A Knowledge-Based Visual Analytics Query System

The present visual analytics query system that is being used to process large quantities of information with complex analytic reasoning processes does not support intelligent selection of best visual display and task analysis. In this paper , Knowledge-Based Visual Analytics Query System (KBVAQS) architecture was designed to support and correct the challenges. The architecture consists of three main parts which include the application layer, logic layer and back layer. These parts were used for querying and displaying data graphically based on User request. In the application layer, the administrator and ordinary users interact directly with the system through the user interface provided by the system. Graphical display based on the decision table is provided and some task analyses are given for the user to interpret. The logic layer handled the full functionality of the implementation while the back layer was used for record keeping. The implementation was achieved by employing object oriented programming language C-Sharp with the data base created in Microsoft Structured Query Language. The effectiveness of KBVAQS tool has been evaluated in surveys carried out at the Nigerian Stock Exchange which deals with stock markets. It shows that users generally viewed KBVAQS tools more positively than using existing Intelligent Visual Query Algorithm (IVQA) technique. These differences were significant to p<0.05. The mean interactions precision and calculated value using expert judge relevance rating show a significant difference between KBVAQS tool and IVQA performance 2.47 against 1.73 for precision with calculated t=6.33. The hypothesis testing revealed that KBVAQS user performed better and achieved acceptable results.