AI论文查重源码

# AI论文查重核心算法示例
import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

class PlagiarismDetector:
    def __init__(self):
        self.vectorizer = TfidfVectorizer()
        
    def train(self, corpus):
        """训练模型"""
        self.tfidf_matrix = self.vectorizer.fit_transform(corpus)
        
    def check_similarity(self, text, threshold=0.8):
        """检测文本相似度"""
        text_vector = self.vectorizer.transform([text])
        similarities = cosine_similarity(text_vector, self.tfidf_matrix)
        max_similarity = np.max(similarities)
        
        if max_similarity >= threshold:
            return {
                'is_plagiarized': True,
                'similarity_score': max_similarity,
                'suggestion': '建议修改相关内容'
            }
        else:
            return {
                'is_plagiarized': False,
                'similarity_score': max_similarity,
                'suggestion': '内容原创性良好'
            }

# 使用示例
detector = PlagiarismDetector()
corpus = ["这是原始论文内容", "这是另一篇论文"]
detector.train(corpus)

result = detector.check_similarity("这是待检测的论文")
print(result)