| Happiness is a key word gradually highlighted with the development of the times and social progress.The study of happiness has been paid more and more attention in academia,and the dimension of the study has been broadened.The study of well-being is gradually integrated with sociology,economics,psychology,statistics and management.There are many achievements in the study of well-being both in content and method.Based on the research of happiness,this paper first elaborates the theory and research methods of happiness,and then uses the macro-micro design ideas to analyze the related factors of workers’ happiness in Fuxin,Liaoning Province,and constructs a relevant quantitative analysis system.The innovation of this study is that on the premise of using crawlers to crawl the network content,the naive Bayesian algorithm is used to quantify the happiness keywords.At the same time,the rough set algorithm is used to quantify the questionnaire results of the factors affecting workers’ happiness,and the results obtained by the two methods are compared and analyzed.The specific research contents are three parts.In the first part,this chapter designs a blog crawler analysis system based on Naive Bayesian algorithm,classifies the crawled content and arranges the related vocabulary frequency.The vocabulary related to happiness is obtained through high frequency vocabulary,and the related factors are analyzed,and the results are summarized.The second part investigates the influencing factors of happiness through questionnaires.After that,the results of crawling related words are compared and analyzed,and four major categories affecting happiness are concluded.Furthermore,the factor system of employee happiness index should be constructed.Deeply excavate all kinds of factors affecting workers’ happiness,establish a reasonable and correct evaluation system,improve the accuracy of predicting the development trend of workers,and establish a first-level system of factors affecting happiness index.In the third part,the data collected by online and offline methods are processed to meet the purpose of data analysis,which mainly includes several links: data acquisition and data discretization.Finally,attribute reduction of the collected data is carried out.According to the requirement of rough set,a decision information system is constructed,and the attribute reduction algorithm based on relative identification degree proposed in this chapter is used to find out the key factors affecting the well-being of trade union workers.Through comparative analysis of data based on naive Bayesian classification and rough reduction,it is concluded that young people have a high degree of happiness related to money,housing,mobile phones and other popular topics,while as workers,they have a strong happiness related to health,pension,medical insurance,social security and other contents.In the future quantitative study of human psychology,we can use reptiles or offline data collection methods.At the same time,using Naive Bayesian Classification or Rough Set Reduction and other methods to reduce the data analysis,has a certain methodological significance for this kind of research.This paper contains 12 figures,6 tables and 52 references... |
