Electronic Theses and Dissertations

Date

2019

Date of Award

2019

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Computer Science

Committee Chair

Sajjan Shiva

Committee Member

Deepak Venugopal

Committee Member

Scott Fleming

Committee Member

Mark Gillenson

Abstract

Crowdsourcing is an approach whereby requesters call for workers with different capabilities to process a task for monetary reward. The emergence of crowdsourcing has drawn increasing attention in recent years as a revolutionary phenomenon. Although crowdsourcing is still considered a developing approach, early signs expectations are promising. The advantage of crowdsourcing resides in its ability to facilitate access to diverse skilled workers to process the outsourced tasks at reduced time and cost. Moreover, crowdsourcing helps reduce unemployment by offering on-demand employment opportunities. With the vast amount of tasks posted every day, satisfying the workers, requesters, and service providers who are the stakeholders of any crowdsourcing system is critical to its success. To achieve this, the system should address three objectives: (1) match the worker with suitable tasks that fit the workers interests and skills and raise the workers rewards and rating, (2) give the requester qualified solutions at lower cost and time and raise the employer rating, and (3) raise the task acceptance rate, which will raise the aggregated commissions accordingly. For these objectives, we present a mechanism design capable of achieving holistic satisfaction using a multi-objective recommendation system. The proposed model is designed as an interactive system where every worker and employer could set the parameters that meet their goals. In contrast, all previous crowdsourcing recommendation systems have been designed to address one stakeholder. Moreover, no previous crowdsourcing recommendation systems have considered the other partys behavior to provide more qualified recommendations as we have done. Furthermore, we conducted a survey of one type of macrotask, namely a cloud application development to emphasize the importance of using crowdsourcing for macrotask. We identified its challenges and explored the facilities that support addressing these challenges. We also reviewed two widespread existing approaches for software development crowdsourcing and propose a novel approach. Additionally, we evaluated our proposed approach for its ability to address these challenges and provide future adopters with a list of attributes to assist in choosing the right crowdsourcing service. Finally, we evaluated our model with synthesized datasets. The experimental simulation showed the superiority of the proposed model compared with two other baseline models.

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest

Share

COinS