Identification of semantic patterns in e learning by automatic constructing taxonomy using Bayesian rose tree

Abstract

E-Learning can be viewed as a learning process created by interaction newlinewith digitally delivered content, services, and support to learners. It involves newlineintensive usage of Information and Communication Technology (ICT) to newlineserve, facilitate, and revolutionize the teaching learning process. The rapidly newlinegrowing use of ICT in e-learning, is changing the way in which knowledge is newlinecreated, organized, stored, managed, and disseminated. Knowledge newlineManagement and e-learning are closely related, since, e-learning learners newlineneed a suitable amount of knowledge management that can help them to newlineobtain the related, relevant and complete content they require. Knowledge newlineManagement oriented e-learning has become the effective tool that improves newlinethe learning experience of the learners.The main objective of my thesis is to enhance the knowledge newlinemanagement based e-learning services using semantic approaches. In newlineparticular, the research deals with automatically constructs taxonomy from a newlineset of keywords for data sharing reuse and data search wherein the newlineconstructed taxonomy should be independent from different records type. newlineAdditionally, weighted technique is used to mine the web contents with focus newlineon ranking. newlineA deployment approach used in building taxonomy is Bayesian Rose newlineTree and K-mean nearest neighbor classifier, so that the number of distinct newlinevalues will increase the performance of data mining version in terms of newlinecategory accurateness and by inducing a lexical taxonomy for a given list of newlinewords newlineSemantic search engine is developed to enhance the performance of newlinesearch using semantic web concepts. It is aimed to retrieve the information in newlinea well-defined meaning, highly specific and machine understandable. The newlineproposed work also filters the inappropriate files based on user s query. The newlinesemantic search engine system consists of four modules such as Semantic newlineparser, Semantic reasoner, Document retriever and Document ranker.