| With the advancement of Internet technology and the increasing improvement of logistics distribution system,online shopping may have become the mainstream way,fresh ecommerce as one of them has also been rapidly rising and developing,consumers online shopping and online reviews are not only for the experience of the product sharing,but also contains a large number of product meaning outside the evaluation information,such as price,customer service and logistics services,which can be used to provide reference and decision support to other consumers.Due to perishable attributes,fresh products are closely related to the quality of logistics services,especially when fresh e-commerce operates,and there is a large amount of information related to logistics service quality in online review data,which is directly related to consumers’ shopping experience,so the logistics service quality of fresh products is very important for fresh e-commerce.Therefore,how to extract useful and related information related to logistics service quality from a large number of online review data is the key to this study,and how to optimize the theoretical model with the help of output evaluation results is also a further expansion of this thesis.The main research work in this thesis includes:(1)Analysis of Logistics Service Quality Evaluation Index System of Fresh Products:Based on the LSQ model,combined with the characteristics of fresh products,this thesis provides theoretical support for the evaluation of logistics service quality from six dimensions:freshness preservation integrity,logistics response ability,personnel communication quality,error handling,logistics service cost and information quality.(2)Data collection and processing based on online reviews: Firstly,the online review data of various fresh products is obtained from platforms such as JD Fresh,Taobao Fresh and Suning.com,so as to establish the data basis of this thesis;Then,the collected online comment data is pre-processed such as word segmentation,word removal,and custom dictionary construction,and the TF-IDF value of logistics keywords in each dimension is calculated by Python language and the weight of the indicator is obtained,which reflects the importance of each dimension or indicator to a certain extent.(3)Fresh product logistics service quality evaluation and supply chain model optimization: The sentiment analysis algorithm based on deep learning obtains the sentiment value of each comment,calculates the sentiment score of the indicators in each dimension,and summarizes the logistics service quality evaluation results together with the weight of the indicators and discusses and analyzes.Based on the practical results of logistics service quality evaluation,representative dimensions of "freshness integrity" and "logistics responsiveness" were selected to optimize the supply chain model of fresh products.Based on online commentary,this thesis proposes an evaluation index system for fresh products and uses text analysis technology to evaluate logistics service quality,which provides a theoretical reference for improving the logistics service quality of fresh ecommerce industry.Then,according to the practical results,the supply chain optimization is carried out,and the results show that the performance of the evaluation dimension stems from consumers’ preference for them,and the influence ability of "freshness integrity" is also explained from the theoretical model level. |
PDF Full Text Request