文本预处理中索引技术概述

1import pandas as pd
2
3df = pd.read_csv('reviews.csv')
4software_reviews = df['review_text'].tolist()
5vocabulary = {}
6indexed_reviews = [] 
7for review in software_reviews:
8    indexed_review = []
9    for word in review.split():
10        if word not in vocabulary:
11            vocabulary[word] = len(vocabulary)
12        indexed_review.append(vocabulary[word])
13    indexed_reviews.append(indexed_review) 
14for i, indexed_review in enumerate(indexed_reviews, start=1):
15    print(f"评论 {i}: {indexed_review}") 
16print("\n词到索引的映射字典：")
17print(vocabulary)

1import pandas as pd
2from sklearn.feature_extraction.text import CountVectorizer
3 
4df = pd.read_csv('reviews.csv')
5reviews = df['review_text']
6vectorizer = CountVectorizer()
7X = vectorizer.fit_transform(reviews)
8print("词汇表（词到索引的映射）：")
9print(vectorizer.vocabulary_)
10print("\n特征矩阵（词袋表示）：")
11print(X.toarray())