Mahout实现基于用户的协同过滤算法
Mahout中对协同过滤算法进行了封装,看一个简单的基于用户的协同过滤算法。
基于用户:通过用户对物品的偏好程度来计算出用户的在喜好上的近邻,从而根据近邻的喜好推测出用户的喜好并推荐。
程序中用到的数据都存在MySQL数据库中,计算结果也存在MySQL中的对应用户表中。
package com.mahout.helloworlddemo; import java.sql.Connection; import java.sql.DatabaseMetaData; import java.sql.ResultSet; import java.sql.SQLException; import java.sql.Statement; import java.util.HashSet; import java.util.List; import org.apache.mahout.cf.taste.impl.model.jdbc.MySQLJDBCDataModel; import org.apache.mahout.cf.taste.impl.neighborhood.NearestNUserNeighborhood; import org.apache.mahout.cf.taste.impl.recommender.GenericUserBasedRecommender; import org.apache.mahout.cf.taste.impl.similarity.PearsonCorrelationSimilarity; import org.apache.mahout.cf.taste.model.DataModel; import org.apache.mahout.cf.taste.model.JDBCDataModel; import org.apache.mahout.cf.taste.neighborhood.UserNeighborhood; import org.apache.mahout.cf.taste.recommender.RecommendedItem; import org.apache.mahout.cf.taste.recommender.Recommender; import org.apache.mahout.cf.taste.similarity.UserSimilarity; import com.mahout.util.DBUtil; import com.mysql.jdbc.jdbc2.optional.MysqlDataSource; /** * *@author wxisme *@time 2015-9-13 下午6:25:26 */ public class RecommenderIntroFromMySQL { public static void main(String[] args) throws Exception { //连接MySQL MysqlDataSource dataSource = new MysqlDataSource(); dataSource.setServerName("localhost"); dataSource.setUser("root"); dataSource.setPassword("1234"); dataSource.setDatabaseName("mahoutdemo"); //获取数据模型 JDBCDataModel dataModel = new MySQLJDBCDataModel(dataSource, "taste_preferences", "user_id", "item_id", "preference","time"); DataModel model = dataModel; //计算相似度 UserSimilarity similarity = new PearsonCorrelationSimilarity(model); //计算阈值 UserNeighborhood neighborhood = new NearestNUserNeighborhood(2,similarity,model); //推荐 Recommender recommender = new GenericUserBasedRecommender(model,neighborhood,similarity); Connection con = DBUtil.getConnection(); Statement stmt = con.createStatement(); //获取每个用户的推荐数据并存入数据库 for(int i=0; i<5; i++) { List<RecommendedItem> recommendations = recommender.recommend(i, 3); String tableName = "user_" + i; for (RecommendedItem recommendation : recommendations) { //如果是第一次推荐就创建该用户的数据表 if(!doesTableExist(tableName)) { String createSQL = "create table " + tableName + " (item_id bigint primary key,value float);"; stmt.execute(createSQL); } String insertSQL = "insert into " + tableName + " values (" + recommendation.getItemID() + "," + recommendation.getValue() + " );"; //插入用户的推荐数据 stmt.execute(insertSQL); System.out.println(recommendation); } } } /** * 是否存在这个数据表 * @param tablename * @return * @throws SQLException */ public static Boolean doesTableExist(String tablename) throws SQLException { HashSet<String> set = new HashSet<String>(); Connection con = DBUtil.getConnection(); DatabaseMetaData meta = con.getMetaData(); ResultSet res = meta.getTables(null, null, null, new String[]{"TABLE"}); while (res.next()) { set.add(res.getString("TABLE_NAME")); } DBUtil.close(res, con); return set.contains(tablename); } }
测试数据:
1,101,5
1,102,3
1,103,2.5
2,101,2
2,102,2.5
2,103,5
2,104,2
3,101,2.5
3,104,4
3,105,4.5
3,107,5
4,101,5
4,103,3
4,104,4.5
4,106,4
5,101,4
5,102,3
5,103,2
5,104,4
5,105,3.5
5,106,4
运行结果:
更多Mahout和协同过滤算法的介绍与分析:
http://www.cnblogs.com/dlts26/archive/2011/08/23/2150225.html
http://www.tuicool.com/articles/FzmQziz
http://www.ibm.com/developerworks/cn/web/1103_zhaoct_recommstudy2/
作者:Pickle
声明:对于转载分享我是没有意见的,出于对博客园社区和作者的尊重一定要保留原文地址哈。
致读者:坚持写博客不容易,写高质量博客更难,我也在不断的学习和进步,希望和所有同路人一道用技术来改变生活。觉得有点用就点个赞哈。
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112835275-168981902.gif)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112841431-2047172832.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112847494-1544911856.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112904385-918783221.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111120143119-1302805212.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112856369-1466718103.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111112830416-794416355.jpg)
![](https://images2015.cnblogs.com/blog/735119/201701/735119-20170111113919510-1155220901.gif)