An Effective Metric Measuring the Degree of Web Page Changes by Reliance Evaluation

A number of document similarity metrics have been used to measure the degree of web page changes. When a web page changes, the metrics often represent the change differently. In this paper, we first define criteria for evaluating the effectiveness of the metrics in terms of six important types of web page changes. The criteria satisfy ISO/IED 15408 (CC). Second, we propose a new document similarity metric appropriate for measuring the degree of web page changes. In our experiment, we evaluate the five existing metrics (i.e., the byte-wise comparison, the TF·IDF cosine distance, the word distance, the edit distance, and the shingling) and ours under the proposed criteria. The experimental result shows that our metric represents the changes more effectively than other metrics. This work was supported by Korea Research Foundation Grant (KRF-2004-005-D00172).