AutoAIViz: Opening the Blackbox of Automated Artificial Intelligence with Conditional Parallel Coordinates

TL;DR

AutoAIViz is a visualization tool designed to make AutoAI processes transparent, helping data scientists understand and trust automated machine learning model generation.

Contribution

This paper introduces AutoAIViz, a novel visualization system that enhances transparency and user understanding of AutoAI model selection and generation processes.

Findings

AutoAIViz improves user understanding of AutoAI processes.

Users report increased trust in AutoAI outputs.

The system helps users complete data science tasks more effectively.

Abstract

Artificial Intelligence (AI) can now automate the algorithm selection, feature engineering, and hyperparameter tuning steps in a machine learning workflow. Commonly known as AutoML or AutoAI, these technologies aim to relieve data scientists from the tedious manual work. However, today's AutoAI systems often present only limited to no information about the process of how they select and generate model results. Thus, users often do not understand the process, neither do they trust the outputs. In this short paper, we provide a first user evaluation by 10 data scientists of an experimental system, AutoAIViz, that aims to visualize AutoAI's model generation process. We find that the proposed system helps users to complete the data science tasks, and increases their understanding, toward the goal of increasing trust in the AutoAI system.