Towards better analysis of machine learning models: A visual analytics perspective

Xiting Wang, Mengchen Liu, Jun Zhu

Visual Informatics · 2017 · 396 citations · 53 references

DOIFull text

Open access

Concepts

TL;DR

Interactive model analysis—using visualization to understand, diagnose, and refine machine learning models—is crucial for solving real-world AI and data mining problems, especially as big data analytics has expanded the range of such tasks. This paper provides a comprehensive analysis of interactive model analysis and discusses future research opportunities. The authors classify interactive model analysis work into three categories: understanding, diagnosis, and refinement. Each category is illustrated by recent influential studies.

Abstract

Interactive model analysis, the process of understanding, diagnosing, and refining a machine learning model with the help of interactive visualization, is very important for users to efficiently solve real-world artificial intelligence and data mining problems. Dramatic advances in big data analytics have led to a wide variety of interactive model analysis tasks. In this paper, we present a comprehensive analysis and interpretation of this rapidly developing area. Specifically, we classify the relevant work into three categories: understanding, diagnosis, and refinement. Each category is exemplified by recent influential work. Possible future research opportunities are also explored and discussed.

References

53