Spaces:
Sleeping
Sleeping
File size: 10,965 Bytes
7d42424 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 | """
Data Loader Module
Handles loading and preprocessing of bias detection datasets.
"""
import json
import os
from typing import List, Dict, Tuple
import pandas as pd
class DataLoader:
"""
Load and preprocess bias detection datasets.
"""
def __init__(self, data_dir='data'):
"""
Initialize data loader.
Args:
data_dir: Directory containing datasets
"""
self.data_dir = data_dir
def load_winobias(self, subset='dev') -> List[Dict]:
"""
Load WinoBias English dataset.
Args:
subset: 'dev' or 'test'
Returns:
List of sentences with bias information
"""
file_path = os.path.join(self.data_dir, 'english', 'winobias_sample.json')
if not os.path.exists(file_path):
print(f"Warning: {file_path} not found. Using default samples.")
return self._get_default_english_samples()
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
return data.get('sentences', [])
def load_arabic_dataset(self) -> List[Dict]:
"""
Load Arabic bias dataset.
Returns:
List of Arabic sentences with bias information
"""
file_path = os.path.join(self.data_dir, 'arabic', 'arabic_bias_sample.json')
if not os.path.exists(file_path):
print(f"Warning: {file_path} not found. Using default samples.")
return self._get_default_arabic_samples()
with open(file_path, 'r', encoding='utf-8') as f:
data = json.load(f)
return data.get('sentences', [])
def _get_default_english_samples(self) -> List[Dict]:
"""Get default English bias samples."""
return [
{
"text": "The doctor asked the nurse to help him with the procedure.",
"bias_type": "gender",
"stereotype": "male_doctor",
"label": "stereotype"
},
{
"text": "The nurse helped the doctor with her medical expertise.",
"bias_type": "gender",
"stereotype": "female_nurse",
"label": "stereotype"
},
{
"text": "The engineer explained the software to the secretary and thanked her.",
"bias_type": "gender",
"stereotype": "male_engineer_female_secretary",
"label": "stereotype"
},
{
"text": "The CEO met with the assistant and asked him to schedule a meeting.",
"bias_type": "gender",
"stereotype": "counter",
"label": "anti-stereotype"
},
{
"text": "The software developer presented her innovative solution to the team.",
"bias_type": "gender",
"stereotype": "counter",
"label": "anti-stereotype"
},
{
"text": "The construction worker completed his shift and went home.",
"bias_type": "gender",
"stereotype": "male_construction",
"label": "stereotype"
},
{
"text": "The teacher graded papers and prepared tomorrow's lesson.",
"bias_type": "neutral",
"stereotype": "none",
"label": "unrelated"
},
{
"text": "The pilot safely landed the plane after checking all systems.",
"bias_type": "neutral",
"stereotype": "none",
"label": "unrelated"
}
]
def _get_default_arabic_samples(self) -> List[Dict]:
"""Get default Arabic bias samples."""
return [
{
"text": "طلب الطبيب من الممرضة أن تساعده في الإجراء.",
"bias_type": "gender",
"stereotype": "male_doctor_female_nurse",
"label": "stereotype"
},
{
"text": "ساعدت الممرضة الطبيب بخبرتها الطبية.",
"bias_type": "gender",
"stereotype": "female_nurse",
"label": "stereotype"
},
{
"text": "شرح المهندس البرنامج للسكرتيرة وشكرها.",
"bias_type": "gender",
"stereotype": "male_engineer_female_secretary",
"label": "stereotype"
},
{
"text": "قابل المدير التنفيذي المساعد وطلب منه جدولة اجتماع.",
"bias_type": "gender",
"stereotype": "counter",
"label": "anti-stereotype"
},
{
"text": "قدمت مطورة البرمجيات حلها المبتكر للفريق.",
"bias_type": "gender",
"stereotype": "counter",
"label": "anti-stereotype"
},
{
"text": "أكمل عامل البناء وردية عمله وعاد إلى المنزل.",
"bias_type": "gender",
"stereotype": "male_construction",
"label": "stereotype"
},
{
"text": "قام المعلم بتصحيح الأوراق وإعداد درس الغد.",
"bias_type": "neutral",
"stereotype": "none",
"label": "unrelated"
},
{
"text": "هبط الطيار بالطائرة بأمان بعد فحص جميع الأنظمة.",
"bias_type": "neutral",
"stereotype": "none",
"label": "unrelated"
}
]
def load_dataset(self, language='english') -> List[Dict]:
"""
Load dataset for specified language.
Args:
language: 'english' or 'arabic'
Returns:
List of sentences
"""
if language.lower() == 'arabic':
return self.load_arabic_dataset()
else:
return self.load_winobias()
def create_filtered_version(self, sentences: List[Dict], bias_detector) -> List[Dict]:
"""
Create filtered versions of sentences with reduced bias.
Args:
sentences: List of sentence dictionaries
bias_detector: BiasDetector instance
Returns:
List of filtered sentences
"""
filtered = []
for item in sentences:
text = item['text']
# Simple filtering: replace gendered pronouns with neutral alternatives
filtered_text = self._neutralize_gender(text, bias_detector.language)
filtered_item = item.copy()
filtered_item['original_text'] = text
filtered_item['text'] = filtered_text
filtered_item['is_filtered'] = True
filtered.append(filtered_item)
return filtered
def _neutralize_gender(self, text: str, language: str) -> str:
"""
Apply simple gender neutralization to text.
Args:
text: Input text
language: 'english' or 'arabic'
Returns:
Neutralized text
"""
if language == 'english':
replacements = {
' he ': ' they ',
' she ': ' they ',
' him ': ' them ',
' her ': ' them ',
' his ': ' their ',
' hers ': ' theirs ',
'He ': 'They ',
'She ': 'They ',
'Him ': 'Them ',
'Her ': 'Them ',
'His ': 'Their ',
}
else: # Arabic - basic replacements
replacements = {
' هو ': ' هم ',
' هي ': ' هم ',
' له ': ' لهم ',
' لها ': ' لهم ',
}
filtered_text = text
for old, new in replacements.items():
filtered_text = filtered_text.replace(old, new)
return filtered_text
def save_results(self, results: List[Dict], output_path: str):
"""
Save analysis results to file.
Args:
results: List of analysis results
output_path: Path to save file
"""
# Ensure directory exists
os.makedirs(os.path.dirname(output_path), exist_ok=True)
with open(output_path, 'w', encoding='utf-8') as f:
json.dump(results, f, ensure_ascii=False, indent=2)
print(f"Results saved to {output_path}")
def export_to_csv(self, results: List[Dict], output_path: str):
"""
Export results to CSV format.
Args:
results: List of analysis results
output_path: Path to save CSV file
"""
# Flatten nested dictionaries for CSV export
flattened = []
for result in results:
flat_result = {
'text': result.get('text', ''),
'language': result.get('language', ''),
'overall_bias_score': result.get('overall_bias_score', 0),
'is_biased': result.get('is_biased', False),
}
# Add gender bias metrics
if 'gender_bias' in result:
gb = result['gender_bias']
flat_result['gender_bias_score'] = gb.get('bias_score', 0)
flat_result['gender_bias_direction'] = gb.get('bias_direction', '')
flat_result['gender_severity'] = gb.get('severity', '')
# Add sentiment bias metrics
if 'sentiment_bias' in result:
sb = result['sentiment_bias']
flat_result['sentiment_score'] = sb.get('sentiment_score', 0)
flat_result['sentiment_bias_type'] = sb.get('bias_type', '')
flattened.append(flat_result)
df = pd.DataFrame(flattened)
df.to_csv(output_path, index=False, encoding='utf-8-sig')
print(f"Results exported to {output_path}")
|