Sharing Economy Credit Evaluation Research On The Seller Of Online Short Term Renting Mode

With the rise of Internet technology and the transformation of consumption patterns,the concept of the sharing economy has gradually gained popular support.The integrated utilization of resources under the sharing model has the characteristics of environmental protection and sustainable development,and has become a new driving force for economic development.The sharing economy boom has brought new development opportunities.In2011,the domestic market ushered in the beginning of the online short-term rental industry.In recent years,the number of online short-term rental users and the size of the industry have increased significantly,entering a period of rapid development,and becoming a new industry that meets social needs.The core of the development of the online short-term rental industry is to establish a good credit system.In actual operation,frequent occurrence of credit accidents due to information asymmetry,the credit management of online short-term rental industry has attracted the attention of scholars at home and abroad.This article obtains actual data from the domestic representative platform Xiaozhu.Selecting variables from three aspects of housing facilities,seller attributes and user feedback to conduct seller credit evaluation and analyze influencing factors.Based on the topic extraction of LDA,a quantitative index of text is constructed through the semantic analysis of sentiment dictionary,and establish Decision Tree model to analyze the influencing factors based on the text information.In terms of prediction,the integrated learning algorithms of Boosting and Bagging are used to establish a Random Forest and XGBoost model.Compared with the 10-fold cross-validation accuracy rate,F-score and other related indicators,the contribution of the text quantification index is verified,and the model prediction efficiency is improved based on the classification tree.Studies have shown that with the change in the consumption model of the mobile internet development,online short-term rental users have put forward new requirements for products and services,expecting more emphasis on unique experiences and personalized services.The text frequency shows that users pay more attention to the landlord than the house itself.The results of the Decision Tree model show that the house's occupancy experience,transportation location,and in-room facilities,the total number of orders received by the seller,and service attitude have an effect on building trust,and product service optimization should focus on these aspects.The addition of text quantitative indicators has a significant contribution to the seller's credit prediction.The method of constructing quantitative indicators provides ideas for improving the credit system.The industry credit management should pay attention to the information of user reviews and feedback,and related platforms can improve the efficiency of credit management accordingly.