Personalized Visualization Recommendation

Visualization recommendation work has focused solely on scoring visualizations based the underlying dataset, and not actual user their past visualization feedback. These systems recommend same for every user, despite that interests, intent, preferences are likely to be fundamentally different, yet vitally important. In this work, we formally introduce problem of personalized present a generic learning framework solving it. particular, focus recommending each individual interactions (e.g., viewed, clicked, manually created) along with data from those visualizations. More importantly, can learn relevant other users, even if generated completely different datasets. Experiments demonstrate effectiveness approach as it leads higher quality recommendations tailored specific intent preferences. To support research new problem, release our user-centric corpus consisting 17.4k users exploring 94k datasets 2.3 million attributes 32k user-generated