Context-aware OLAP for textual data warehouses

Decision Support Systems (DSS) that leverage business intelligence are based on numerical data and On-line Analytical Processing (OLAP) is often used to implement it. However, business decisions are increasingly dependent on textual data as well. Existing research work on textual data warehouses has the limitation of capturing contextual relationships when comparing only strongly related documents. This paper proposes an Information System (IS) based context-aware model that uses word embedding in conjunction with agglomerative hierarchical clustering algorithms to dynamically categorize documents in order to form the concept hierarchy. The results of the experimental evaluation provide evidence of the effectiveness of integrating textual data into a data warehouse and improving decision making through various OLAP operations.