Improved Integrated Mining Of Heterogeneous Data In Decision Support Systems

ABSTRACT

The volume of information available on the Internet and corporate intranets continues to increase along with the corresponding increase in the data (structured and unstructured) stored by man organizations. In customer relationship management, information is the raw material for decision making. For this to be effective there is need to discover knowledge from the seamless integration of structured and unstructured data for completeness and comprehensiveness. This study addresses two unique challenges experienced in business decision support systems, the first one is how to transform and analyze unstructured data alongside structured data. Secondly, the need to improve result obtained from the integrated mining system in order to reduce decision failure. There is also a necessity to solve the challenge of Customer Relationship Management, in terms of the ability to differentiate useful information from chatter or even disinformation. There is also further need to have a holistic view to mining from structured and unstructured information sources towards a better Customer Relationship Management. Improved Integrated Mining Architecture (IIMA), our approach to solving the above challenges, consists of three major phases; the first phase is the Extraction and Integration phase. This phase is aimed at optimizing the performance of the knowledge mining phase. It consists of unstructured data (text) preprocessing which includes lexical analysis, stemming, application of weighing schemes and finally transforming the documents to an XML format.