Today's businesses are facing a number of difficulties in statistics, analysis, and processing of their data sources to make appropriate business decisions. The reason is that the data source of the enterprise is stored discretely in many file types with different structures and is not unified. In this paper, we design and build a model of a centralised storage system, using big data mining to provide business data analysis functions according to their business requirements. To test and evaluate the effectiveness of the proposed model, we use the input data, which are the actual business data set of an accessory business. The results when the model is applied show that the source data sets are organised, stored, and analyzed with many different criteria, and are displayed on the charts in an obvious and detailed way. Furthermore, we also compare the proposed model with some existing models. The results show that the proposed model is easy to use for end-users. It has high scalability and fault tolerance, and a faster processing speed compared to traditional models.
Keyword
business intelligence, data warehouse, data lake, big data, distributed storage, and processing.