Empirical evaluation of anonymized data with ARX Anonymization Tool

• Item type: SIG Technical Reports(1)
• 公開日: 2023-07-17
• タイトル: Empirical evaluation of anonymized data with ARX Anonymization Tool
• タイトル / 言語: en
• タイトル: Empirical evaluation of anonymized data with ARX Anonymization Tool
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: CSEC
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Institute of Statistical Mathematics
• 著者所属: Institute of Statistical Mathematics
• 著者所属(英): en
• 著者所属(英): Institute of Statistical Mathematics
• 著者所属(英): en
• 著者所属(英): Institute of Statistical Mathematics
• 著者名: Takumi, Sugiyama; Kazuhiro, Minami
• 著者名(英): Takumi, Sugiyama; Kazuhiro, Minami
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: We examine the feasibility of evaluating the risk and utility of anonymized data using the open resource anonymization tool called ARX. As a first step, we study the risk of k-anonymized medical data based on the privacy metric of l-diversity. Our experiments show that significant portions of equivalence classes in k-anonymized data fail to satisfy the diversity requirement of l-diversity. However, we also find that anonymized data based on l-diversity further reduces the utility of anonymized data in comparison with that based on k-anonymity. We finally discuss some lessons from using ARX from the practical viewpoint.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: We examine the feasibility of evaluating the risk and utility of anonymized data using the open resource anonymization tool called ARX. As a first step, we study the risk of k-anonymized medical data based on the privacy metric of l-diversity. Our experiments show that significant portions of equivalence classes in k-anonymized data fail to satisfy the diversity requirement of l-diversity. However, we also find that anonymized data based on l-diversity further reduces the utility of anonymized data in comparison with that based on k-anonymity. We finally discuss some lessons from using ARX from the practical viewpoint.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA12628305
• 書誌情報: 研究報告セキュリティ心理学とトラスト（SPT）; 巻 2023-SPT-52, 号 48, p. 1-7, 発行日 2023-07-17
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8671
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会