NLVIZ: Exploring Research Opportunities for Natural Language, Text, and Data Visualization

Natural language processing (NLP) has evolved as a promising field for helping users in their exploration and interaction with data for visual analysis tasks. The applications of NLP for supporting various aspects of the visual analysis workflow include employing text processing methods for data preparation, supporting interaction modalities to help people naturally “ask” questions of their data, and generating captions and textual summaries to help readers take away key information from the chart or dashboard, among others. As the field of NLP matures, computers now have an increased capability of interpreting natural language and engaging in conversations with people. But can NLP techniques and interactive visualizations work in concert to support an analytical conversation? If so, how? How can the increasing number of domain-specific corpora and their semantics be better utilized for supporting smarter data transformations and visualization defaults? How can we develop models to effectively summarize visualizations and highlight key data findings? Can text and charts be interactively linked to support interpretation and data-driven communication? Addressing these questions calls for research at the intersection of human-computer interaction, information visualization, and NLP, three fields with natural connections (pun intended) but rather infrequent meetings.

This workshop will assemble an interdisciplinary community that promotes collaboration across these fields, explore research opportunities and challenges, and establish an agenda for NLP research specifically for data visualization.