Research On Collaborative Assignment Of Software Crowdsourcing Task Based On Fuzzy Measurement Of Workers' Capability

As a new type of swarm computing model,crowdsourcing has become one of the hot research topic in recent years,also software crowdsourcing is one of the typical application of crowdsourcing.Due to the diversity of its task requirements and the dynamics and uncertainties of workers' abilities,the workers' recommendation,ability assessment and task allocation in the process of software crowdsourcing face many challenges.According to the characteristics of software crowdsourcing task requirements,in order to improve the accuracy and objectivity of worker capacity measurement and ensure the final overall completion quality of the task set on the software crowdsourcing platform,this paper mainly focuses on collaborative recommendation of software crowdsourcing workers who match the task requirements,the fuzzy comprehensive measurement of the capabilities of software crowdsourcing workers,and the collaborative assignment method of crowdsourcing tasks based on role collaboration are studied:(1)According to the diversity characteristics of software crowdsourcing task requirements,matching numerical and text are given respectively the method of calculating the similarity of workers with high-demand needs selects high-similar workers by weighted synthesis,so as to recommend a set of workers that match the task requirements for a specific crowdsourcing task,thereby providing a source of candidate workers for subsequent task assignment.(2)Using interval number theory,a fuzzy measurement method for workers' comprehensive capabilities is proposed.It calculates a fuzzy evaluation matrix based on the indicators of workers' capabilities and task expectations,to carry out multidimensional evaluation of candidate workers for crowdsourcing tasks,and screen out high quality,high-capacity workers.Thus providing a basis for decision-making for the optimization method of collaborative research task allocation.(3)In order to ensure the final overall completion quality of the task set on the crowdsourcing platform,a collaborative distribution method of software crowdsourcing tasks is proposed.By applying Role-Based Collaboration(RBC)theory,the E-CARGO(Environments-Classes,Agents,Roles,Groups,Objects)model is used to model the crowdsourcing task allocation problem as a combination optimization problem based on role collaboration.Taking into account the cooperation conflict between workers and the compatibility between tasks,A solution method based on the IBM CPLEX algorithm optimization package is also given.Cases and experimental analysis verify the effectiveness and feasibility of the proposed method,thus providing strong support for ensuring the quality of task completion.This paper has carried out related research on software crowdsourcing worker recommendation,worker comprehensive ability assessment,and collaborative assignment of group tasks.The proposed method can provide new ideas and important references for software crowdsourcing task collaborative assignment research.