Design And Implementation Of Personalization Marketing Algorithm Based On Data Mining

Recent years, as the development of the e-commerce, personalization marketing is becoming an important part of enterprise marketing activities. Personalization marketing is a new marketing method that faces to consumers directly and provides the right things to every consumer. One of the common methods that used by personalization marketing is personalization recommendation system. This kind of system can help the enterprises analyze the consumer behaviors and recommend the right commodities to the right person. By doing this, enterprises can raise the sales volume dramatically and attract more consumers. The core of the personalization recommendation system is the recommendation algorithms. A good recommendation algorithm can discover the potential interests by mining the associations, this is quite critical in recommendation systems.Most recommendation algorithms are based on data mining technology. These algorithms can discover the user behavior mode by mining large amount of user behavior data. The usual methods of advertising recommendation are recommendations using collaborative filtering method and association rule mining method. However, these algorithms are mostly used by e-commerce websites that many commodities are for recommendation. There is another condition that there is only one commodity for recommendation. This condition can be seen when mobile service operators want to recommend new service to the mobile users. This kind of user recommendation of one target commodity can be seen as a user selection problem. As more and more services are provided to the mobile users today, they now are facing more and more choices. It's the key problem for mobile operators to win more users choosing their service. So when the mobile operators provide a new service, they may need user selection method to recommend this new service to the users. To solve this problem, this paper presents a novel recommendation algorithm based on the data mining technology that can solve the problem above.First, this paper studies the related data of a certain mobile service operator when they recommend a new service. After studying this data, we make it as a standard data set and publish it on the Internet. This data set can be used by researchers and scholars when study the related problems. Secondly, we present a novel user selection method of advertising recommendation. This method can provide a way to handle the mobile user recommendation problem. Compare with the ordinary method, the method presented in this paper can improve the success rate dramatically and reduce the amount of advertisements. As a result, a lower cost and higher success rate can make the mobile providers gain more new users.