Design Patterns for Machine Learning-Based Systems With Humans in the Loop

Jakob Smedegaard Andersen

doi:10.1109/MS.2023.3340256

Design Patterns for Machine Learning-Based Systems With Humans in the Loop

2024-07-01

3

PRE

AI

J

Jakob Smedegaard Andersen *

W

Walid Maalej

DOI：10.1109/MS.2023.3340256

Original

Original request for help

Save

Abstract

En 中文

Human involvement in machine learning (ML) is a promising paradigm to overcome the limitations of purely automated predictions and improve the applicability of ML. We compile a catalog of design patterns to guide developers to select and implement human-in-the-loop solutions. OOP PATTERNS

Keywords:

Data models

Training

Data visualization

Training data

Task analysis

Software

Labeling

Design Patterns for Machine Learning-Based Systems With Humans in the Loop

Abstract

Journal

Organization