A Review on Human-AI Interaction in Machine Learning and Insights for Medical Applications.

Hadi Akbarzadeh Khorshidi,Uwe Aickelin,Mansoureh Maadi

doi:10.3390/ijerph18042121

Hadi Akbarzadeh Khorshidi, Uwe Aickelin + Show 1 more

Open Access

https://doi.org/10.3390/ijerph18042121

Copy DOI

Abstract

Objective: To provide a human–Artificial Intelligence (AI) interaction review for Machine Learning (ML) applications to inform how to best combine both human domain expertise and computational power of ML methods. The review focuses on the medical field, as the medical ML application literature highlights a special necessity of medical experts collaborating with ML approaches. Methods: A scoping literature review is performed on Scopus and Google Scholar using the terms “human in the loop”, “human in the loop machine learning”, and “interactive machine learning”. Peer-reviewed papers published from 2015 to 2020 are included in our review. Results: We design four questions to investigate and describe human–AI interaction in ML applications. These questions are “Why should humans be in the loop?”, “Where does human–AI interaction occur in the ML processes?”, “Who are the humans in the loop?”, and “How do humans interact with ML in Human-In-the-Loop ML (HILML)?”. To answer the first question, we describe three main reasons regarding the importance of human involvement in ML applications. To address the second question, human–AI interaction is investigated in three main algorithmic stages: 1. data producing and pre-processing; 2. ML modelling; and 3. ML evaluation and refinement. The importance of the expertise level of the humans in human–AI interaction is described to answer the third question. The number of human interactions in HILML is grouped into three categories to address the fourth question. We conclude the paper by offering a discussion on open opportunities for future research in HILML.

Highlights

Published: 22 February 2021Recent impressive developments of Machine Learning (ML) methods have created many star applications in various fields of science and industry
“Why should humans be in the loop?”, “Where does human–Artificial Intelligence (AI) interaction occur in the ML processes?”, “Who are the humans in the loop?” and “How do humans interact with ML in Human-In-the-Loop ML (HILML)?” are questions that by answering them, can guide researchers to comprehend strengths, weaknesses and application scenarios of human–AI interaction in ML applications
The key contributions of this paper are threefold in that we conduct a comprehensive review on human roles in human–AI interaction in ML applications, we provide an overview for the current research in human–AI interaction area, and we discuss the role of humans in collaboration with ML

Summary

Introduction

Recent impressive developments of Machine Learning (ML) methods have created many star applications in various fields of science and industry. ML methods as a part of Artificial Intelligence (AI) are tools that automatically learn from sample data (training data) and provide insightful knowledge. Different types of ML methods can be categorised into supervised, unsupervised and semi-supervised ML methods. Supervised ML methods use training data with labels and assign labels to all feasible inputs. Unsupervised ML methods use a training data set without labels and group data by finding similarities among training data. Semi-supervised ML methods use a training data set consisting of labelled data and unlabelled data (mostly unlabelled data) and are the combination of supervised and unsupervised ML methods.

Methods

Results

Conclusion