import pandas as pd

# 加載校友數(shù)據(jù)

alumni_data = pd.read_csv('alumni.csv')

# 去除重復項

cleaned_data = alumni_data.drop_duplicates()

# 填補缺失值

cleaned_data.fillna(value={'email': 'unknown@university.edu'}, inplace=True)

# 保存清理后的數(shù)據(jù)

cleaned_data.to_csv('cleaned_alumni.csv', index=False)

]]>

from transformers import pipeline

# 初始化問答模型

qa_model = pipeline('question-answering', model='distilbert-base-cased-distilled-squad')

# 定義問題和上下文

question = "誰是2010年的畢業(yè)生？"

context = "Alumni Database contains records of all graduates since 2010."

# 獲取答案

answer = qa_model(question=question, context=context)

print(f"Answer: {answer['answer']}")

]]>

import numpy as np

from sklearn.metrics.pairwise import cosine_similarity

# 示例校友特征向量

alumni_features = np.array([[...], [...]])

# 用戶興趣向量

user_interests = np.array([...])

# 計算相似度

similarity_scores = cosine_similarity([user_interests], alumni_features)[0]

# 獲取最相似的校友索引

top_indices = np.argsort(similarity_scores)[-5:]

recommended_alumni = [alumni_list[i] for i in top_indices]

print(f"Recommended Alumni: {recommended_alumni}")

]]>