Towards a Distributed Worker-Job Matching
Architecture for Crowdsourcing

[pdf-embedder url=”https://wellapets.com/wp-content/uploads/2019/06/Towards-a-Distributed-Worker-Job-Matching.pdf” title=”Towards a Distributed Worker-Job Matching”]

Abstract— While the crowdsourcing paradigm facilitates the use of human-enacted resources from large groups of individuals, matching workers with jobs is limited by the need for these potential workers to proactively subscribe to various networks.
This subscription phase is part of an “open call model” that reduces the ability for crowdsourcing platforms to scale or retain crowd-oriented workers. Leveraging collaborative filtering techniques, in this paper, we propose an alternative model that seeks to address the issue through a recommendation technique and system that exploits a push-pull model.

Crowdsourcing [1], through the advent of the Internet and Web 2.0 technologies, has provided a new paradigm for employment, to harness mass human computation and has given new avenues for businesses and researchers to quickly distribute work across a global cross-section of potential workers [1][2]. As defined by Howe [3], the paradigm entails an open call model via the Internet to anonymous individuals to solicit services for work, usually at a much cheaper cost than traditional outsourcing [4]. Labor markets [5] such as Amazon Mechanical Turk, Microworkers and UpWork (formerly ODesk) exhaustively use this model. The model however has a significant deficiency [6]; that is the challenge of attracting and maintaining a crowd [7][8]. Via the open call model, tasks requiring human intelligence or HITs are posted for workers to accept relevant task offerings. Those being exposed to the offering are typically members or subscribers of a labor market’s community pool or workers [6]. However, there exist massive crowds of potential workers outside of the subscribed labor market pools and currently the open call model is not capable of leveraging this untapped pool of workers [6][9].
In this paper, augmented by collaborative filtering, we propose a service-oriented architecture based on an open pushpull worker-job matching model capable of harnessing the wisdom and labor potential of diverse communities external to current labor markets. The architecture incorporates transactional web services that implement services pertinent to crowdsourcing including recruitment, job allocation and compensation. We continue by reviewing the open call model including current recruitment strategies and techniques, and worker-job recommender strategies within and external to the paradigm of crowdsourcing. We follow by presenting our proposed collaborative filtering augmented architecture for open push-pull worker-job matching.