A spatial crowdsourcing quality control model is based on location privacy protection and decessor detection

Abstract

The invention provides a spatial crowdsourcing quality control model based on position privacy protection and decessor detection. The method specifically comprises the following steps: taking a crowdsourcing platform as a credible third party, firstly formulating a position privacy protection strategy of a worker according to a task issued by a requester, then carrying out k anonymity fuzzy on position privacy submitted by the worker, and transmitting protected position information to the requester; using the trained ELM to carry out common worker and spammer classification detection on the test data; and estimating an error rate by using an EM algorithm, and estimating the error rate by using an expected maximum algorithm. In one aspect, a space k is employed And the anonymous algorithmis used for protecting the location privacy of general space outsourcing personnel. In another aspect, a spoofer is detected using an ELM algorithm, and an error rate is estimated using an EM algorithm. The efficiency of the model is simulated by selecting different parameters, and the result shows that the space crowdsourcing model provided by the invention can ensure the quality of crowdsourcingprojects on the premise of protecting the privacy of employees.

Description

[0001] 【Technical field】 [0002] The invention relates to the technical field of spatial crowdsourcing, in particular to a spatial crowdsourcing quality control model based on location privacy protection and cheater detection. [0003] 【Background technique】 [0004] The basic concepts of spatial crowdsourcing are: [0005] (1) Task requester: The task requester registers and uses the space crowdsourcing platform to complete the design and release of space tasks, reject or receive answers from crowdsourcing workers, and organize a series of tasks such as crowdsourcing workers' answers. A task requester is usually defined as R=<L R ,T R >, where L R Indicates the location information of the task requester, T R Indicates a task posted by a task requester. [0006] (2) Spatial task: A spatial task is usually a special task with geographic location and time attributes, generally defined as a quadruple T=<L T , t begin , t end , P T >. Among them, L T Indicat...

AI Technical Summary

Problems solved by technology

Research conclusions on the binary model are not difficult to extend and apply to other task types

Deceptive workers (spammer) will not seriously submit low-quality task results in order to maximize their benefits; even diligent and conscientious workers may submit wrong results due to ability or accident, and there is a certain error rate

It is precisely because of the existence of spammer and error rate that the quality problems of crowdsourcing tasks are caused

Images