Presentation 2005/7/25
LooM : An Anonymity Quantification Method for Privacy Protection
Miyuki IMADA, Masakatsu OHTA, Masayasu YAMAGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a novel anonymity quantification method which calls LooM. The LooM is a method that evaluates the privacy protection levels every property of user collection for realizing a disclosure negotiation function of private information for user agents. Its main feature is that it can quantitatively control anonymity by a single value (disclosure threshold value) using a classification algorithm of the decision tree. In this paper, we show that the LooM is hardly affected by size of privacy information database or the attribute values distribution of users' private information by using artificial database. In order to decide the disclosure threshold value on practical systems, we show a method for setting the value based on web questionnaire data.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Privacy Protection / Decision Tree / Anonymity / Equilibrium Model
Paper # AI2005-9
Date of Issue

Conference Information
Committee AI
Conference Date 2005/7/25(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) LooM : An Anonymity Quantification Method for Privacy Protection
Sub Title (in English)
Keyword(1) Privacy Protection
Keyword(2) Decision Tree
Keyword(3) Anonymity
Keyword(4) Equilibrium Model
1st Author's Name Miyuki IMADA
1st Author's Affiliation NTT Network Innovation Labs.()
2nd Author's Name Masakatsu OHTA
2nd Author's Affiliation NTT Network Innovation Labs.
3rd Author's Name Masayasu YAMAGUCHI
3rd Author's Affiliation NTT Network Innovation Labs.
Date 2005/7/25
Paper # AI2005-9
Volume (vol) vol.105
Number (no) 224
Page pp.pp.-
#Pages 6
Date of Issue