Use this URL to cite or link to this record in EThOS:
Title: Interactive machine learning for user-innovation toolkits : an action design research approach
Author: Bernardo, Francisco
ISNI:       0000 0004 9349 8732
Awarding Body: Goldsmiths, University of London
Current Institution: Goldsmiths College (University of London)
Date of Award: 2020
Availability of Full Text:
Access from EThOS:
Access from Institution:
Machine learning offers great potential to developers and end users in the creative industries. However, to better support creative software developers' needs and empower them as machine learning users and innovators, the usability of and developer experience with machine learning tools must be considered and better understood. This thesis asks the following research questions: How can we apply a user-centred approach to the design of developer tools for rapid prototyping with Interactive Machine Learning? In what ways can we design better developer tools to accelerate and broaden innovation with machine learning? This thesis presents a three-year longitudinal action research study that I undertook within a multi-institutional consortium leading the EU H2020 -funded Innovation Action RAPID-MIX. The scope of the research presented here was the application of a user-centred approach to the design and evaluation of developer tools for rapid prototyping and product development with machine learning. This thesis presents my work in collaboration with other members of RAPID-MIX, including design and deployment of a user-centred methodology for the project, interventions for gathering requirements with RAPID-MIX consortium stakeholders and end users, and prototyping, development and evaluation of a software development toolkit for interactive machine learning. This thesis contributes with new understanding about the consequences and implications of a user-centred approach to the design and evaluation of developer tools for rapid prototyping of interactive machine learning systems. This includes 1) new understanding about the goals, needs, expectations, and challenges facing creative machine-learning non-expert developers and 2) an evaluation of the usability and design trade-offs of a toolkit for rapid prototyping with interactive machine learning. This thesis also contributes with 3) a methods framework of User-Centred Design Actions for harmonising User-Centred Design with Action Research and supporting the collaboration between action researchers and practitioners working in rapid innovation actions, and 4) recommendations for applying Action Research and User-Centred Design in similar contexts and scale.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available