Абстрактный

Automatic Annotation Wrapper Generation and Mining Web Database Search Result

V.Yogam, K.Umamaheswari

Search result record (SRR) is the result page obtained from web database (WDB) and these records are used to display the result for each query. Each SRR contain multiple data units which need to be label semantically for machine processable. In this paper we present the automatic annotation approach which involve three phases to annotate and display the result. In first phase the data units in result record are identified and aligned to different groups such that the data in same group have the same semantics. In the second phase, for each group we annotate it from different aspects and aggregate the different annotations to predict a final annotation label for it. In third phase, an annotation wrapper for the search site is automatically constructed and can be used to annotate new result pages from the same web database. This approach is highly effective. From the annotated search result, frequently used websites are identified by using apriori Algorithm which involve pattern mining. The advantage of this new technique is fast operation on dataset containing items and provides facilities to avoid unnecessary scans to the database