改进的TF-IDF算法，将IDF值改为自定义文件输入的每行热度权重python代码

以下是将IDF值改为自定义文件输入的每行热度权重的Python代码：

import math

def tfidf(docs):
    # 创建词频字典
    word_freq = {}
    for doc in docs:
        for word in doc.split():
            if word not in word_freq:
                word_freq[word] = 1
            else:
                word_freq[word] += 1
    
    # 计算TF值
    tf_values = []
    for doc in docs:
        doc_words = doc.split()
        doc_freq = {}
        for word in doc_words:
            if word not in doc_freq:
                doc_freq[word] = 1
            else:
                doc_freq[word] += 1
        tf_dict = {}
        for word in doc_words:
            tf_dict[word] = doc_freq[word] / len(doc_words)
        tf_values.append(tf_dict)
    
    # 读取自定义热度权重文件
    weight_dict = {}
    with open("weight.txt") as f:
        for line in f:
            word, weight = line.strip().split()
            weight_dict[word] = float(weight)
    
    # 计算IDF值并乘以热度权重
    idf_values = {}
    for word in word_freq:
        doc_containing_word = 0
        for doc in docs:
            if word in doc:
                doc_containing_word += 1
        idf_values[word] = math.log(len(docs) / (1 + doc_containing_word)) * weight_dict[word]
    
    # 计算TF-IDF值
    tfidf_values = []
    for tf_dict in tf_values:
        tfidf_dict = {}
        for word in tf_dict:
            tfidf_dict[word] = tf_dict[word] * idf_values[word]
        tfidf_values.append(tfidf_dict)
    
    return tfidf_values

其中，`docs`是一个包含多个文档的列表，每个文档是一个字符串。自定义热度权重文件`weight.txt`中每行包含一个词和一个浮点数，表示该词的热度权重值。`tfidf_values`是一个列表，其中每个元素代表一个文档的TF-IDF值，是一个字典，键为词，值为TF-IDF值。