{"title":"A Data Anonymization Method to Mitigate Identity Attack in Transactional Database Publishing","authors":"Dedi Gunawan","doi":"10.1109/ICoICT49345.2020.9166262","DOIUrl":null,"url":null,"abstract":"Publishing transactional database becomes more recognized for many institutions such as retails and groceries. Many of them share or publish their data to other institutions as an effort to gain more revenue for their business. However, publishing such a database is problematic since irresponsible parties may associate records in database with specific individuals to disclose personal identity known as identity attack. Data anonymization is an effective technique to protect database from the threat. Unfortunately, applying data anonymization method in transaction database using generalization and suppression based techniques may reduce data utility significantly and cause severe distortion to database properties. A solution to mitigate such drawbacks has been proposed by replacing item with another item instead of applying those techniques. However, selecting an item to replace another item causes other problems specifically when the selected item for the replacement process is not the optimum one. Therefore, in this paper we propose a data anonymization method which performs item replacement that utilizes weighted scoring method to select an optimal item with respect to minimize information loss and maintaining database properties. Experimental results show that the proposed method guarantee higher privacy protection compared with an existing method and it successfully generates an anonymized database while at the same time it maintains data utility by minimizing information loss more than 50% compared with that of an existing method. In addition, the data property of the anonymized database can be well maintained.","PeriodicalId":113108,"journal":{"name":"2020 8th International Conference on Information and Communication Technology (ICoICT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 8th International Conference on Information and Communication Technology (ICoICT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICoICT49345.2020.9166262","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Publishing transactional database becomes more recognized for many institutions such as retails and groceries. Many of them share or publish their data to other institutions as an effort to gain more revenue for their business. However, publishing such a database is problematic since irresponsible parties may associate records in database with specific individuals to disclose personal identity known as identity attack. Data anonymization is an effective technique to protect database from the threat. Unfortunately, applying data anonymization method in transaction database using generalization and suppression based techniques may reduce data utility significantly and cause severe distortion to database properties. A solution to mitigate such drawbacks has been proposed by replacing item with another item instead of applying those techniques. However, selecting an item to replace another item causes other problems specifically when the selected item for the replacement process is not the optimum one. Therefore, in this paper we propose a data anonymization method which performs item replacement that utilizes weighted scoring method to select an optimal item with respect to minimize information loss and maintaining database properties. Experimental results show that the proposed method guarantee higher privacy protection compared with an existing method and it successfully generates an anonymized database while at the same time it maintains data utility by minimizing information loss more than 50% compared with that of an existing method. In addition, the data property of the anonymized database can be well maintained.