首页|How to select crowdsourcing teams with limited information? A heterogeneous information network embedding approach
How to select crowdsourcing teams with limited information? A heterogeneous information network embedding approach
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Springer Nature
Crowdsourcing has become a widely accepted approach to leverage crowds to solve business problems, and how to find proper solvers has been widely discussed. Existing studies mainly focus on matching tasks and individuals for simple tasks. However, the increasing complexity of projects calls for the crowdsourcing team selection, whereas the difficulty lies in the limited background information of team members which leads to data sparsity and cold-start problems. Motivated by this problem, we develop CT-HIN, a heterogeneous information network embedding method to evaluate skill matching and communication by similarity searching based on the pair-wise random walk model from multi-dimension. An empirical evaluation with input data collected from a real-world crowdsourcing platform is conducted to justify our proposed approach.
CrowdsourcingHeterogeneous information networkProject team selection
Yuanyuan Lai、Min Li、Junjun Liu、Huimin Liu
展开 >
School of Management and Engineering, Nanjing University, 22 Hankou Road, Gulou District, Nanjing 210093, China
NETL
NSTL
Springer Nature
