Creating an effective dashboard is a challenge for both analysts and designers. It is difficult to pick the best set of base visualizations for the task and data. This can lead to unsatisfactory analytics results and insights especially when creating a dashboard for a large user base with different preferences. The proposed Intelligent Dashboard addresses this issue.
Automated knowledge discovery and dashboard building system, but allows users to drive the composition
Integrates data visualization knowledge with AI and ML
Finds interesting statistical patterns between attributes in the multivariate data and assigns an interestingness score to them
Associates attributes with 10 well studied data analysis tasks (e.g. find correlations or anomalies)
Generates visualizations based on tasks and ranks them based on interestingness score
Enforces consistency among the base visualization for good storytelling