Presentation 2011-03-03
An Efficient Pattern Evaluation method using ZDDs for Mining Lifelogs
Masaaki NISHINO, Toru KOBAYASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) To use multiple and large amount of lifelogs effectively, we need methods that can efficiently analyze these data. In this report, we propose a novel data mining method that enables us to analyze the data with subtle relationships between different kind of data. We show a method that reveals both the properties of combinations of the items have, and the set consists of the similar terms if a subset of the items are given as a query pattern. The effectiveness of our methods comes from the use of Zero-Suppressed Binary Decision Diagrams(ZBDDs) : a data structure that can represent the set of the combinations of the items with less memory and we can execute some kinds of set operations on them. We show our method can be executed effectively with the use of ZBDDs in experiments using MovieLens data sets.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Lifelog / Data Mining / ZDD
Paper # LOIS2010-69
Date of Issue

Conference Information
Committee LOIS
Conference Date 2011/2/24(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 Life Intelligence and Office Information Systems (LOIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Efficient Pattern Evaluation method using ZDDs for Mining Lifelogs
Sub Title (in English)
Keyword(1) Lifelog
Keyword(2) Data Mining
Keyword(3) ZDD
1st Author's Name Masaaki NISHINO
1st Author's Affiliation NTT Cyber Solutions Laboratories, NTT Corporation()
2nd Author's Name Toru KOBAYASHI
2nd Author's Affiliation NTT Cyber Solutions Laboratories, NTT Corporation
Date 2011-03-03
Paper # LOIS2010-69
Volume (vol) vol.110
Number (no) 450
Page pp.pp.-
#Pages 6
Date of Issue