Anjana Arunkumar

Anjana Arunkumar

Postdoc @ Northeastern

About

I am a Postdoctoral Researcher in the Khoury Vis Lab at Northeastern University, working with Lace Padilla. I graduated from Arizona State University, supervised by Chris Bryan. I have also previously interned at the National Renewable Energy Laboratory (NREL).

My research lies at the intersection of visual cognition, natural language, and data visualization. I focus on creating inclusive data visualizations that democratize> both creation and consumption for underrepresented populations. Specifically, I examine the influence of multicultural and linguistic factors, as well as cognitive biases, on how people engage with and make decisions using visualizations. Additionally, I have extensively developed visual analysis platforms to support non-expert users in developing large language models and engineering prompts in natural language processing.

Looking ahead, I am excited to expand my research at the intersection of visualization, decision science, and natural language processing, to develop data communication paradigms such as patient decision aids, personalized learning pathways, and risk management frameworks, that empower low data literacy populations to take informed decisions.

I am always open to discussing new ideas and potential collaborations. Please feel free to email me to chat!

I am currently on the job market, and looking for tenure track positions!

News

July 2024Our submission Mind Drifts, Data Shifts: Utilizing Mind Wandering to Track the Evolution of User Experience with Data Visualizations has been accepted as a full paper in IEEE VIS 2024
June 2024Excited to join Northeastern University as a Post-Doctoral researcher working with Lace Padilla!
May 2024Officially a Doctor! Graduated from Arizona State University!
August 2023Our submission Comparing Collaborative Visualization Behaviors in Desktop and Augmented Reality Environments has been accepted as a full paper in VRST 2023
July 2023Our submission Image or Information? Examining the Nature and Impact of Visualization Perceptual Classification has been accepted as a full paper in IEEE VIS 2023
July 2023Our submissions Examining the Effect of Visualization Proxemics on Decision Making using Touchscreen Dashboards and Automated Construction of Business Intelligence Dashboards Using Artificial Intelligence Tools have been accepted to Decision Sciences Institute Annual Conference 2023
March 2023Our submission LINGO : Visually Debiasing Natural Language Instructions to Support Task Diversity has been accepted as a full paper in Eurovis 2023
Jan 2023Our submission Real-Time Visual Feedback to Guide Benchmark Creation: A Human-and-Metric-in-the-Loop Workflow has been accepted as a full paper in EACL 2023
Dec 2022Our submissions Examining The Effects of Visual Redundancy on Understanding Uncertainty Representations in Schedule Risk Analysis, and Examining the Effect of Reconfiguration Interactions on Understanding Uncertainty in Supply Chain Disruption have been accepted to Southwest Decision Sciences Institute Annual Conference 2023
Nov 2022Our submissions Examining the Effect of Visual Redundancy on the Understanding of Visualizations in Procurement and Sourcing, and Examining the Effect of Reconfiguration Interactions on the Understanding of Anomalies in Supply Chain Management have been accepted to Decision Sciences Institute Annual Conference 2022
Oct 2022Our work PMU Tracker: A Visualization Platform for Epicentric Event Propagation Analysis in the Power Grid has been accepted as a full paper in IEEE VIS 2022
Apr 2022Our work Benchmarking generalization via in-context instructions on 1,600+ language tasks has been accepted as a full paper in EMNLP 2022
Apr 2022Our work PMUVis: A Large Scale Platform to Assist Power System Operators in a Smart Grid has been published in IEEE Computer Graphics and Applications 2022
Apr 2022Our submissions Evaluation of Visual Uncertainty Representations in Schedule Risk Analysis, and Modelling the Complexity of Risk Centered Supply Chain Network Visualization have been accepted to Midwest Decision Sciences Institute Annual Conference 2022
Nov 2021Our submissions Impact of Uncertainty Representation on Decision Making for Business Network Visualization, Bayesian Modelling of Uncertainty in Schedule Risk Analysis using Gantt Charts, and Modelling the Complexity of Supply Chain Network Visualization have been accepted to Decision Sciences Institute Annual Conference 2021
Oct 2021Our work Bayesian Modelling of Alluvial Diagram Complexity has been accepted as a short paper in IEEE VIS 2021
Mar 2021Our submission Strategy-Enterprise System Alignment and Firm Performance has been accepted to Northeast Decision Sciences Institute Annual Conference 2021
Feb 2021Our work How Robust are Model Rankings : A Leaderboard Customization Approach for Equitable Evaluation has been accepted as a full paper in AAAI 2021
Oct 2020Our submission TotemFinder: A Visual Analytics Approach for Image-based Key Players Identification won an honorable mention in the VAST Challenge, in IEEE Vis 2020
Oct 2020Our submission A Visual Exploration of Fair Evaluation for ML - Bridging the Gap Between Research and the Real World has been accepted to the VisxAI workshop, in IEEE Vis 2020
July 2020Our submissions DQI: A Guide to Benchmark Evaluation and Our Evaluation Metric Needs an Update to Encourage Generalization have been accepted to the Uncertainty & Robustness in Deep Learning Workshop, in ICML 2020
Dec 2020Our submission VAIDA: An Educative Benchmark Creation Paradigm using Visual Analytics for Interactively Discouraging Artifacts has been accepted to the Crowd Science Workshop: Remoteness, Fairness, and Mechanisms as Challenges of Data Supply by Humans for Automation Workshop, in NeurIPS 2020
Dec 2020Our submission Front Contribution instead of Back Propagation has been accepted to the Beyond Backpropagation: Novel Ideas for Training Neural Architectures, in NeurIPS 2020
Dec 2020Our submission Real-Time Visual Feedback for Educative Benchmark Creation: A Human-and Metric-in-the-Loop Workflow has been accepted to the HAMLETS (Human And Machine in-the-Loop Evaluation and Learning Strategies) Workshop, in NeurIPS 2020
Dec 2020Our submission Is High Quality Data All You Need? has been accepted to The pre-registration experiment: an alternative publication model for machine learning research Workshop, in NeurIPS 2020
Nov 2020Our submissions Investigating the Influence of Top Management Support in Systemic Enterprise System Deployments, Examining Data Visualization's Impact on Decision Making, and Investigating the Influence of Chart Embellishment on Decision Making have been accepted to Decision Sciences Institute Annual Conference 2020
Nov 2019Our submissions Abstractive Summarization Through Sentiment Analysis Of User Product Reviews - An RNN Approach, Integration Mechanisms, Absorptive Capacity, & Enterprise Systems, Relational Capital & Enterprise Systems, and The Role of Consultants in Enterprise Systems have been accepted to Decision Sciences Institute Annual Conference 2019




Selected Publications

Anjana Arunkumar, Lace Padilla, and Chris Bryan. Mind Drifts, Data Shifts: Utilizing Mind Wandering to Track the Evolution of User Experience with Data Visualizations accepted to IEEE VIS 2024.
Paper | Supplementary Materials
Anjana Arunkumar, Lace Padilla, Gi-Yeul Bae, and Chris Bryan. Image or Information? Examining the Nature and Impact of Visualization Perceptual Classification accepted to IEEE VIS 2023.
Paper | Supplementary Materials
Anjana Arunkumar, Shubham Sharma, Rakhi Agrawal, Sriram Chandrasekaran, and Chris Bryan. LINGO: Visually Debiasing Natural Language Instructions to Support Task Diversity Computer Graphics Forum, 42.3, 2023.
DOI | Paper | Video
Anjana Arunkumar, Swaroop Mishra, Bhavdeep Sachdeva, Chitta Baral, and Chris Bryan. Real-Time Visual Feedback to Guide Benchmark Creation: A Human-and-Metric-in-the-Loop Workflow in press, accepted to EACL 2023.
DOI | Paper
Anjana Arunkumar, Andrea Pinceti, Lalitha Sankar, and Chris Bryan. PMU Tracker: A Visualization Platform for Epicentric Event Propagation Analysis in the Power Grid in press, accepted to IEEE Vis 2022.
DOI | Paper
Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Anjana Arunkumar, and 35 others. Benchmarking generalization via in-context instructions on 1,600+ language tasks in press, accepted to EMNLP 2022.
arXiv preprint
Anjana Arunkumar, Shashank Ginjpalli, and Chris Bryan. Bayesian Modelling of Alluvial Diagram Complexity accepted to 2021 IEEE Visualization Conference (VIS), pp 51-55, 2021.
DOI | Paper
Swaroop Mishra and Anjana Arunkumar. How Robust are Model Rankings : A Leaderboard Customization Approach for Equitable Evaluation accepted to 2021 AAAI Conference on Artificial Intelligence, 35(15), 13561-13569.
DOI | Paper