Extract Deep Web-Detail Pages with Simple Tree Match

In this paper, we provide a method to extract data from Deep Web-Detail Pages. The method use the Simple Tree Match to compute the max match value between two trees, and use the Hungarian algorithm to trace the result of the STM compute, after this we use tree merge method to generate Wrapper. At last, we use Term Frequency to optimize the Wrapper. In experimental, we use the Wrapper to extract data; the results show that our method compared with other methods is feasible and effective.