Introduction to A Novel Algorithm for Hiding Sensitive Frequent Item Sets Project:
This paper discussed about fast hiding sensitive frequent itemsets (FHSFI) approach to hide sensitive frequent itemsets with one database scan and generate limited side effects.
Overview:
In a database, with an association analyzer, if an itemset with support above a given minimal support then that itemset is called a frequent itemset. There is a need to analyze the correlations between the sensitive itemsets and each transaction in a database. The transactions order to be altered can be decided based on the each transaction weight which is given by heuristic function.
So by this there is a possibility of saving time by only consider the actual required transactions to modified and then hide the given sensitive frequent itemsets. There is no need to deal with all the transactions to hide the sensitive frequent itemsets. FHSFI provides the facility to hide the SFI without generating many side effects but still there is loss in rule sets, so researches have been takes place to overcome this issue. The main goal of FHSFI is to hide sensitive frequent itemsets by allowing minimum support thresholds, by limited side effects and by executed only one database scan.
Conclusions:
In network and data mining world, there is a severe need to protect the confidentiality of sensitive information in a database. The relationships hidden among large data sets are in a form of frequent item sets or association rules. Privacypreserving data mining is an important issue which needs the immediate attentions in the current industry.
When the support of each given sensitive itemsets are reduced then it will help to hide sensitive frequent itemsets, and this can be achieved by modifying transaction in database but at the same time will have enough side effects due to it. So FHSFI emerged to hide sensitive frequent itemsets with limited side effects.
Download A Novel Algorithm for Hiding Sensitive Frequent Item Sets.