Facilitating Machine Learning Model Comparison and Explanation through a Radial Visualisation

Fang Chen,Weidong Huang,Jianlong Zhou

doi:10.3390/en14217049

Abstract

Building an effective Machine Learning (ML) model for a data set is a difficult task involving various steps. One of the most important steps is to compare a substantial amount of generated ML models to find the optimal one for deployment. It is challenging to compare such models with a dynamic number of features. Comparison is more than only finding differences of ML model performance, as users are also interested in the relations between features and model performance such as feature importance for ML explanations. This paper proposes RadialNet Chart, a novel visualisation approach, to compare ML models trained with a different number of features of a given data set while revealing implicit dependent relations. In RadialNet Chart, ML models and features are represented by lines and arcs, respectively. These lines are generated effectively using a recursive function. The dependence of ML models with a dynamic number of features is encoded into the structure of visualisation, where ML models and their dependent features are directly revealed from related line connections. ML model performance information is encoded with colour and line width in RadialNet Chart. Taken together with the structure of visualisation, feature importance can be directly discerned in RadialNet Chart for ML explanations. Compared with other commonly used visualisation approaches, RadialNet Chart can help to simplify the ML model comparison process with different benefits such as the following: more efficient in terms of helping users to focus their attention to find visual elements of interest and easier to compare ML performance to find optimal ML model and discern important features visually and directly instead of through complex algorithmic calculations for ML explanations.

Highlights

IntroductionTogether with the dramatic advances in Machine Learning (ML), obtaining insights from these “Big Data” and data analytics-driven solutions are increasingly in demand for different purposes
We have witnessed a rapid boom of data in recent years from various fields such as infrastructure, transport, energy, health, education, telecommunications, and finance.Together with the dramatic advances in Machine Learning (ML), obtaining insights from these “Big Data” and data analytics-driven solutions are increasingly in demand for different purposes
We explore an approach based on the structure of visualisation in addressing challenges of comparison ML models with a dynamic number of features: while height information of bars and lines in commonly used visualisation approaches only encode one-dimensional information in a 2-dimensional (2D) space, it is possible to encode ML

Summary

Introduction

Together with the dramatic advances in Machine Learning (ML), obtaining insights from these “Big Data” and data analytics-driven solutions are increasingly in demand for different purposes. Taking a data set with multiple features for ML training as an example, multiple features can be grouped differently as the input for an ML algorithm to train different ML models. Given a fixed number of features, it is possible to use different features and their groups to train machine learning algorithms resulting in various machine learning models. Obtaining optimal results out of machine learning models requires truly understanding all models. Each data set with a large number of features can have hundreds or even thousands of ML models, rendering it nearly impossible to understand all models based on different feature groups in an intuitive fashion. This section investigates various visualisations from the perspectives of multi-attribute data visualisation, visualisation in explanation of machine learning, and comparison visualisation in order to demonstrate the state-of-art approaches and challenges for the comparison of machine learning models with visualisation

Objectives

Methods

Results

Discussion

Conclusion