Research On Re-assignment Problem Based On E-CARGO Model

The E-CARGO model is an extremely flexible role-based cooperative learning system model that has emerged in recent years.Using the E-CARGO model can solve many problems that cannot be solved in the traditional cooperative science.In terms of allocation problems and re-assignment issues,E-CARGO model also has great advantages.Based on the original allocation problem of the E-CARGO model,this paper formalizes the re-assignment problem,formalizes the re-assignment problem as a constrained optimization problem,and proposes two different formalized optimization.The approach will have different effects on the final in the cooperation.For the algorithm to solve the re-assignment problem,this paper proposes two algorithms,one is to change the re-assignment problem to the initial allocation,and then use the Hungarian algorithm to calculate.Hungary can obtain an exact solution but does not fully utilize the results of the initial distribution.Another algorithm is based on the greedy strategy to solve the re-assignment problem approximation algorithm,greedy algorithm makes full use of the results of the initial allocation,the time complexity is low,but does not guarantee that the exact result.This article tests the two proposed algorithms by setting different conditions.These conditions include the distribution of the matrix in the initial state,divided into a uniform distribution,a normal distribution and a skewness distribution.For the number of people who have allocated tasks in cooperation,different data are also set,and there are 1%,5%,10%,and 30%.A large number of experiments were conducted and a large amount of data on re-assignment was obtained.The experimental data was analyzed in detail,including qualitative analysis and quantitative analysis.In the case of any distribution,when the number of missing persons is less than 10%,the score difference between the Hungarian algorithm and the greedy algorithm is less than 0.1 points,and when the number of missing persons is more than 10%and the ratio of the number of people to the number of tasks exceeds 1/2,The score difference between the Hungarian algorithm and the greedy algorithm is less than 1 point.