锦中人工智能助手

# 简单的问答机器人示例

questions = {

"你好": "你好！欢迎来到校园问答系统！",

"你是谁": "我是校园问答机器人，可以帮助你解答各种问题。",

"怎么去图书馆": "图书馆在教学楼东侧，步行大约5分钟。",

"今天天气怎么样": "抱歉，我无法获取实时天气信息。",

}

def answer_question(question):

for q in questions:

if question.lower() == q.lower():

return questions[q]

return "抱歉，我不太明白你的问题。"

user_input = input("请输入你的问题：")

print(answer_question(user_input))

from sklearn.feature_extraction.text import TfidfVectorizer

from sklearn.metrics.pairwise import cosine_similarity

answers = [

"你好！欢迎来到校园问答系统！",

"我是校园问答机器人，可以帮助你解答各种问题。",

"图书馆在教学楼东侧，步行大约5分钟。",

"抱歉，我无法获取实时天气信息。",

]

def get_most_similar_answer(question):

vectorizer = TfidfVectorizer()

tfidf_matrix = vectorizer.fit_transform(answers + [question])

similarity_scores = cosine_similarity(tfidf_matrix[-1], tfidf_matrix[:-1])

best_index = similarity_scores.argmax()

return answers[best_index]

user_input = input("请输入你的问题：")

print(get_most_similar_answer(user_input))

def score_answers(question, answers):

scores = []

for answer in answers:

# 基于关键词匹配的得分

keyword_score = sum(1 for word in question.split() if word in answer) / len(question.split())

# 基于余弦相似度的得分

vectorizer = TfidfVectorizer()

tfidf_matrix = vectorizer.fit_transform([question, answer])

cos_sim = cosine_similarity(tfidf_matrix[0], tfidf_matrix[1])[0][0]

# 总得分（可以调整权重）

total_score = keyword_score * 0.4 + cos_sim * 0.6

scores.append((answer, total_score))

# 按得分排序，降序排列

scores.sort(key=lambda x: x[1], reverse=True)

return scores

user_input = input("请输入你的问题：")

ranked_answers = score_answers(user_input, answers)

for i, (answer, score) in enumerate(ranked_answers):

print(f"第{i+1}个答案：{answer}（得分：{score:.2f}）")