Font Size: a A A

Analysis Of APP User Characteristics Of Train Ticket

Posted on:2020-11-22Degree:MasterType:Thesis
Country:ChinaCandidate:Y LuoFull Text:PDF
GTID:2427330596481765Subject:Master of Applied Statistics
Abstract/Summary:PDF Full Text Request
Nowadays,the penetration rate of network users has reached a new high,and the 5G era is about to open.People begin to consume spiritual culture,and experience has become the highest level of consumption field pursued by people.Every holiday and other travel peak,"ticket grabbing" has become a compulsory course for many travelers.Qunar,Ctrip,Gaotieguanjia and other major travel enterprises have emerged ticket grabbing services.Wechat small procedures also can invite friends to grab tickets to accelerate.With the increase of mobile devices and the increase of network connection speed,the development of mobile phone mainly depends on APP products.The APP market is highly competitive.For users,software for purchasing train tickets is not indispensable.In the case that software programming techniques tend to be homogeneous,there is a strong substitution in the intermediary role of software.Making certain unique features of the software itself and the unique user experience become the competitive soft power of major businesses,which is an important indicator to measure the success of a product.Therefore,we focuses on the user experience characteristics of urban residents in Wuhan about train ticketing software.Firstly,we takes Wuhan urban residents as the research object,and obtains data through questionnaire survey,which provides data basis for the follow-up understanding of user characteristics of ticketing software.The survey process is divided into two stages: pre-survey and formal survey.Through pre-survey,users' usage is preliminarily understood,and reliability and validity tests are combined to measure the reliability of the questionnaire in order to carry out formal investigation effectively.The formal survey uses multi-stage PPS sampling method to obtain the survey,in order to prepare for descriptive statistical analysis and modeling analysis.Then,the descriptive characteristics of user's basic situation are analyzed with R software.Through the analysis of histogram and pie chart of data by R software,the basic situation characteristics of samples are fully and directly understood and mastered.After that,Lasso variable selection,complexity adjustment are carried out for basic situation variables.Then,combined with the disordered multi-classification logistic model,the basic features of the users who choose the ticket software are modeled,and the factors that affect the users' choice of the ticket software are extracted,and valuable user characteristics are analyzed by comparison.Finally,a user experience cellular model with value realization as its core is established.Six latent variables,usefulness,agreeability,availability,searchability,accessibility and reliability,are selected.Among them,usefulness and agreeability directly affect the value.Usability,searchability,accessibility and reliability pass through usefulness and agreeability indirectly influences value.The structural equation model(SEM)is used to analyze the path data.The parameter estimation is based on the maximum likelihood method and is implemented by AMOS software.The user is analyzed to consider the characteristics of the software characteristics when using the software,and the user's preference feature is extracted.The results of the study indicate that the two assumptions that “accessibility has a significant positive effect on usefulness” and “usefulness has a significant positive effect on desirability” are not valid,and other assumptions are true.Among the two factors that directly affect value,the effect of desirability on value is greater,while the direct effect of usefulness on value is less user characteristics.Among the six factors that directly or indirectly affect value,in terms of total effect,the effect of desirability on value is the greatest,and the usefulness has the least influence on value.
Keywords/Search Tags:Lasso variable selection, disordered multi-classification logistic model, structural equation model
PDF Full Text Request
Related items