Integration of Data Warehouse and Unstructured Business Documents

The profusion of unstructured data forced organizations to manage and take advantage of such data especially in the decision making process. The feasibility of integrating or mapping unstructured data to a data warehouse is becoming significant to bridge this gap and take the full potential of these data. In this paper, we propose a multi-layer schema for mapping structured data stored in a data warehouse and unstructured data in business-related documents. The multi-layer schema facilitates the mapping between the two different data. Linguistically correlated data is identified using Word Net to enable the integration between both data sources. We also propose a generic XML schema for business-related unstructured documents to assist the mapping. The use Word Net to identify the matching result is promising in the absence of schema-instance and without the need to domain specific knowledge.