MultiVision: Designing Analytical Dashboards with Deep Learning Based Recommendation
This addresses the challenge for data workers in efficiently creating meaningful dashboards, though it is incremental as it builds on prior automated visualization recommendation methods.
The paper tackles the problem of automating the design of analytical dashboards from data tables, which is tedious for data workers, by proposing a deep-learning-based method that recommends multiple-view visualizations and integrates into a mixed-initiative system, with results showing improvements over existing rule-based approaches in user studies.
We contribute a deep-learning-based method that assists in designing analytical dashboards for analyzing a data table. Given a data table, data workers usually need to experience a tedious and time-consuming process to select meaningful combinations of data columns for creating charts. This process is further complicated by the need of creating dashboards composed of multiple views that unveil different perspectives of data. Existing automated approaches for recommending multiple-view visualizations mainly build on manually crafted design rules, producing sub-optimal or irrelevant suggestions. To address this gap, we present a deep learning approach for selecting data columns and recommending multiple charts. More importantly, we integrate the deep learning models into a mixed-initiative system. Our model could make recommendations given optional user-input selections of data columns. The model, in turn, learns from provenance data of authoring logs in an offline manner. We compare our deep learning model with existing methods for visualization recommendation and conduct a user study to evaluate the usefulness of the system.