Extracting TARs from XML for Efficient Query Answering

The massive amount of datasets expressed in different formats,such as relational, XML, and RDF, avail-able in several realapplications, may cause some difficulties to non-expert userstrying to access these datasets without having sufficientknowledge on their content and structure. Moreover, theprocesses of query composition, especially in the absence of aschema, and interpretation of the obtained answers may benon-trivial. The existing data mining process is often guided bythe designer, who determines the portion of a dataset whereuseful patterns can be extracted based on his/her deepknowledge of the application scenario. In this paper, wepropose efficient mining techniques to mine hiddeninformation from huge datasets, and then use it in order to gainuseful knowledge which helps inexperienced users to accesshuge XML datasets. We also describe XML mining tool whichimplemented using Java encompasses two main features 1) itmines all the frequent association rules from input documentswithout any a-priori specification of the desired results 2) itprovides quick, summarized, thus often approximate answersto user’s queries, by using the previously mined knowledge.