File size: 8,189 Bytes
3d015cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
"""Research/Science Domain Plugin

Scores research competency based on:
- Publication record (papers, citations)
- Lab experience and duration
- Research project depth
- Thesis/dissertation summaries
"""
import re
import time
import logging
from typing import Dict, List
from .base_plugin import BaseDomainPlugin, DomainScore
from .plugin_factory import register_plugin

logger = logging.getLogger(__name__)


@register_plugin('research')
class ResearchPlugin(BaseDomainPlugin):
    """Research/Science domain scoring plugin"""
    
    def __init__(self):
        super().__init__()
        # Research indicators
        self.publication_venues = [
            'journal', 'conference', 'proceedings', 'ieee', 'acm',
            'springer', 'elsevier', 'nature', 'science', 'arxiv'
        ]
        self.research_methods = [
            'experiment', 'methodology', 'hypothesis', 'literature review',
            'data collection', 'statistical analysis', 'simulation', 'survey'
        ]
    
    def _get_domain_type(self) -> str:
        return 'research'
    
    def _get_feature_weights(self) -> Dict[str, float]:
        return {
            'publication_score': 0.35,
            'lab_experience_score': 0.25,
            'research_depth_score': 0.25,
            'thesis_quality_score': 0.15
        }
    
    def get_required_fields(self) -> List[str]:
        return ['research_description']
    
    def get_optional_fields(self) -> List[str]:
        return ['publications_text', 'lab_experience_text', 'thesis_summary']
    
    def score(self, evidence_data: Dict) -> DomainScore:
        """Calculate research domain score"""
        start_time = time.time()
        features = {}
        
        # Publication analysis
        publications = evidence_data.get('publications_text', '')
        features['publication_score'] = self._analyze_publications(publications)
        
        # Lab experience
        lab_exp = evidence_data.get('lab_experience_text', '')
        features['lab_experience_score'] = self._analyze_lab_experience(lab_exp)
        
        # Research depth from main description
        research_desc = evidence_data.get('research_description', '')
        features['research_depth_score'] = self._analyze_research_depth(research_desc)
        
        # Thesis quality
        thesis = evidence_data.get('thesis_summary', '')
        features['thesis_quality_score'] = self._analyze_thesis(thesis)
        
        # Calculate weighted score
        score = sum(features[k] * self.feature_weights[k] for k in features.keys())
        
        # Calculate confidence
        confidence = self.calculate_confidence(evidence_data)
        
        processing_time = (time.time() - start_time) * 1000
        
        return DomainScore(
            domain_type='research',
            score=min(score, 1.0),
            confidence=confidence,
            raw_features=features,
            processing_time_ms=processing_time
        )
    
    def _analyze_publications(self, publications_text: str) -> float:
        """
        Analyze publication record
        Returns: 0-1 score based on number and quality of publications
        """
        if not publications_text or len(publications_text) < 30:
            return 0.0
        
        text_lower = publications_text.lower()
        score = 0.0
        
        # Count publication mentions (by common patterns)
        # Pattern: "Paper title" or [1] Reference format
        title_patterns = [
            r'"([^"]+)"',  # Quoted titles
            r'\[\d+\]',     # Numbered references
            r'\d{4}\.\s',   # Year format (2023. Title...)
        ]
        
        pub_count = 0
        for pattern in title_patterns:
            matches = re.findall(pattern, publications_text)
            pub_count = max(pub_count, len(matches))
        
        # Score based on publication count
        count_score = min(pub_count / 5, 0.6)  # 5+ pubs = 0.6
        score += count_score
        
        # Venue quality bonus
        venue_count = sum(1 for venue in self.publication_venues if venue in text_lower)
        venue_score = min(venue_count / 3, 0.4)  # 3+ venues = 0.4
        score += venue_score
        
        logger.info(f"Publication score: {score:.2f} ({pub_count} pubs, {venue_count} venues)")
        return min(score, 1.0)
    
    def _analyze_lab_experience(self, lab_text: str) -> float:
        """
        Analyze laboratory experience
        Returns: 0-1 score based on duration and depth
        """
        if not lab_text or len(lab_text) < 30:
            return 0.0
        
        text_lower = lab_text.lower()
        score = 0.0
        
        # Extract duration (months/years)
        duration_patterns = [
            (r'(\d+)\s*years?', 12),  # Convert years to months
            (r'(\d+)\s*months?', 1),
        ]
        
        max_duration = 0
        for pattern, multiplier in duration_patterns:
            matches = re.findall(pattern, text_lower)
            if matches:
                duration = max([int(m) * multiplier for m in matches])
                max_duration = max(max_duration, duration)
        
        # Duration score (12 months = max)
        duration_score = min(max_duration / 12, 0.5)
        score += duration_score
        
        # Lab quality indicators
        quality_keywords = ['research lab', 'professor', 'phd', 'equipment', 'experiment', 'protocol']
        quality_count = sum(1 for kw in quality_keywords if kw in text_lower)
        quality_score = min(quality_count / 4, 0.5)
        score += quality_score
        
        logger.info(f"Lab experience: {score:.2f} ({max_duration} months)")
        return min(score, 1.0)
    
    def _analyze_research_depth(self, research_desc: str) -> float:
        """
        Analyze research methodology depth
        Returns: 0-1 score based on methodology sophistication
        """
        if not research_desc or len(research_desc) < 50:
            return 0.0
        
        text_lower = research_desc.lower()
        score = 0.0
        
        # Research method mentions
        method_count = sum(1 for method in self.research_methods if method in text_lower)
        method_score = min(method_count / 4, 0.5)
        score += method_score
        
        # Technical depth indicators
        technical_terms = [
            'algorithm', 'model', 'framework', 'dataset', 'validation',
            'baseline', 'benchmark', 'evaluation', 'metrics', 'results'
        ]
        tech_count = sum(1 for term in technical_terms if term in text_lower)
        tech_score = min(tech_count / 5, 0.3)
        score += tech_score
        
        # Length as depth proxy
        length_score = min(len(research_desc) / 1000, 0.2)
        score += length_score
        
        logger.info(f"Research depth: {score:.2f}")
        return min(score, 1.0)
    
    def _analyze_thesis(self, thesis_text: str) -> float:
        """
        Analyze thesis/dissertation quality
        Returns: 0-1 score based on structure and depth
        """
        if not thesis_text or len(thesis_text) < 100:
            return 0.0
        
        text_lower = thesis_text.lower()
        score = 0.0
        
        # Thesis structure keywords
        structure_keywords = [
            'abstract', 'introduction', 'methodology', 'results',
            'discussion', 'conclusion', 'references', 'chapter'
        ]
        structure_count = sum(1 for kw in structure_keywords if kw in text_lower)
        structure_score = min(structure_count / 5, 0.5)
        score += structure_score
        
        # Academic rigor indicators
        rigor_keywords = [
            'research question', 'objective', 'contribution', 'limitation',
            'future work', 'significance', 'novelty', 'finding'
        ]
        rigor_count = sum(1 for kw in rigor_keywords if kw in text_lower)
        rigor_score = min(rigor_count / 4, 0.3)
        score += rigor_score
        
        # Length bonus
        length_score = min(len(thesis_text) / 2000, 0.2)
        score += length_score
        
        logger.info(f"Thesis quality: {score:.2f}")
        return min(score, 1.0)