Low cost scheme for Crowdsourced Top-k Queries
Crowdsourced queries processing is a new processing technology.It is invented through human intelligence.Crowdsourced queries have solved a lot of challenging problems.The basic idea is divided a problem into many small tasks (e.g., pairwise comparison) and then published them to be answered by people.After collecting peoples answers, propose the solution of the problem.It is to solve a problem that is difficult for people to calculate.This paper deals with the comparison.It based on the collected response after the assignment of crowdsourcing.Finally, the top-k can be found with the lowest monetary cost.The feedback of each answer is the preference score for each object.In this paper, we adopt a new sort method to obtain top-k.It can effectively reduce the cost.We define this scheme as TLS.TLS chooses pairwise preference judgment.It can reduce the workload of comparison.We can see from the experiment that the select partition sort algorithm is superior to other algorithms.And the TLS can reduce the total cost significantly.
select partition sort monetary cost top-k
Zongmin Cui Yu Gao Zhuolin Mei Bin Wu Guangyong Gao Zhiqiang Zhao
School of Information Science and Technology Jiujiang University,Jiujiang, Jiangxi, China
国际会议
郑州
英文
382-390
2018-09-21（万方平台首次上网日期，不代表论文的发表时间）