双12文档识别推荐

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

# OCR识别函数
def ocr_recognition(image_path):
    image = Image.open(image_path)
    text = pytesseract.image_to_string(image)
    return text

# 示例：读取商品描述数据
products_df = pd.read_csv('products.csv')

# OCR识别商品描述
image_path = 'product_image.jpg'
product_description = ocr_recognition(image_path)

# 将新商品描述加入数据集
new_product = {'name': 'New Product', 'description': product_description}
products_df = products_df.append(new_product, ignore_index=True)

# 使用TF-IDF向量化商品描述
tfidf = TfidfVectorizer(stop_words='english')
tfidf_matrix = tfidf.fit_transform(products_df['description'])

# 计算相似度矩阵
cosine_sim = linear_kernel(tfidf_matrix, tfidf_matrix)

# 推荐相似商品
def get_recommendations(title, cosine_sim=cosine_sim):
    idx = products_df.index[products_df['name'] == title].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[1:6]  # 取前5个最相似的商品
    product_indices = [i[0] for i in sim_scores]
    return products_df['name'].iloc[product_indices]

recommended_products = get_recommendations('New Product')
print("Recommended Products:", recommended_products)