledge discovery system which searches hypotheses from general ones to specific ones. We also assume that each hypothesis can be represented as a logic program. Then the discovery method is to correct hypotheses generated by the computers and then merge them. For the merge of hypotheses we use the most general specialization of clauses. We also validate each clause in the merged hypotheses. We implement our method on the EVLD system, which we developed previously, and call it Distributed EVLD (Distributed EVLD). In the paper we report how our method works well with a well-known data set.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) inductive logic programming / distributed processing / specialization
Paper # AI2002-64
Date of Issue

Conference Information
Committee AI
Conference Date 2003/3/6(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) Distributed Knowledge Discovery based on Merging Hypotheses
Sub Title (in English)
Keyword(1) inductive logic programming
Keyword(2) distributed processing
Keyword(3) specialization
1st Author's Name Toshiyuki KIKUCHI
1st Author's Affiliation MemeMedia Laboratory, Hokkaido University()
2nd Author's Name Akihiro YAMAMOTO
2nd Author's Affiliation MemeMedia Laboratory, Hokkaido University
Date 2003/3/6
Paper # AI2002-64
Volume (vol) vol.102
Number (no) 709
Page pp.pp.-
#Pages 6
Date of Issue