帮我用python实现一个1)从给定的语料集中提取出带有类别标签的句子并进行预处理（分词、去除停用词），构建候选特征词集S。 2)对候选特征词集S中的所有词汇w，计算其特征得分s(w)，计算方法分别使用文档频率

（DF）、逆文档频率（IDF）、TF-IDF和信息增益（IG）四种方法。最终输出每种方法下得分最高的前n个特征词汇。

首先，需要准备好语料集和停用词表。假设语料集为corpus.txt，停用词表为stopwords.txt，可以使用以下代码读取：

with open('corpus.txt', 'r', encoding='utf-8') as f:
    corpus = f.readlines()

with open('stopwords.txt', 'r', encoding='utf-8') as f:
    stopwords = f.read().splitlines()

接下来，进行预处理，分词并去除停用词。可以使用jieba库进行中文分词，使用以下代码实现：

import jieba

def preprocess(text):
    words = jieba.lcut(text)
    words = [w for w in words if w not in stopwords]
    return words

sentences = []
labels = []

for line in corpus:
    label, sentence = line.split('\t')
    sentences.append(preprocess(sentence))
    labels.append(label)

这样就得到了带有类别标签的句子列表sentences和对应的标签列表labels。下一步是构建候选特征词集S，可以使用Python的set类型，将所有句子中出现的词汇加入集合中，即可得到候选特征词集S：

candidate_words = set()
for sentence in sentences:
    candidate_words.update(sentence)

接下来，分别使用文档频率（DF）、逆文档频率（IDF）、TF-IDF和信息增益（IG）四种方法计算特征得分。这里使用sklearn库计算IDF和TF-IDF，使用自己编写的函数计算DF和IG：

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_selection import mutual_info_classif
import numpy as np

def df(word, sentences):
    return sum(1 for sentence in sentences if word in sentence)

def idf(word, sentences):
    return np.log(len(sentences) / (1 + df(word, sentences)))

def tfidf(word, sentences):
    vectorizer = TfidfVectorizer(vocabulary=[word])
    tfidf_matrix = vectorizer.fit_transform(sentences)
    return tfidf_matrix.sum()

def ig(word, sentences, labels):
    N = len(sentences)
    N11 = sum(1 for i in range(N) if word in sentences[i] and labels[i] == '1')
    N10 = sum(1 for i in range(N) if word in sentences[i] and labels[i] == '0')
    N01 = sum(1 for i in range(N) if word not in sentences[i] and labels[i] == '1')
    N00 = sum(1 for i in range(N) if word not in sentences[i] and labels[i] == '0')
    N1_ = N11 + N10
    N0_ = N01 + N00
    N_1 = N11 + N01
    N_0 = N10 + N00
    N__ = N1_ + N0_
    p11 = N11 / N
    p10 = N10 / N
    p01 = N01 / N
    p00 = N00 / N
    p1_ = N1_ / N
    p0_ = N0_ / N
    p_1 = N_1 / N
    p_0 = N_0 / N
    H_ = -p1_ * np.log2(p1_) - p0_ * np.log2(p0_)
    H_1 = -p11 * np.log2(p11) - p10 * np.log2(p10) if N1_ > 0 else 0
    H_0 = -p01 * np.log2(p01) - p00 * np.log2(p00) if N0_ > 0 else 0
    return H_ - p_1 * H_1 - p_0 * H_0

def score(word, sentences, labels, method):
    if method == 'DF':
        return df(word, sentences)
    elif method == 'IDF':
        return idf(word, sentences)
    elif method == 'TF-IDF':
        return tfidf(word, sentences)
    elif method == 'IG':
        return ig(word, sentences, labels)

methods = ['DF', 'IDF', 'TF-IDF', 'IG']
n = 10

for method in methods:
    scores = [(word, score(word, sentences, labels, method)) for word in candidate_words]
    top_n = sorted(scores, key=lambda x: x[1], reverse=True)[:n]
    print(f"Top {n} words using {method}:")
    for word, score in top_n:
        print(f"{word}: {score:.4f}")

这样就可以分别使用DF、IDF、TF-IDF和IG四种方法计算特征得分，并输出每种方法下得分最高的前n个特征词汇。