Evolutionary Approach for XML Data Mining

Abstract

Extensible Markup Language(XML) has emerged as a medium for interoperability over the Internet. The XML technology, with its self-describing and extensible tags, is significantly contributing to the next generation semantic web. The present search techniques used for HTML and text documents are not efficient to retrieve relevant XML documents. In chapter four, Self Adaptive Genetic Algorithms for XML Search(SAGAXSearch) is presented to learn about the tags, which are useful in indexing. The indices and relationship strength metrics are used to extract fast and accurate semantically related elements in the XML documents. Experiments are conducted on the DataBase systems and Logic Programming (DBLP) XML corpus and are evaluated for precision and recall. The proposed SAGAXSearch outperforms the existing algorithms XSEarch and XRank with respect to accuracy and query execution time.

As the number of documents published in the form of XML is increasing, there is a need for selective dissemination of XML documents based on user interests. In the proposed technique, a combination of Self Adaptive Genetic Algorithms and multi class Support Vector Machine is used to learn a user model. Based on the feedback from the users, the system automatically adapts to the users preference and interests. The user model and a similarity metric are used for selective dissemination of a continuous stream of XML documents. Experimental evaluations performed over a wide range of XML documents, indicate that the proposed approach significantly improves the performance of the selective dissemination task, with respect to accuracy and efficiency.

On similar grounds, there is a need for categorization of XML documents into specific user interest categories. However, manually performing the categorization task is not feasible due to the sheer amount of XML documents available on the Internet. A machine learning approach to topic categorization which makes use of a multi class SVM for exploiting the semantic content of XML documents is also presented. The SVM is supplemented by a feature selection technique which is used to extract the useful features. Experimental evaluations performed over a wide range of XML documents indicate that the proposed approach significantly improves the performance of the topic categorization task, with respect to accuracy and efficiency.