Research And Implementation Of Shopping System Based On Improved Mahout Recommendation Engine

With the continuous expansion of the scale of e-commerce,online shopping system as a form of e-commerce has also become increasingly popular with the development of e-commerce.With the prevalence of online shopping,for consumers,the number of goods has become more and more,and there are more and more complex and diverse species,so they need to spend a lot of time and effort to select from thousands of merchandise information when they want to buy something.For businesses,everyone wants to increase sales and sales of their products in the shopping system.Therefore,the personalized recommendation of goods for the complex and diverse,very large amount of information shopping system is essential,not only it can help consumers to pick out their desired goods from a large number of information,but also can help businesses improve consumer's praise of the business,so it can help businesses to improve business sales.The thesis first introduces the background and significance of the topic,followed by the analysis of the status quo of e-commerce,and then introduced the development of recommended technology background and research content,and then studied the Mahout collaborative filtering recommendation algorithm,and finally introduced the shopping system code implementation and completed the testing of the entire shopping system to validates the improved collaborative filtering algorithms and the feasibility of the entire system.To sum up,this thesis mainly completed the following four aspects of the work:1.With research and analysis of the Mahout collaborative filtering recommendation engine,we get the different combinations of different recommended algorithm.Compared with different combinations of algorithms,we get the highest rate of recall and precision of a combination algorithms to apply to the shopping system in this thesis;2.Analysis the demand of the shopping system and the design of the overall architecture,combined with the needs analysis and the overall design of the conclusion,completed the system of database logic design,and completed the design of the database table;3.According to the database structure design and database management tool Mysql,the thesis completed the SSH environment construction and realize the main module of the shopping system,and according to the Mahout collaborative filtering recommendation algorithm combination applied to the shopping system in this thesis,the thesis improve the Mahout collaborative filtering recommendation algorithm;4.Set up the shopping system test environment,do the detailed testing of the entire system including product recommendation module to ensure the good operation of the entire shopping system.Test results show that the thesis' s shopping system is a complete,practical and user-friendly shopping system based on the improved Mahout recommendation engine.