Privacy-Preserving-Outsourced Association Rule Mining on Vertically Partitioned Databases

Project Discription

methodology can help associating knowledge gaps in human understanding. Such as analysis of any student dataset gives a better student model yields better instruction, which leads to improved learning. More accurate skill diagnosis leads to better prediction of what a student knows which provides better assessment. Better assessment leads to more efficient learning overall. The main objectives of data mining in practice tend to be prediction and description [4, 5]. Predicting performance involves variables, IAT marks and assignment grades etc. in the student database to predict the unknown values. Data mining is the core process of knowledge discovery in databases. It is the process of extracting of useful patterns from the large database. In order to analyze large amount of information, the area of Knowledge Discovery in Databases (KDD) provides techniques by which the interesting patterns are extracted. Therefore, KDD utilizes methods at the cross point of machine learning, statistics and database systems.

Whole work is a combination of two steps where first include site data creation (encryption, hiding association rule etc.) while second include finding patterns from the encrypted data from various data owners. Explanation of whole work is shown in fig.

As original dataset need to be perturbed so it should be read from the stored file format to the matrix in which each row and column is same as the dataset but here it can be read and written much easier way. Some time dataset is represented as the sequence of data where pattern need to be found in order to generate the set while text on the basis of the pattern one can judge that this data is of same column and  same row, this can be understood as the let Dr is the rough dataset which can be represented in table.

As scientists are chipping away at various field out of which finding a powerful vertical examples is measure issue with this becoming advanced world. has proposed an secured information distribution algorithm. By the utilization of Paillers encryption calculation security of the information at server side get increases. Results demonstrates that proposed work execution time get decrease. While batch passed time get decrease. By the utilization of programmed vertical example space cost is additionally diminish.

Project Sample Image