TY - JOUR
T1 - Efficient and Flexible Crowdsourcing of Specialized Tasks with Precedence Constraints
AU - Chatterjee, Avhishek
AU - Borokhovich, Michael
AU - Varshney, Lav R.
AU - Vishwanath, Sriram
N1 - Funding Information:
Manuscript received April 5, 2016; revised February 21, 2017 and November 29, 2017; accepted January 31, 2018; approved by IEEE/ACM TRANSACTIONS ON NETWORKING Editor A. Eryilmaz. Date of publication March 21, 2018; date of current version April 16, 2018. This work was supported by the National Science Foundation under Grant CCF-1623821. Part of the material in this paper was presented at IEEE INFOCOM 2016, San Francisco, CA, USA [1]. (Corresponding author: Avhishek Chatterjee.) A. Chatterjee is with the Electrical Engineering Department, IIT Madras, Chennai 600036, India (e-mail: avhishek@ee.iitm.ac.in ).
Publisher Copyright:
© 1993-2012 IEEE.
PY - 2018/4
Y1 - 2018/4
N2 - Many companies now use crowdsourcing to leverage external as well as internal crowds to perform specialized work, and so methods of improving efficiency are critical. Tasks in crowdsourcing systems with specialized work have multiple steps and each step requires multiple skills. Steps may have different flexibilities in terms of obtaining service from one or multiple agents due to varying levels of dependency among parts of steps. Steps of a task may have precedence constraints among them. Moreover, there are variations in loads of different types of tasks requiring different skill sets and availabilities of agents with different skill sets. Considering these constraints together necessitate the design of novel schemes to allocate steps to agents. In addition, large crowdsourcing systems require allocation schemes that are simple, fast, decentralized, and offer customers (task requesters) the freedom to choose agents. In this paper, we study the performance limits of such crowdsourcing systems and propose efficient allocation schemes that provably meet the performance limits under these additional requirements. We demonstrate our algorithms on data from a crowdsourcing platform run by a nonprofit company and show significant improvements over current practice.
AB - Many companies now use crowdsourcing to leverage external as well as internal crowds to perform specialized work, and so methods of improving efficiency are critical. Tasks in crowdsourcing systems with specialized work have multiple steps and each step requires multiple skills. Steps may have different flexibilities in terms of obtaining service from one or multiple agents due to varying levels of dependency among parts of steps. Steps of a task may have precedence constraints among them. Moreover, there are variations in loads of different types of tasks requiring different skill sets and availabilities of agents with different skill sets. Considering these constraints together necessitate the design of novel schemes to allocate steps to agents. In addition, large crowdsourcing systems require allocation schemes that are simple, fast, decentralized, and offer customers (task requesters) the freedom to choose agents. In this paper, we study the performance limits of such crowdsourcing systems and propose efficient allocation schemes that provably meet the performance limits under these additional requirements. We demonstrate our algorithms on data from a crowdsourcing platform run by a nonprofit company and show significant improvements over current practice.
KW - Crowdsourcing
KW - human resource management
KW - scheduling algorithms
UR - http://www.scopus.com/inward/record.url?scp=85044269234&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044269234&partnerID=8YFLogxK
U2 - 10.1109/TNET.2018.2811736
DO - 10.1109/TNET.2018.2811736
M3 - Article
AN - SCOPUS:85044269234
SN - 1063-6692
VL - 26
SP - 879
EP - 892
JO - IEEE/ACM Transactions on Networking
JF - IEEE/ACM Transactions on Networking
IS - 2
ER -