Investigating Aspects of Platform Growth on Crowdsource Driven Recommendation Services

With the advent of the Internet, crowdsourcing platforms are a viable option for the solicitation of labor and distribution of tasks, online. As platforms develop, expand, and evolve, they employ various approaches to assist job owners and workers in searching for suitable tasks where workers' skillsets and job requirements are aligned and compatible. Recommender systems provide a suitable candidate for such purposes and can assist in worker-job matching. In this paper, through experimentation, we investigate the effect that growth in the labor force and job catalog has on the times to produce recommendations, the jobs recommended to workers and the precision of our collaborative-filtering-oriented bottom-up recommender.