Home > Research > Publications & Outputs > GP-selector

Associated organisational unit


Text available via DOI:

View graph of relations

GP-selector: a generic participant selection framework for mobile crowdsourcing systems

Research output: Contribution to journalJournal articlepeer-review

<mark>Journal publication date</mark>1/05/2018
<mark>Journal</mark>World Wide Web
Issue number3
Number of pages24
Pages (from-to)759-782
Publication StatusPublished
Early online date2/09/17
<mark>Original language</mark>English


Participant selection is a common and crucial function for mobile crowdsourcing (MCS) systems or platforms. This paper introduces a generic framework, named GP-Selector, to handle the participant selection from MCS task creation time to runtime. Compared to existing approaches, ours has the following two unique features. 1) In the task creation time, it assists task creators with diverse levels of programming skills to define basic requirements of participant selection. 2) In the runtime, it adopts a two-phase selection process to select participants who not only meet the basic requirements but also are willing to accept the task. Specifically, we utilize the state-of-the-art techniques including ontology modeling, end-user programming and multi-classifier fusion to implement GP-Selector. We evaluate GP-Selector extensively in three aspects: the end-user task creation, the expressiveness of the core ontology model, and the willingness-based selection algorithm. The evaluation results demonstrate the usability and effectiveness.