基于数据挖掘的自动化推荐系统算法

结合人工神经网络中的自适应共振理论(ART)及数据挖掘(Data Mining)技术来建构一个可自动聚类族群特征且能挖掘出关联特质的自动化在线推荐系统。探讨如何有效地运用数据挖掘技术从大量的数据库中挖掘出完整知识,以推荐适当的信息给使用者,帮助他们在浩大的信息流中找到真正需要、有用的文件或信息。整合ART及数据挖掘技术,并针对推荐系统的特性提出一种改进的ART算法(MART算法)。实例验证了算法的有效性。

Algorithm of Automatic Recommender System Based on Data Mining

A typical online recommendation system is described.By using ART neural network and data mining technology,it can automatically cluster population characteristics and can dig out the associated characteristics.Aim at the online recommendation system applied on network,how effectively use data mining techniques to mine the complete knowledge from a large number of databases is discussed,then the appropriate information is recommended to users to help them to find really needed and useful documents or information in the vast flow of information.A pattern is put forward that combines ART neural network and data mining technology.Aim at the characteristics of recommendation system,a new modified ART algorithm(MART algorithm) is proposed.The result shows that the proposed algorithm is effective.