The Challenges of Knowledge Combination in ML-based Crowdsourcing – The ODF Killer Shrimp Challenge using ML and Kaggle

Date

2021-01-05

Contributor

Advisor

Department

Instructor

Depositor

Speaker

Researcher

Consultant

Interviewer

Narrator

Transcriber

Annotator

Journal Title

Journal ISSN

Volume Title

Publisher

Volume

Number/Issue

Starting Page

4930

Ending Page

Alternative Title

Abstract

Organizations are increasingly using digital technologies, such as crowdsourcing platforms and machine learning, to tackle innovation challenges. These technologies often require the combination of heterogeneous technical and domain-specific knowledge from diverse actors to achieve the organization’s innovation goals. While research has focused on knowledge combination for relatively simple tasks on crowdsourcing platforms and within ML-based innovation, we know little about how knowledge is combined in emerging innovation approaches incorporating ML and crowdsourcing to solve domain-specific innovation challenges. Thus, this paper investigates the following: What are the challenges to knowledge combination in domain-specific ML-based crowdsourcing? We conducted a case study of an environmental challenge – how to use ML to predict the spread of a marine invasive species, led by the Swedish consortium, Ocean Data Factory Sweden using the crowdsourcing platform Kaggle. After discussing our results, we end the paper with recommendations on how to integrate crowdsourcing into domain-specific digital innovation processes.

Description

Keywords

Emerging Trends in Crowd Science, crowdsourcing, digital innovation, knowledge combination, machine learning, problem-solving

Citation

Extent

10 pages

Format

Geographic Location

Time Period

Related To

Proceedings of the 54th Hawaii International Conference on System Sciences

Related To (URI)

Table of Contents

Rights

Attribution-NonCommercial-NoDerivatives 4.0 International

Rights Holder

Local Contexts

Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.