Information retrieval from web databases using semantic similarity

In the recent years, the database access is done by the HTML form based interfaces easily. The textual data deal with the natural languages, knowledge acquisition and finally information retrieval. The data units are returned from the databases and encoded into result pages for human browsing. For the encoded data processable, they need to assign a label for each group. In this paper, an automatic annotation approach aligns the resultant data into a similar group of data units. For each group, we annotate the results and assign the label for it. Then survey and classify the search result records for evaluate their concept similarity and dissimilarity. Finally the annotation wrapper is constructed automatically.