Crowd computing leverages the input of a crowd of online users to collaboratively solve complex problems. Humans and machines are seen as programmable computational units capable of executing tasks; both require seamless interactions even though their requirements may differ. In this paper, we aim at finding the optimal interplay between machine computing elements (MCEs) and human computing elements (HCEs). We propose an elastic computing framework, a system that leverages both MCEs and HCEs, through a uniformed interface, in order to optimally solve a complex task. Our work investigates three main research questions:
• RQ1 – When human and machine elements are capable of performing the same task, is there a general model that can define and evaluate their respective performance outcomes simultaneously?
• RQ2 - Can experimentation in a specific domain, such as face recognition, uncover the most appropriate, shared evaluative attributes that have cross-domain applicability?
• RQ3 - Can the specific performance variations in real-life experimentation enhance our overall understanding and ultimately lead to a more generalized elastic model?
To address these questions, we devise an elastic model and supporting architecture that governs the provisioning of MCE’s, HCE’s or both (hereinafter referred to as Elastic Computing Elements (ECEs)) given a specific task. The algorithms will use the elasticity model to ascertain the complexity of the task and the current operating environment and then proactively enforce constraints on its successful completion. The main idea is to extract the elasticity attributes of a certain task and use these attributes to orchestrate the use of HCEs and MCEs. Our work differs from other related projects in the fact that we focus on situations where the task for humans and for machine elements are exactly the same. Related projects focus on a variety of tasks for humans and machine elements where humans and machines might address distinct tasks.
This work is done in collaboration with M. Brian Blake research group at the University of Miami.
Julian Jarrett, Iman Saleh and M. Brian Blake, Rohan Malcolm, Sean Thorpe, Tyrone Grandison, "Combining Human and Machine Computing Elements for Analysis via Crowdsourcing", 10th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing, CollaborateCom'14, October 2014
Julian Jarrett, Iman Saleh, M.Brian Blake, Sean Thorpe, Tyrone Grandison, and Rohan Malcolm, "Mobile Services for Enhancing Human Crowdsourcing with Computing Elements", IEEE 3rd International Conference on Mobile Services, MS’14, June 2014