协调过滤推荐算法代码java

以下是一个基于用户的协同过滤推荐算法的Java代码实现：

import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
import java.util.List;
import java.util.Map;
import java.util.TreeMap;

public class UserBasedCF {
    // 用户-物品评分矩阵
    private double[][] data;

    public UserBasedCF(double[][] data) {
        this.data = data;
    }

    // 计算用户之间的相似度
    private double similarity(int user1, int user2) {
        double sum = 0;
        int count = 0;
        for (int i = 0; i < data[user1].length; i++) {
            if (data[user1][i] != 0 && data[user2][i] != 0) {
                sum += Math.pow(data[user1][i] - data[user2][i], 2);
                count++;
            }
        }
        return count == 0 ? 0 : 1 / (1 + Math.sqrt(sum / count));
    }

    // 为指定用户推荐物品
    public int recommend(int userId, int k) {
        Map<Integer, Double> simMap = new TreeMap<>();
        for (int i = 0; i < data.length; i++) {
            if (i != userId) {
                double sim = similarity(userId, i);
                if (sim > 0) {
                    simMap.put(i, sim);
                }
            }
        }
        List<Map.Entry<Integer, Double>> simList = new ArrayList<>(simMap.entrySet());
        Collections.sort(simList, new Comparator<Map.Entry<Integer, Double>>() {
            @Override
            public int compare(Map.Entry<Integer, Double> o1, Map.Entry<Integer, Double> o2) {
                return -Double.compare(o1.getValue(), o2.getValue());
            }
        });
        double[] predRatings = new double[data[0].length];
        for (int i = 0; i < data[0].length; i++) {
            if (data[userId][i] == 0) {
                double sum = 0, weightSum = 0;
                for (int j = 0; j < k && j < simList.size(); j++) {
                    int otherUser = simList.get(j).getKey();
                    double weight = simList.get(j).getValue();
                    double rating = data[otherUser][i];
                    if (rating != 0) {
                        sum += weight * rating;
                        weightSum += weight;
                    }
                }
                if (weightSum != 0) {
                    predRatings[i] = sum / weightSum;
                }
            }
        }
        int maxIndex = 0;
        for (int i = 1; i < predRatings.length; i++) {
            if (predRatings[i] > predRatings[maxIndex]) {
                maxIndex = i;
            }
        }
        return maxIndex;
    }
}

使用示例：

double[][] data = {
    {5, 3, 0, 1},
    {4, 0, 4, 4},
    {1, 1, 0, 5},
    {0, 0, 4, 0},
    {5, 4, 3, 1},
    {0, 1, 5, 4}
};
UserBasedCF cf = new UserBasedCF(data);
int itemId = cf.recommend(0, 2);
System.out.println(itemId);  // 输出：2

该代码实现了一个基于用户的协同过滤推荐算法，通过计算用户之间的相似度来预测用户对未评分物品的评分。具体实现中，首先计算每个用户与其他用户的相似度，然后选取相似度最大的k个用户的评分，计算每个未评分物品的预测评分，并返回预测评分最高的物品。