TaskFinder: A Semantics-Based Methodology for Visualization Task Recommendation

Abstract: Data visualization has entered the mainstream, and numerous visualization recommender systems have been proposed to assist visualization novices, as well as busy professionals, in selecting the most appropriate type of chart for their data. Given a dataset and a set of user-defined analytical tasks, these systems can make recommendations based on expert coded visualization design principles or empirical models. However, the need to identify the pertinent analytical tasks beforehand still exists and often requires domain expertise. In this work, we aim to automate this step with TaskFinder, a prototype system that leverages the information available in textual documents to understand domain-specific relations between attributes and tasks. TaskFinder employs word vectors as well as a custom dependency parser along with an expert-defined list of task keywords to extract and rank associations between tasks and attributes. It pairs these associations with a statistical analysis of the dataset to filter out tasks irrelevant given the data. TaskFinder ultimately produces a ranked list of attribute–task pairs. We show that the number of domain articles needed to converge to a recommendation consensus is bounded for our approach. We demonstrate our TaskFinder over multiple domains with varying article types and quantities

Teaser: The TaskFinder interface:

The TaskFinder interface takes in a dataset along with related textual documents via inputs on the top left and produces a set of recommended visualizations that can be used to explore the most important features of the data. Users can control the recommendations by selecting the tasks they are interested in and the visualizations they are familiar with via the task and visualizations panels on the left.

Paper: D. Coelho, B. Ghai, A. Krishna, M. Velez-Rojas, S. Greenspan, S. Mankovski, K. Mueller, “TaskFinder: A Semantics-Based Methodology for Visualization Task Recommendation,” Analytics, 3(3), 255-275. 2024. October 2023 PDF