Key Issues of Security and Integrity in Third Party Association Rule Mining

• Main Authors: Sethi, K.K., Mishra, D.K., Solanki, G., Mishra, B.
• Format: Conference Proceeding
• Language: English
• Subjects: Association rules; Data engineering; Data mining; Data security; Decision making; Educational institutions; Electronic mail; Information security; Itemsets; Transaction databases
• ISSN: 2157-0477; 2157-0485
• DOI: 10.1109/ICETET.2009.136

Association rule mining discovers interesting association or correlations among a large set of data items. Association rule mining make decision making process easier by providing the important information to the user. The most of the work done in this field is concerned with mining in an isolated form, but it is not only the case and many times more than one parties are involved in this process. Data mining task may be done by third party as a service provider. The database is made available to service provider and he returns the computed association rule to the owner of the database. For example, number of companies dealing in same type of product or services, may work together to identify the current trends in the industry & their customer. The organizations have to share the data & information with their friend or competitor to identify the global trends. There are many situations where more than one parties involve in the data mining process. In this situation we require security of data so that information stealing by the dishonest party from dataset can be stopped and integrity of result be ensured by stopping dishonest party to corrupt the result or process. In this paper, we have summarized the integrity & security issues related to association rule mining in third party computation.