Table2Charts: Recommending Charts by Learning Shared Table Representations

Abstract

It is common for people to create different types of charts to explore a multi-dimensional dataset (table). However, to recommend commonly composed charts in real world, one should take the challenges of efficiency, imbalanced data and table context into consideration. In this paper, we propose Table2Charts framework which learns common patterns from a large corpus of (table, charts) pairs. Based on deep Q-learning with copying mechanism and heuristic searching, Table2Charts does table-to-sequence generation, where each sequence follows a chart template. On a large spreadsheet corpus with 165k tables and 266k charts, we show that Table2Charts could learn a shared representation of table fields so that recommendation tasks on different chart types could mutually enhance each other. Table2Charts outperforms other chart recommendation systems in both multi-type task (with doubled recall numbers R@3=0.61 and R@1=0.43) and human evaluations.

9 + 2(appendix) pages, accepted by KDD'21 conference

References in corpus (1)

• Embedding Projector: Interactive Visualization and Interpretation of Embeddings

Cited by in corpus (10)

• Towards Natural Language Interfaces for Data Visualization: A Survey
• ChartGPT: Leveraging LLMs to Generate Charts from Abstract Natural Language
• LightVA: Lightweight Visual Analytics with LLM Agent-Based Task Planning and Execution
• HAIChart: Human and AI Paired Visualization System
• Visual Data Analysis with Task-based Recommendations
• RL-LABEL: A Deep Reinforcement Learning Approach Intended for AR Label Placement in Dynamic Scenarios
• Agnostic Visual Recommendation Systems: Open Challenges and Future Directions
• Reinforced Labels: Multi-Agent Deep Reinforcement Learning for Point-Feature Label Placement
• Adversarial Attacks on Machine Learning-Aided Visualizations
• NoteFlow: Leveraging Charts as Sight Glasses for Consistent and Continuous Data Flow Tracing