Analysis And Forecast Of Beijing’s B&B Sharing Economy In The Post-Epidemic Era

With the rapid development of society,consumers’ consumption concepts,consumption patterns and consumption levels are also changing.More and more consumers are seeking personalized travel methods and travel services,and the demand for accommodation links in tourism is also on the rise.As one of the options for tourism accommodation,B&Bs can meet tourists’ basic accommodation needs as well as their needs for personalized services.As time goes by,B&Bs are gaining more consumer recognition and the B&B market is rapidly expanding and continuing to heat up.At the same time,the new economic model of sharing economy,which realizes the supply and demand of idle resources through the Internet platform,demonstrates the rationality and effectiveness of resource allocation and has greatly contributed to the development of the B&B industry.There are mant examples of this model being applied at home and abroad,such as Airbnb,Tujia and Piggy Short Rent.In the context of COVID-19 sweeping the world,the catering and tourism industry has suffered a huge impact,and the nationals have become more cautious about traveling out for accommodation,preferring accommodation with less contact and clean and hygienic environments,which has led to a decline in the attractiveness of B&Bs to travelers and an increase in competitive pressure on B&Bs,thus affecting the sales of B&Bs.Therefore,it is important to study the sharing economy of B&Bs in the post-epidemic era.In this paper,we firstly understand the relevant previous studies on shared B&Bs,and then explain the theoretical knowledge of text analysis,Lasso regression and XGBoost methods.In order to analyze the influencing factors of the sales of shared B&Bs in the post-epidemic era in a targeted manner,Airbnb China is chosen as the research platform,and the relevant data provided by the official website of Airbnb are first analyzed through the method of text analysis of online reviews to get the B&B-related features that users pay attention to,then the data are pre-processed according to the results of text analysis,and then Lasso regression is used to We then used Lasso regression to filter out the relatively important influencing factors among the many features,based on which we constructed an XGBoost model,and finally used the model to predict the sales of B&Bs in Beijing,so as to suggest optimizations and improvements to the B&B sharing platform and operators,in order to attract more consumers and promote the development of the B&B sharing industry.The results show that the Lasso-XGBoost model constructed in this paper fits well.According to the fitting results of the model,the main influencing factors of shared B&B sales include ratings(review_scores_cleanliness,review_scores_checkin,review_scores_rating),house attributes(room_type,neighbourhood_cleansed,amenities)and host attributes(host_is_superhost,host_response_rate,host_total_listings_count),in addition to other types of ratings that also have an impact on shared B&B sales.It is clear from these key influencing factors that when booking a shared B&B in the post-epidemic era,users will mainly refer to the ratings of B&Bs from users who have already completed their stays,and are most concerned about the hygiene of the shared B&B and whether the stay is convenient and efficient.