Mahout基于用户的协同过滤算法的例子

每行测试数据分别标识用户id(uid),物品id(itemid),评分(rating),评分时间(time)

3464,2502,3,973282547

3464,3160,2,973282494

3464,2505,3,967175070

3464,1703,2,967248043

3464,1704,5,967246680

3464,3163,1,967174266

3464,2369,4,973282339

3464,1569,4,967247436

3464,896,3,967247557

3464,3316,3,973282934

3464,2517,3,967174139

3464,3174,4,967174266

3464,3175,2,973282421

3464,3176,3,967174298

3464,1573,3,967247865

3464,3178,4,967247587

3464,105,3,967248019

3464,3325,4,973282547

3464,1721,3,967247042

3464,3327,4,973282892

3464,3185,3,967174298

3464,1727,4,967248268

3464,111,5,967174438

3464,3186,4,967242949

3464,1729,3,967247165

3464,1584,3,967247078

3464,2387,3,967247884

3464,2389,4,967175256

3464,1589,4,967248019

3464,1732,4,967247306

3464,2391,4,967246935

3464,2395,4,973282625

3464,2396,5,967246752

3464,1597,4,967174960

3464,2541,3,967247865

package userBased;

import java.io.File;
import java.util.List;

import org.apache.mahout.cf.taste.impl.model.file.FileDataModel;
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.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;

/**
* mahout基于用户的协同过滤算法
*
*/
public class UserBased {

public static void main(String[] args) throws Exception {

DataModel model = new FileDataModel(new File("F:/ml-1m/ratings.dat"));
/**
* 用户偏好数据包含评分
欧氏距离：EuclideanDistanceSimilarity
皮尔森距离：PearsonCorrelationSimilarity
余弦距离：UncenteredCosineSimilarity

用户偏好数据不包含评分
曼哈顿距离：CityBlockSimilarity
对数似然距离： LogLikelihoodSimilarity
*/
UserSimilarity similarity = new PearsonCorrelationSimilarity(model);
// 相邻用户UserNeighborhood
/**
* NearestNUserNeighborhood
指定距离最近的N个用户作为邻居。
示例：UserNeighborhood unb = new NearestNUserNeighborhood(10, us, dm);
三个参数分别是： 邻居的个数，用户相似度，数据模型
邻居个数的大小直接决定了推荐结果的近似程度和计算的复杂度
ThresholdUserNeighborhood
指定距离最近的一定百分比的用户作为邻居。
示例：UserNeighborhood unb = new ThresholdUserNeighborhood(0.2, us, dm);
三个参数分别是： 阀值（取值范围0到1之间），用户相似度，数据模型
*/
UserNeighborhood neighborhood = new NearestNUserNeighborhood(500, similarity, model);
//根据数据模型、用户相似度模型、以及邻近值构建推荐引擎
Recommender recommender = new GenericUserBasedRecommender(model, neighborhood, similarity);
// 向用户100推荐2个商品
List<RecommendedItem> recommendations = recommender.recommend(100, 2);
for (RecommendedItem recommendation : recommendations) {
// 输出推荐结果
System.out.println(recommendation);
}
}
}