An Efficient Web Page Change Detection System Based on an Optimized Hungarian Algorithm

This paper describes an efficient Web page change detection system based on three optimizations that were implemented on top of the Hungarian algorithm, which we employ to compare trees that correspond to HTML Web pages. The optimizations attempt to stop the comparator algorithm that employs this O(n
³
) algorithm before it completes all its iterations based on criteria having to do with properties of HTML and heuristics. Analysis and experimental results prove the effectiveness of these optimizations and their ability to render O(n
²
) performance, where n denotes the number of nodes in the tree. A complete system was implemented and used to carry out the performance experiments. This system includes functionalities and interfaces for processing user requests, fetching Web pages from the Internet, allowing users to select zones in Web pages to monitor, and highlighting changes on the Web pages being monitored