Java 实现协同过滤算法推荐算法 - ToB企服应用市场:ToB评测及商务社交产业平台

import java.util.HashMap;
import java.util.Map;
public class CollaborativeFiltering {
// 用户评分矩阵
private static final Map<String, Map<String, Integer>> ratings = new HashMap<>();
static {
ratings.put("UserA", Map.of("Item1", 5, "Item2", 3, "Item4", 1));
ratings.put("UserB", Map.of("Item1", 4, "Item4", 1));
ratings.put("UserC", Map.of("Item2", 1, "Item4", 5));
ratings.put("UserD", Map.of("Item3", 5, "Item4", 4));
}
// 计算余弦相似度
private double cosineSimilarity(Map<String, Integer> ratings1, Map<String, Integer> ratings2) {
double dotProduct = 0.0;
double normA = 0.0;
double normB = 0.0;
for (String item : ratings1.keySet()) {
if (ratings2.containsKey(item)) {
dotProduct += ratings1.get(item) * ratings2.get(item);
}
normA += Math.pow(ratings1.get(item), 2);
}
for (double rating : ratings2.values()) {
normB += Math.pow(rating, 2);
}
normA = Math.sqrt(normA);
normB = Math.sqrt(normB);
return (normA == 0 || normB == 0) ? 0 : dotProduct / (normA * normB);
}
// 为用户推荐物品
public Map<String, Double> recommendItems(String user) {
Map<String, Integer> userRatings = ratings.get(user);
Map<String, Double> scoreMap = new HashMap<>();
for (String otherUser : ratings.keySet()) {
if (!otherUser.equals(user)) {
double similarity = cosineSimilarity(userRatings, ratings.get(otherUser));
for (String item : ratings.get(otherUser).keySet()) {
if (!userRatings.containsKey(item)) {
scoreMap.put(item, scoreMap.getOrDefault(item, 0.0) + similarity * ratings.get(otherUser).get(item));
}
}
}
}
return scoreMap;
}
public static void main(String[] args) {
CollaborativeFiltering cf = new CollaborativeFiltering();
Map<String, Double> recommendations = cf.recommendItems("UserA");
System.out.println("推荐物品给 UserA: " + recommendations);
}
}

复制代码

import java.util.HashMap;
import java.util.Map;
public class ItemBasedCollaborativeFiltering {
private static final Map<String, Map<String, Integer>> ratings = new HashMap<>();
static {
ratings.put("UserA", Map.of("Item1", 5, "Item2", 3, "Item4", 1));
ratings.put("UserB", Map.of("Item1", 4, "Item4", 1));
ratings.put("UserC", Map.of("Item2", 1, "Item4", 5));
ratings.put("UserD", Map.of("Item3", 5, "Item4", 4));
}
// 计算物品之间的余弦相似度
private double cosineSimilarity(Map<String, Integer> item1, Map<String, Integer> item2) {
// 与用户的计算相似
// 省略相似度计算的具体实现
return 0.0; // 这里应返回实际计算的相似度
}
// 为用户推荐物品
public Map<String, Double> recommendItems(String user) {
Map<String, Integer> userRatings = ratings.get(user);
Map<String, Double> scoreMap = new HashMap<>();
// 计算物品之间的相似度
// 省略物品相似度计算和推荐逻辑的实现
return scoreMap;
}
public static void main(String[] args) {
ItemBasedCollaborativeFiltering ibcf = new ItemBasedCollaborativeFiltering();
Map<String, Double> recommendations = ibcf.recommendItems("UserA");
System.out.println("推荐物品给 UserA: " + recommendations);
}
}

复制代码