Visualization Of Furniture Users' Potential Needs Based On Text Mining

As a pillar industry,furniture industry is related to national life.The depression of traditional furniture industry makes the rise of furniture e-commerce.Due to the particularity of furniture products,the e-commerce of furniture is faced with many challenges and difficulties in the fierce competitive environment.Under the background of the era of big data and the booming development of e-commerce platforms,users will post online shopping comments on the e-commerce platform after buying furniture.The comments include product attributes,services,logistics and other aspects,all of which imply users' demand for product improvement.If the furniture e-commerce can quickly grasp the user demand in online shopping reviews,it can timely improve the existing products and services,and find the future development direction of products,so as to find inspiration for rapid iteration,so as to firmly grasp the user stickiness and improve the core competitiveness.Based on the research of theories and key technologies such as text mining and user demand,this paper proposes a set of text mining process suitable for online furniture online shopping comment,preliminarily constructs the evaluation index of furniture online shopping comment,and verifies the usability of the method and obtains valuable conclusions through practical cases.Through empirical analysis,combined with the analysis of online shopping behavior process and the study of user groups,the online shopping comments of online furniture products were deeply excavated,and the potential needs of users were analyzed and Suggestions for improvement were put forward,which verified the feasibility of online furniture users' demand analysis.The main research work of this paper is as follows:First,construct evaluation index based on furniture online shopping review.Through to the furniture industry background and product types and characteristics of the relevant investigation to determine the research object,constructing evaluation index system of the secondary,review content,commentators two level indexes is put forward,accordingly puts forward the online product reviews and objective attribute for the positive comments may positively affect the usefulness of online comments and related assumptions behind to validate.Second,put forward the process of text mining that is applicable to furniture online shopping review.From the definition,development process,main research field of text mining,and the application of the relevant aspects to do academic research and industry application situation,the preliminary established a set of suitable for furniture online reviews the process of text mining,the final list in some text mining tool,and analyses the advantages and disadvantages,so as to make sure the Python as a text mining tool,this article choose PyCharm compiler writing the related programs.Third,the online shopping review of furniture in-depth mining,get effective user demand information.Data acquisition and preprocessing,feature word extraction,emotion analysis and other methods and techniques are used.To be specific,the relevant methods and processes ofdata acquisition are firstly studied.By selecting the online shopping platform selling furniture as the data source,the recent online shopping comments of furniture products are taken as the sample data,and Jieba word segmentation tool is used to pre-process online shopping comments.Second text feature extraction and text characteristic evaluation method is studied,using TF-IDF algorithm to keyword extraction collected furniture online comments,related to high frequency keywords,based on the keyword extraction of co-occurrence analysis,with the help of a DataFrame get word matrix form,further improve the accuracy,and with the aid of Worldcloud generation word cloud,visualization Networkx construct the semantic network diagram;Then it studied the relevant theories of text emotion analysis and listed the existing text emotion analysis techniques.It selected SnowNLP as the emotion analysis tool in this paper to calculate the emotion value of each online shopping comment of relevant furniture products,understood the attitude and opinions of users,and made dynamic interaction charts with the help of ECharts.Finally,the process of user demand and online shopping behavior is studied,so as to obtain the steps of online furniture user demand analysis.In combination with online furniture user group research,the potential demand of furniture user demand is further analyzed,and the feasibility of online shopping comment evaluation index and online furniture user demand analysis method is preliminarily verified.