朴素贝叶斯分类器代码c语言

以下是一个简单的朴素贝叶斯分类器的 C 语言代码示例：

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <math.h>

#define MAX_FEATURES 1000
#define MAX_SAMPLES 10000

int num_features, num_samples;
int feature_counts[MAX_FEATURES][2];
double feature_probs[MAX_FEATURES][2];
int sample_classes[MAX_SAMPLES];
int sample_features[MAX_SAMPLES][MAX_FEATURES];

void read_data(char *filename) {
    FILE *fp = fopen(filename, "r");
    if (!fp) {
        perror("Error opening file");
        exit(EXIT_FAILURE);
    }
    char line[1024];
    while (fgets(line, 1024, fp)) {
        char *token = strtok(line, ",");
        int class = atoi(token);
        sample_classes[num_samples] = class;
        for (int i = 0; i < num_features; i++) {
            token = strtok(NULL, ",");
            int feature = atoi(token);
            sample_features[num_samples][i] = feature;
            if (feature) {
                feature_counts[i][class]++;
            }
        }
        num_samples++;
    }
    fclose(fp);
}

void train() {
    for (int i = 0; i < num_features; i++) {
        int count_0 = feature_counts[i][0];
        int count_1 = feature_counts[i][1];
        int total = count_0 + count_1;
        feature_probs[i][0] = (double) count_0 / total;
        feature_probs[i][1] = (double) count_1 / total;
    }
}

double classify(int *features) {
    double log_prob_0 = 0.0;
    double log_prob_1 = 0.0;
    for (int i = 0; i < num_features; i++) {
        if (features[i]) {
            log_prob_0 += log(feature_probs[i][0]);
            log_prob_1 += log(feature_probs[i][1]);
        } else {
            log_prob_0 += log(1.0 - feature_probs[i][0]);
            log_prob_1 += log(1.0 - feature_probs[i][1]);
        }
    }
    double prob_0 = exp(log_prob_0);
    double prob_1 = exp(log_prob_1);
    double total_prob = prob_0 + prob_1;
    return prob_0 / total_prob;
}

int main(int argc, char **argv) {
    if (argc != 3) {
        printf("Usage: %s <train_file> <test_file>\n", argv[0]);
        exit(EXIT_FAILURE);
    }
    char *train_file = argv[1];
    char *test_file = argv[2];
    FILE *fp = fopen(test_file, "r");
    if (!fp) {
        perror("Error opening file");
        exit(EXIT_FAILURE);
    }
    char line[1024];
    while (fgets(line, 1024, fp)) {
        char *token = strtok(line, ",");
        int class = atoi(token);
        int features[num_features];
        for (int i = 0; i < num_features; i++) {
            token = strtok(NULL, ",");
            features[i] = atoi(token);
        }
        double prob_0 = classify(features);
        int predicted_class = (prob_0 >= 0.5) ? 0 : 1;
        printf("%d,%d\n", class, predicted_class);
    }
    fclose(fp);
    return 0;
}

以上代码实现了一个简单的朴素贝叶斯分类器，包括读取数据、训练模型和分类等功能。其中，特征和样本数的最大值可以通过宏定义进行调整。在主函数中，输入参数为训练数据文件和测试数据文件的文件名。测试数据文件中的每一行表示一个测试样本，第一个数字为其类别，后面的数字为其特征值。程序将对每个测试样本进行分类，并输出其真实类别和预测类别。