Towards a Problem-Oriented Domain Adaptation Framework for Machine Learning

TL;DR

This paper introduces a problem-oriented framework for domain adaptation in machine learning, helping researchers and practitioners identify suitable approaches based on specific scenarios and guiding effective application.

Contribution

It develops a novel, structured framework that categorizes domain adaptation scenarios and offers practical guidelines, validated through multiple evaluations including real-world data and user studies.

Findings

Framework effectively captures diverse domain adaptation problems

Provides clear guidance for selecting adaptation approaches

Validated through artificial, real-world datasets, and user study

Abstract

Domain adaptation is a sub-field of machine learning that involves transferring knowledge from a source domain to perform the same task in the target domain. It is a typical challenge in machine learning that arises, e.g., when data is obtained from various sources or when using a data basis that changes over time. Recent advances in the field offer promising methods, but it is still challenging for researchers and practitioners to determine if domain adaptation is suitable for a given problem -- and, subsequently, to select the appropriate approach. This article employs design science research to develop a problem-oriented framework for domain adaptation, which is matured in three evaluation episodes. We describe a framework that distinguishes between five domain adaptation scenarios, provides recommendations for addressing each scenario, and offers guidelines for determining if a problem falls into one of these scenarios. During the multiple evaluation episodes, the framework is tested on artificial and real-world datasets and an experimental study involving 100 participants. The evaluation demonstrates that the framework has the explanatory power to capture any domain adaptation problem effectively. In summary, we provide clear guidance for researchers and practitioners who want to employ domain adaptation but lack in-depth knowledge of the possibilities.