Finding the Provenance of k-anonymous Data and Adding It to Association-rule Mining

Provenance of data describes the process of the generation and transformation of the date, which is used in data mining, data verification, data recovery and reference and some other areas. For uncertain data, provenance plays an important role in query and analytical processing. K-anonymous data is a special kind of uncertain data, which is derived according to the generalization tree by the definite data. Thus, generalization tree is an essential part of provenance of the k-anonymous data. Through the analysis of the generation of the k-anonymous data, we propose a formal definition of the generalization tree and the construction algorithm from the perspective of recipients of the data. On that basis, we present an algorithm for mining association-rule on the k-anonymous data. Finally, compared with the experiment mining of original data, we can verify that the new algorithm is feasible, and theoretical performance analysis shows that the new algorithm can greatly reduce the time complexities of mining and improve the efficiency of mining.