Software birthmark based theft detection of JavaScript programs using agglomerative clustering and improved frequent subgraph mining

Today software industry is suffering a lot because of software piracy. With the increased use of websites, the plagiarism of softwares and websites is also increasing. This threatens the intellectual property rights of JavaScript developers. Hence it is our small effort to safeguard the intellectual rights of developers by introducing a system which uses the website's unique behavior at run-time as a birthmark. Software Birthmark is a set of unique characteristics or behavior inhibited by software while it is in execution. The heap snapshots play a vital role in this system. Heap Snapshots provide us the behavior of objects required while execution of websites. In order to merge the heap graphs we used agglomerative clustering approach. In order to find the birthmark of the program or website, it is needed to extract frequently occurring nodes. We worked on 3000 combinations of websites on ten different sectors for both FSM (Apriori approach) and Improved FSM (Frequent Pattern Growth approach). We observed that the Improved FSM (Pattern Growth approach) is more optimum with respect to the time of execution and the rate of theft detection than the FSM (Apriori Approach).