With the rapid development of e-commerce,the importance of online product reviews has become increasingly prominent and has become one of the key elements of consumers’purchasing decisions.As an important part of e-commerce,the mining of logistics service-related comments is of great significance to the improvement of consumer satisfaction and enterprise logistics service levels.In view of this,firstly,the web crawler technology was used to collect the logistics comment data of household appliances,fresh goods and electronic goods in JD Mall,then Python 3.7 was used to classify the comment text for emotional polarity,and finally,the word frequency analysis and LDA theme model analysis were carried out for the comment text with different emotional polarities.It is found that the three types of goods logistics service needs of consumers are significantly different.Among them,the installation and after-sales service of household appliances are relatively key,fresh goods have higher requirements on the delivery speed,and electronic goods pay more attention to the return and replacement service.Merchants or enterprises can improve the quality of their logistics services by improving the above aspects so as to improve consumer satisfaction.