### 实现类似知网的论文查重功能 以下是基于 Python 的完整解决方案，涵盖了 Word 文件处理、重复内容高亮显示以及生成 PDF 查重报告的核心逻辑。该方案综合了文本相似度计算、文件转换和可视化技术。 --- #### 1. **Word 文件处理** 利用 `python-docx` 库读取 Word 文件的内容，并将其存储为字符串列表以便进一步分析[^1]。 ```python from docx import Document def read_docx(file_path): document = Document(file_path) paragraphs = [" ".join([run.text for run in paragraph.runs]) for paragraph in document.paragraphs] return "\n".join(paragraphs).strip() ``` --- #### 2. **文本相似度计算** 采用 SimHash 或 TF-IDF 方法计算两个文档间的相似度。此处选用 SimHash 来快速判断大段文本是否高度相似[^3]。 ```python import simhash def calculate_simhash_similarity(text1, text2): hash1 = simhash.Simhash(text1.split()) hash2 = simhash.Simhash(text2.split()) similarity = hash1.distance(hash2) / max(len(hash1.hash), len(hash2.hash)) return similarity ``` --- #### 3. **重复内容标记** 当检测到某部分内容与其他文档存在较高相似度时，在原 Word 文件中标记这些片段为红色字体[^1]。 ```python from docx.shared import RGBColor def highlight_duplicates(input_file, output_file, duplicate_texts): document = Document(input_file) for paragraph in document.paragraphs: runs = [] current_run_text = "" for i, char in enumerate(paragraph.text): current_run_text += char if any(dup_text in current_run_text for dup_text in duplicate_texts): run = paragraph.add_run(current_run_text) font = run.font font.color.rgb = RGBColor(255, 0, 0) # 设置颜色为红色 elif char == " ": # 处理空白符作为分隔条件 continue paragraph.clear() # 清除原有内容 document.save(output_file) ``` --- #### 4. **生成 PDF 报告** 借助 `reportlab` 将最终结果导出至 PDF 格式，便于用户查看整体情况[^5]。 ```python from reportlab.lib.pagesizes import letter from reportlab.pdfgen import canvas def generate_pdf_report(report_data, file_name="detection_report"): c = canvas.Canvas(f"{file_name}.pdf", pagesize=letter) width, height = letter y_position = height - 50 for line in report_data.split("\n"): c.drawString(50, y_position, line[:100]) y_position -= 20 if y_position < 50: c.showPage() y_position = height - 50 c.save() ``` --- #### 完整流程整合 将上述各部分串联起来形成一个完整的查重工具链： ```python if __name__ == "__main__": input_word_file = "example.docx" original_content = read_docx(input_word_file) comparison_files = ["comparison_1.txt", "comparison_2.txt"] all_duplicate_segments = [] for comp_file in comparison_files: compared_content = open(comp_file, 'r').read().strip() similarity_ratio = calculate_simhash_similarity(original_content, compared_content) if similarity_ratio >= 0.8: # 设定阈值 all_duplicate_segments.append(compared_content) highlighted_output = f"highlighted_{input_word_file}" highlight_duplicates(input_word_file, highlighted_output, all_duplicate_segments) final_report = "\n".join(all_duplicate_segments) generate_pdf_report(final_report, "final_detection_report") ``` --- ###