企业画像推荐

import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import linear_kernel

# 假设我们有一个包含企业信息和产品描述的数据集
data = {
    'company': ['Company A', 'Company B', 'Company C'],
    'product_description': [
        'High-quality software development services',
        'Innovative marketing solutions',
        'Efficient supply chain management'
    ]
}

df = pd.DataFrame(data)

# 使用TF-IDF向量化产品描述
tfidf = TfidfVectorizer(stop_words='english')
df['product_description'] = df['product_description'].fillna('')
tfidf_matrix = tfidf.fit_transform(df['product_description'])

# 计算产品描述之间的相似度
cosine_sim = linear_kernel(tfidf_matrix, tfidf_matrix)

# 推荐函数
def get_recommendations(company_name, cosine_sim=cosine_sim):
    idx = df.index[df['company'] == company_name].tolist()[0]
    sim_scores = list(enumerate(cosine_sim[idx]))
    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
    sim_scores = sim_scores[1:3]  # 获取最相似的两个产品
    product_indices = [i[0] for i in sim_scores]
    return df['product_description'].iloc[product_indices]

# 示例调用
print(get_recommendations('Company A'))