Analysis Of Service Quality Generated By Customers And Its Influence Based On Text Mining

In the era of big data,the diversification of information media and the advancement of communication technology have promoted continuous innovation in the form of social economy.As an emerging industry,the sharing economy has sprung up across the country and even globally.It is known that the concept of the sharing economy is to encourage human beings to conduct on collaborative innovation and value creation,which better responds to the call of the Central Committee and the mass.Therefore,people participate in the process of co-creation that has gradually been highly praised,that is,consumers can be transformed from service recipients to service providers to share their idle resources.When China is facing some major challenges in the housing market,housing leasing has begun to be promoted by most people,but service quality in accommodation still has enormous problems.In order to improve user experience,online short-term rental,as a kind of sharing economy in space,has offered some product information for customers to make purchase decision,and they can choose to share their room with other strangers,which can make them have an opportunity to own the different experience for local contact and social interaction.Meanwhile,service quality of house and host in online short-term rental is a key factor for consumers to consider,and it is also the primary condition for hosts to strengthen their core competitiveness.As the platform is quite transparent and consumer voices are protected,the vast amount of information generated by the platform and consumers increases the searching cost for potential customers.However,word-of-mouth generated by consumers is the most reliable way to feedback some important information about products or providers,so consumer review is an valuable channel for mining useful information.Therefore,this paper takes online short-term rental as the research object,and extracts the product service quality from the user-generated content to explore the core needs for consumers and the developing process of their purchase behavior.This article uses python to crawl the accumulated data of Xiaozhu.com on March 2017,from which I obtain some data including the attribute of landlord and tenant,product and order information,and consumer reviews.This paper is divided into two parts: text analysis and quantitative regression.On the one hand,the first part is to abstract some related variables from user experience: from the overall level,I explored user experience from the comments based on LDA topic modeling,and extracted five components,namely emotional component including attitude(e.g.repurchase attitude and recommendation attitude)and evaluation description;sensory component including room comfort(such as aesthetic design and warmth);practical component including house location,house equipment and surrounding facilities;cognitive component including cost-effective,description match,safety,problem-solving ability and thoughtful;and relational component including social interaction(e.g.host-guest interaction and local contact,etc.).From the perspective of topics extracted from Xiaozhu.com,user experience includes several dimensions such as service quality,perceived value,customer loyalty,and satisfaction.From the individual level,I adopted the machine learning classification algorithm model SVM to identify the topics contained in each comment of every product,and then calculated emotional metrics for each topic corresponding to different products based on the sentimental dictionary.Thus,according to these related information,products' service quality(e.g.empathy,tangible,reliable,responsive and assurance)is constructed.On the other hand,the second part is to explore the causal relationship between variables,I extracts the five measurements of service quality from user experience as an independent variable to explore its impact on customer loyalty and potential consumer purchase behavior,and to observe whether the herding effect and information disclosure will moderate the impact of customers' loyalty on purchase behavior.The results show that service quality of product and provider in the short-term rental(e.g.empathy,tangible,reliable,responsive and assurance)has a positive impact on customer loyalty and purchase behavior,while the customer loyalty will act as a mediator between perceived quality and consumer behavior;In addition,herd effect acts as a kind of learning reference,and information disclosure will convey the providers' altruistic signal,which will significantly regulate the impact of customers' loyalty on potential consumers' purchasing behavior;most importantly,under the context of short-term rental,empathy(e.g.social interaction)in the level of service quality has the strongest impact on consumers' attitudes,and the signals generated by consumers will be significantly stronger than the signals of platforms or providers.This paper explores the impact of service quality users feedback in the short-term rental on consumer behavior has several theoretical and practical significance.From a theoretical perspective,I explore the important components of user experience to observe how the signal generated by the customers,based on the service qualityperceived value-satisfaction-trust chain,affects customer loyalty and consumer purchasing behavior;in addition,I also observe whether herd effect and information disclosure can moderate the impact of customer loyalty on potential customers' purchase behavior,which will be meaning for future research in the field of customer behavior.From a practical perspective,it is suggested that service providers can increase the communication with customers to improve the product strategy,platform can optimize the layout design according to customers' concern,consumers can look through user-generated signals to capture useful information,which can reduce the search cost.