test_openalex_integration.py

#!/usr/bin/env python3
"""
测试OpenAlex集成到推荐系统的功能
"""

import sys
import os
sys.path.append(os.path.dirname(os.path.abspath(__file__)))

from reviewer_recommendation import (
    PaperInfo, OpenAlexSearcher, DynamicAcademicSearcher, LLMRecommendationEngine
)

def test_openalex_integration():
    """测试OpenAlex集成功能"""
    print("=== 测试OpenAlex集成到推荐系统 ===")
    print("=" * 60)
    
    # 1. 测试OpenAlexSearcher
    print("\n--- 测试OpenAlexSearcher ---")
    openalex_searcher = OpenAlexSearcher(limit=5)
    
    # 测试基本搜索
    results = openalex_searcher.search("machine learning", sort_by_citations=True)
    print(f"OpenAlex搜索返回 {len(results)} 个结果")
    
    if results:
        print("前3个结果:")
        for i, result in enumerate(results[:3], 1):
            title = result.get('title', 'N/A')[:50] + '...' if len(result.get('title', '')) > 50 else result.get('title', 'N/A')
            cited_count = result.get('citedByCount', 0)
            print(f"  {i}. {title} (引用量: {cited_count})")
    
    # 2. 测试DynamicAcademicSearcher with OpenAlex
    print("\n--- 测试DynamicAcademicSearcher with OpenAlex ---")
    dynamic_searcher = DynamicAcademicSearcher(openalex_searcher=openalex_searcher)
    
    # 创建测试论文
    test_paper = PaperInfo(
        title="Deep Learning for Protein Structure Prediction",
        abstract="This paper presents a novel deep learning approach for predicting protein structures using neural networks and molecular dynamics simulations.",
        keywords=["deep learning", "protein structure", "neural networks"],
        authors=["John Smith", "Jane Doe"],
        affiliations=["MIT", "Stanford University"]
    )
    
    # 执行动态查询
    print("执行动态查询...")
    channel1_candidates, channel2_candidates = dynamic_searcher.search_with_dynamic_queries(
        paper=test_paper, 
        num_queries=2
    )
    
    print(f"通道1（高引用量）: {len(channel1_candidates)} 个候选")
    print(f"通道2（相关性）: {len(channel2_candidates)} 个候选")
    
    # 显示通道1的前几个结果
    if channel1_candidates:
        print("\n通道1前3个结果（按引用量排序）:")
        for i, candidate in enumerate(channel1_candidates[:3], 1):
            title = candidate.get('title', 'N/A')[:50] + '...' if len(candidate.get('title', '')) > 50 else candidate.get('title', 'N/A')
            cited_count = candidate.get('citedByCount', 0)
            print(f"  {i}. {title} (引用量: {cited_count})")
    
    # 显示通道2的前几个结果
    if channel2_candidates:
        print("\n通道2前3个结果（按相关性排序）:")
        for i, candidate in enumerate(channel2_candidates[:3], 1):
            title = candidate.get('title', 'N/A')[:50] + '...' if len(candidate.get('title', '')) > 50 else candidate.get('title', 'N/A')
            cited_count = candidate.get('citedByCount', 0)
            print(f"  {i}. {title} (引用量: {cited_count})")
    
    # 3. 测试完整的推荐流程
    print("\n--- 测试完整推荐流程 ---")
    engine = LLMRecommendationEngine()
    
    # 合并候选者
    all_candidates = channel1_candidates + channel2_candidates
    print(f"总候选者数量: {len(all_candidates)}")
    
    if all_candidates:
        # 执行LLM分析
        print("执行LLM分析...")
        try:
            reviewers = engine.analyze_candidates(
                paper=test_paper,
                channel1_candidates=channel1_candidates,
                channel2_candidates=channel2_candidates,
                reviewer_count=3
            )
            
            print(f"推荐了 {len(reviewers)} 个审稿人:")
            for i, reviewer in enumerate(reviewers, 1):
                print(f"  {i}. {reviewer.name} ({reviewer.affiliation})")
                print(f"     推荐理由: {reviewer.reason}")
                print(f"     匹配度: {reviewer.relevance_score}")
                print()
                
        except Exception as e:
            print(f"LLM分析失败: {str(e)}")
    
    print("=" * 60)
    print("OpenAlex集成测试完成")

if __name__ == "__main__":
    test_openalex_integration()