//注释的部分是基于文件也可以理解为基于日志文件的，
//DataModel 可以有很多种，实现abstractDataMode的子类，原则上都可以作为数据源，个人觉得，不管是那种DataMode各自有优缺点
//应该视情况而定，
package com.test.mahout;
import java.util.List;
import org.apache.mahout.cf.taste.impl.common.FastByIDMap;
import org.apache.mahout.cf.taste.impl.model.GenericDataModel;
import org.apache.mahout.cf.taste.impl.model.GenericUserPreferenceArray;
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.PreferenceArray;
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;
class RecommenderIntro {
public static void main(String[] args) throws Exception {
/* FastByIDMap<PreferenceArray> preferences =new FastByIDMap<PreferenceArray>();
*
PreferenceArray prefsForUser1 = new GenericUserPreferenceArray(10); A
prefsForUser1.setUserID(0, 1L);
prefsForUser1.setItemID(0, 101L); B
prefsForUser1.setValue(0, 3.0f); B
prefsForUser1.setItemID(1, 102L);
prefsForUser1.setValue(1, 4.5f);
… (8 more)
preferences.put(1L, prefsForUser1); C
DataModel model = new GenericDataModel(preferences); D
//A Set up PreferenceArray for user 1
//B Add the first of 10 preferences
//C Attach user 1’s preference*/
// laod Data File
// DataModel model = new FileDataModel(new File("C:\\mahout_data.txt"));
FastByIDMap<PreferenceArray> preferences =new FastByIDMap<PreferenceArray>();
/*FastByIDMap是经过优化了的key-value数据结够，在这里用来存储基本的元数据，《mahout in action》书中对这个数据结构做了详细的解释*/
PreferenceArray prefsForUser1=new GenericUserPreferenceArray(3);//注意这里的数字
// 这里是用来存储一个用户的元数据，这些元数据通常来自日志文件，比如浏览历史，等等，不同的业务场合，它的业务语义应该是不一样
prefsForUser1.setUserID(0, 1);/*个人对这里也感觉到非常模糊，其实这里保存用户ID完全用不着key-value结构，也用不着用一个long型的ID，这里个人觉得是为了以后扩展或者，保持统一的格式，注意用就是了*/
prefsForUser1.setItemID(0, 101);
prefsForUser1.setValue(0, 5.0f);
prefsForUser1.setItemID(1, 102);
prefsForUser1.setValue(1, 3.0f);
prefsForUser1.setItemID(2, 103);
prefsForUser1.setValue(2, 2.5f);
preferences.put(1l, prefsForUser1);//在这里添加数据
PreferenceArray prefsForUser2=new GenericUserPreferenceArray(4);
prefsForUser2.setUserID(0, 2);
prefsForUser2.setItemID(0, 101);
prefsForUser2.setValue(0, 2.0f);
prefsForUser2.setItemID(1, 102);
prefsForUser2.setValue(1,2.5f);
prefsForUser2.setItemID(2, 103);
prefsForUser2.setValue(2,5.0f);
prefsForUser2.setItemID(3,104);
prefsForUser2.setValue(3,2.0f);
preferences.put(2l, prefsForUser2);
PreferenceArray prefsForUser3=new GenericUserPreferenceArray(4);
prefsForUser3.setUserID(0, 3);
prefsForUser3.setItemID(0, 101);
prefsForUser3.setValue(0, 2.5f);
prefsForUser3.setItemID(1, 104);
prefsForUser3.setValue(1, 4.0f);
prefsForUser3.setItemID(2, 105);
prefsForUser3.setValue(2, 4.5f);
prefsForUser3.setItemID(3, 107);
prefsForUser3.setValue(3, 5.0f);
preferences.put(3l, prefsForUser3);
PreferenceArray prefsForUser4=new GenericUserPreferenceArray(4);
prefsForUser4.setUserID(0, 4);
prefsForUser4.setItemID(0,101);
prefsForUser4.setValue(0, 5.0f);
prefsForUser4.setItemID(1,103);
prefsForUser4.setValue(1, 3.0f);
prefsForUser4.setItemID(2,104);
prefsForUser4.setValue(2,4.5f);
prefsForUser4.setItemID(3, 106);
prefsForUser4.setValue(3, 4.0f);
preferences.put(4l, prefsForUser4);
PreferenceArray prefsForUser5=new GenericUserPreferenceArray(6);
prefsForUser5.setUserID(0, 5);
prefsForUser5.setItemID(0, 101);
prefsForUser5.setValue(0, 4.0f);
prefsForUser5.setItemID(1, 102);
prefsForUser5.setValue(1, 3.0f);
prefsForUser5.setItemID(2, 103);
prefsForUser5.setValue(2, 2.0f);
prefsForUser5.setItemID(3, 104);
prefsForUser5.setValue(3, 4.0f);
prefsForUser5.setItemID(4, 105);
prefsForUser5.setValue(4, 3.5f);
prefsForUser5.setItemID(5, 106);
prefsForUser5.setValue(5, 4.0f);
preferences.put(5l, prefsForUser5);
DataModel model=new GenericDataModel(preferences) ;//DataModel的建立
UserSimilarity similarity = new PearsonCorrelationSimilarity(model);
UserNeighborhood neighborhood = new NearestNUserNeighborhood(2,similarity, model);
// Create the recommender engine
Recommender recommender = new GenericUserBasedRecommender(model,neighborhood, similarity);
// C For user 1, recommend 1 item
List<RecommendedItem> recommendations = recommender.recommend(1, 1);
for (RecommendedItem recommendation : recommendations) {
System.out.println(recommendation);
}
}
}
转载请注明原文地址: https://ju.6miu.com/read-1000007.html
