Research On Visualization Based On The Information Feedbacked By Users In E-commence

With the massification of E-ecommerce,more and more users choose online shopping.Data shows that e-commence transaction volume has reached the level of trillion,and each transaction will generate corresponding review data,which accumulates a mass of e-commerce customer feedback information data.In the field of e-commence,the feedback information of comment plays an significant role in the purchase decision of customer.Users need to browse a large amount of review data to judge the actual information of products.In the context of massive and largely descriptive review data,customers will confront the following trouble when browsing product reviews:1?due to the large amount of comment data,users cannot effectively browse all feedback information about review,which will cause the problem of information loss;2?faced with a large amount of text description information,customers must spend lots of energy to scan and screen valuable message,which can lead to inefficient information acquisition for users;3?Due to the huge review data on the opposite side,which causes great interference barriers for users to filter valid comment.Combining the above factor,anyone may lead to wrong purchase decisions for users.To solve above problem,this paper proposes an solution of information visualization to work out that user may confront barrier during browsing product comment.Firstly,in the article,the user research method is carried out to explore the problem encountered by users in the process of browsing comments and master the expected goal of customer to obtain information.Secondly,the web crawler is used to grasp data from the e-commence platform and analyze the raw data.Subsequently,the original data is structured and divided into basic review data and textual review data.Based on the semantic analysis of text data by NLP and the training of basic corpus by Word2 Vec model,the text data that can be directly used for visualization is obtained.Finally,this paper adopts Pyechats drawing tool and Matplotlib visualization tool to map the structured data and get the final visual chart.In the period of estimating information visualization,the paper uses mobile application as carrier to design and test the effect of infographic.According to the method of User Experience Design to design the visual interface,and then by the method of Usability testing to verify the effectiveness of the visualization and user satisfaction,the final evaluation results will feedback to the design of visual chart,then improve it.In the paper,information visualization solutions effectively reduce the cost of reading textual information and enhance the efficiency of browsing review for customer,which make the shopping process tend to be more fast and efficient.Eventually,this paper's solution of information visualization obtains high score.At the same time,this subject belongs to interdisciplinary research,which effectively integrates the application of e-commence,data processing,information visualization,NLP,design and other fields to work out the emerging problem in the era of big data.