基于word2vec的文本分类 完整代码

基于word2vec的文本分类的完整代码如下所示：

import numpy as np
from gensim.models import Word2Vec
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import LabelEncoder
from sklearn.svm import SVC

# 1. 准备数据
sentences = [['I', 'love', 'this', 'movie'],
             ['This', 'is', 'an', 'awesome', 'movie'],
             ['Great', 'movie', 'ever'],
             ['I', 'like', 'it'],
             ['This', 'is', 'terrible']]

labels = ['positive', 'positive', 'positive', 'positive', 'negative']

# 2. 训练Word2Vec模型
model = Word2Vec(sentences, min_count=1)

# 3. 获取词向量
word_vectors = model.wv

# 4. 将文本转换为向量表示
X = []
for sentence in sentences:
    vector = np.zeros(100)  # 假设词向量维度为100
    for word in sentence:
        vector += word_vectors[word]
    X.append(vector)

# 5. 标签编码
label_encoder = LabelEncoder()
y = label_encoder.fit_transform(labels)

# 6. 划分训练集和测试集
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 7. 训练分类器
classifier = SVC()
classifier.fit(X_train, y_train)

# 8. 预测
y_pred = classifier.predict(X_test)

# 9. 输出预测结果
for i in range(len(X_test)):
    print("Text:", sentences[i])
    print("True Label:", labels[i])
    print("Predicted Label:", label_encoder.inverse_transform([y_pred[i]]))
    print()