Privacy Protection Using Sensitive Data Protection Algorithm In Frequent Itemset Mining Of Medical Datasets

Frequent Itemset Mining (FIM) is one of the most eminent techniques in the Data mining systems. The exploration of Frequent Itemset Mining distills the recurring knowledge from the incessant data. Explosion of Frequent Itemset Mining in the field of Data Analysis and Data Mining becomes an inescapable one. The paper focuses on “searching the accurate records of efficient database queries without compromising the breach of trust using Sensitive Data Protection Algorithm”. This algorithm divides the database into several partitions. Each partition holds the frequent items of the database in a ranked manner. In data protection perspective, the data gets luxated rather than adding blowing of noise. Next, a user-defined threshold is located to retrieve the necessitate records from the datasets which reduces the time consumed for scanning the whole database. The above process is executed for only authorized user, if any violation, an alert is activated and forwarded to the specified user. In experimental view, it is checked on “Doctor online” dataset that widely used for analysis of medical database queries. Performance metrics such as Precision and Recall is analyzed. Experimental results prove that the medical records are easily retrieved and protected with an improved sensitive data protection.