Web Document Segmentation Using Frequent Term Sets for Summarization

Abstract

Query sensitive summarization aims at extracting the query relevant contents from web documents. Web page segmentation focuses on reducing the run time overhead of the summarization systems by grouping the related contents of a web page into segments. At query time, query relevant segments of the web page are identified and important sentences from these segments are extracted to compose the summary. DOM tree structures of the web documents are utilized to perform the segmentation of the contents. Leaf nodes of DOM tress are merged to form segments according to the statistical and linguistic similarity measure. The proposed system has been evaluated by intrinsic approach making use of user satisfaction index. The performance of the system is compared with summarization without using preprocessed segments. Performance of this system is more promising than the other measures like cosine similarity, jaccard measure which make use of sparse term-frequent vectors, since the most frequent term sets are considered to measure the relevance. Relevant segments alone need to be processed at run time for summarization which reduces the time complexity of the summarization process.

Web Page Segmentation Revisited

“…The most common one (though used in only four publications) is that of a visual "block" with coherent content [9,24,26,37]. Other definitions characterize segments by their edges [12,13], as being semantically self-contained [16], as distinct [30], or as labeled with a heading [28]. Only two papers resort to HTML/DOM elements or sub-trees as segment building blocks [9,24].…”

Selection of Ontology for Web Service Description Language to Ontology Web Language Conversion

“…WSDL provide the foundation for composition of web service, by providing the support in information exchange between the service, it is not rich enough to specify the semantic of the composition and they are not understand by machine. Pasupathi et al (2012), focused on segment the content of web document that highly related with query. It is an simple attempt made over the text comparision.…”