Interpreting the structure and results of a data warehouse model using ontology and machine learning techniques

Abstract

In this paper, we present an intelligent methodology for assisting decision-makers in both understanding the structure of a data warehouse model and making decisions. The support module proposed by our method comprises three operations: (i) transforming a data warehouse model into an ontology, allowing for the display of the different terminology related to a specific domain as well as the different semantic relationships between them, (ii) recommending a series of queries to the decision-maker that enables an understanding of the reasoning logic based on the ontology’s structure, (iii) enriching the different results obtained from some analysis tools through the use of advanced machine learning techniques. The originality of our proposed methodology lies in its ability to influence a decision-maker’s thinking in order to encourage him to take full advantage of the services provided by the data warehouse model. We apply our proposed methodology to an extended data warehouse model that enables the analysis of social media data related to people with personality disorders (PD). The primary goal of this model is to provide decision-makers with suitable services that allow them to make meaningful decisions for people with personality disorders around the world. This task was carried out by analyzing the activities and content of people on social media. In addition, one of the main advantages of this model is the use of various artificial intelligence (AI) and natural language processing (NLP) techniques. Our proposed methodology is implemented and the results achieved are evaluated using both quantitative and qualitative methods.