数值和文本值的组合之间的余弦相似度

import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

# 示例数据
data = [
    {"id": 1, "text": "hello world", "value": 10},
    {"id": 2, "text": "hello python", "value": 20},
    {"id": 3, "text": "goodbye world", "value": 30}
]

# 将文本转换为TF-IDF向量
vectorizer = TfidfVectorizer()
texts = [item["text"] for item in data]
tfidf_matrix = vectorizer.fit_transform(texts)

# 将数值转换为向量
values = np.array([item["value"] for item in data]).reshape(-1, 1)

# 组合文本和数值向量
combined_matrix = np.hstack((tfidf_matrix.toarray(), values))

# 计算余弦相似度
similarity_matrix = cosine_similarity(combined_matrix)

print(similarity_matrix)