LAP Lambert Academic Publishing Development of a Summarisation System for Web Pages

Development of a Summarisation System for Web Pages is an MSc project that designed a summarisation system for search engine results by looking into the web pages and summarising it to a page. The purpose of this work is to provide information to the solution of multi web page summarisation problem by providing a system that will summarise two to four web pages at a time to easy researchers in getting right information from the internet at a minimal time. The work produces a temporary webpage that contains extract of salient sentences from pages got from query posed on Google search engine. Uniform Resource Locator (URL) data harvested from Google search engine was used to develop and test the summarisation system. The web page summarisation system was designed by first analysing and characterising web pages using Text difference function to check for relevancy of the web pages to be summarised. Pair wise bipartite graph theory was used to check for redundancy and score the sentences. The result from the system shows that time to glance through all web pages before getting concrete summary is reduced using this package.