Research On User Satisfaction Evaluation Of Taxi Software Based On SEM

With the development of the mobile terminal equipment,a large number of APP software has been greatly promoted and developed,so that our daily life has been greatly convenient.Taxi software as a convenient public travel,sharing the typical representative of the economy,greatly easing people's daily travel "taxi difficult" problem.But with the reduction of preferential taxi software platform,and the introduction of the new deal makes the taxi difficult and other issues highlighted,the impact on the user's satisfaction is great,affecting the normal development of mobile taxi software.This paper analyzes the concept of taxi software and taxi software user satisfaction research status at home and abroad.And integrated technology receiving model,finally determine the taxi software user satisfaction factors,build a comprehensive taxi software user satisfaction model.The questionnaire was used to collect the data of customer satisfaction.The reliability and validity of the data were analyzed to verify the stability and accuracy of the data,and the structural equation was constructed to test the hypothesis.Finally determine the software information,software design,map services,vehicle services,orders service,account security,vehicle safety,information security and billing,nine taxi software user satisfaction factors.Through empirical analysis to determine the impact of taxi software satisfaction factors as the input variables based on the principal component analysis of the limit learning machine model.In order to make the score more close to the actual situation,the data of the questionnaire were pre processed,and the average of the indicators of each factor was taken as the evaluation index of taxi software user satisfaction.Using the method of analysis of the factors affecting the high correlation dimension of principal components,the reduction obtained five factors and vehicle service as the simulation experiment of the data input variables of extreme learning machine algorithm,with high accuracy of the simulation results.Finally,on the basis of empirical research and simulation experiments,the paper analyzes the results from two angles of policy supervision and operator management,and puts forward relevant opinions on the analysis results.