Incorporating Optional Labeling And Dynamic Tagging With Combining Tag And Value Similarity

Query result pages are generated from the web databases based on the query given by the user. From these query result pages data are extracted automatically. The data extraction is based on the combined tag and value similarity technique. The Query Result Record (QRRs) in the query result pages are identified and segmented. The segmented QRRs are then aligned in to a table. The non contiguous QRRs are also considered which is induced by the auxiliary information. We propose new techniques called Optional Labeling and Dynamic Tag structuring which improves the efficiency in data extraction. Initially all the tags are stored temporarily in a database, from where relevant tags are extracted. The tag structuring is handled dynamically so that more accurate extraction is made possible.