The Product Review Mining Of Taobao And Its Application Research

With the maturity of online shopping,the major electronic commerce platform competes fiercely,which makes the marketing strategy design and management mode become higher and higher status in internet sales.How to help users to pick out their desired goods quickly which plays a crucial role in the competition of the electronic commerce platforms.Research shows that users often hesitant in several commodities,at this time most users will browse the buyer's comments on this commodity to help themselves to make choice.So,the other buyers' comments play an important role to the coming user make the final decision.Based on the above phenomenon,analyze the users' comments data and mine the valuable information by the existing recommendation strategy,this paper put forward a feasible recommendation strategy to help users to solve the difficult choice within the less number of commodities.The main function includes:First of all,a web crawler is implemented.The crawler can effectively acquire the user comment data,and use emotion analysis technology,extract the key information from the comments,effectively distinguish the good,middle and bad comments,finally realizes the acquisition,mining and analysis to information.Secondly,a Taobao scoring algorithm is proposed.Based on the true comment data,this algorithm considered synthetically the weight to the good,middle and bad comments,the positive and negative information in each comment,and the user viewpoints,calculate the final score and recommend the commodity to users by the scoring result.At last this algorithm is used to acquire the comments data on Taobao page,compare the algorithm results with the sale amount in a near month,it shows that the recommendation result is consistent with the current purchase trend to Taobao users.Finally,a Taobao comment mining system is designed and implemented.According to the idea of software engineering,complete the analyzing requirement,detailed design and specific utilization under system data preparation module,scoring algorithm module and commodity recommendation module.