Text Mining Method For Continuous Survey Data

"My Century-old Family History" is a continuous research project of social practice promoted to college students.It aims to complete the record and narration of their own family history through the comprehensive and practical continuous investigation conducted by college students,so as to inspire and guide young students to look back at history,examine society,discover themselves and realize growth.With1517 individual family history research reports completed by college students in eight consecutive years,this paper hopes to explore and understand the characteristics and rules of intergenerational inheritance of ordinary families in modern China from the perspective of statistics,and reflect on the reasons for the rise and fall of a family and the fundamental driving force for the long-term development of a country from the perspective of a family,and feels the great influence of macro-social economic development on a family from the micro perspective.Stimulating college students’ interest in family history research through the mining and visualization analysis of family history text data.This paper uses the event history analysis method to combine the personal growth experience and family development and change process with the national development and the background of The Times,which lays the basic logic of combining the "family history" and "country history" in this paper.Secondly,this paper uses emotion analysis and topic mining in text mining as the main research methods.Through these two methods,this paper completes the analysis of the influence of related events,key time nodes and significant historical events on family inheritance in the family history investigation,the emotion fluctuations under different time spans,and the text mini9 ng analysis under different subject categories.Finally,this paper uses the shiny package to build a bridge between R language and HTML,then design and complet a simple and easy to operate family history analysis system based on text mining to facilitates the display of related work results.The research shows that since modern times,the reproduction and development of Chinese families have been greatly influenced by major historical events.From the Revolution of 1911,to the founding of the Communist Party of China,to the War of Resistance against Japanese aggression,to the War of Liberation,to War to Resist U.S.Aggression and Aid Korea,socialist construction,to the reform and opening up and to the decade of the New Era,every small family has made great achievements in the independence,liberation and prosperity of the country.From the survey report of family history,it can be seen that with the change of time,the time span and theme of the recording of family history by college students of different grades have changed greatly.The articles related to the Anti-Japanese War,resisting the United States and aiding the Dynasty in the family history are on the decrease,while the articles about the reform and opening up are on the increase.In the research on the change of themes of The Times,it is found that the subject and focus of writing under different themes have also changed.At the same time,the analysis of family history theme mining and modeling shows that,for the topic analysis of large-scale continuous research,the event history analysis method and text mining method must take the core idea of the text as the basic logic to get valuable analysis conclusions,while the simple technical analysis and data model cannot reflect the real conclusions of text narration and theme iteration.Finally,this paper also designs a small database system for the family history continuity research project,which can help us succinctly and effectively complete the preliminary analysis and extract some valuable analysis topics.