File size: 19,908 Bytes
dcc24f8 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 |
"""
Financial Entity Extractor Module.
This module provides enterprise-grade extraction of financial entities from
transaction emails across multiple Indian banks and payment platforms.
Supported Banks:
- HDFC Bank
- ICICI Bank
- State Bank of India (SBI)
- Axis Bank
- Kotak Mahindra Bank
Supported Payment Platforms:
- PhonePe
- Google Pay (GPay)
- Paytm
- BHIM UPI
Example:
>>> from src.data.extractor import EntityExtractor
>>> extractor = EntityExtractor()
>>> result = extractor.extract("Rs.2500 debited from A/c 3545 on 05-01-26")
>>> print(result.to_dict())
{'amount': '2500', 'type': 'debit', 'account': '3545', 'date': '05-01-26'}
Author: Ranjit Behera
License: MIT
"""
from __future__ import annotations
import logging
import re
from dataclasses import dataclass, field, asdict
from typing import (
Dict,
List,
Optional,
Pattern,
Tuple,
Any,
ClassVar,
)
# Configure module logger
logger = logging.getLogger(__name__)
@dataclass
class FinancialEntity:
"""
Represents extracted financial entities from a transaction message.
This dataclass holds all extracted fields from a financial transaction
notification, including amount, type, account details, and optional
merchant/category information.
Attributes:
amount: Transaction amount as string (preserves decimal precision).
type: Transaction type - 'debit' or 'credit'.
account: Account number (usually last 4 digits, masked).
date: Transaction date in original format from message.
reference: UPI/IMPS/NEFT reference number.
merchant: Identified merchant name (e.g., 'swiggy', 'amazon').
payment_method: Payment method - 'upi', 'neft', 'imps', 'card'.
category: Transaction category - 'food', 'shopping', 'transport', etc.
bank: Source bank name if identified.
balance: Available balance after transaction.
raw_text: Original text used for extraction (for debugging).
Example:
>>> entity = FinancialEntity(
... amount="2500.00",
... type="debit",
... account="3545",
... date="05-01-26",
... merchant="swiggy",
... category="food"
... )
>>> entity.is_valid()
True
>>> entity.to_dict()
{'amount': '2500.00', 'type': 'debit', ...}
"""
amount: Optional[str] = None
type: Optional[str] = None
account: Optional[str] = None
date: Optional[str] = None
reference: Optional[str] = None
merchant: Optional[str] = None
payment_method: Optional[str] = None
category: Optional[str] = None
bank: Optional[str] = None
balance: Optional[str] = None
raw_text: str = field(default="", repr=False)
# Validation constants
VALID_TYPES: ClassVar[set] = {"debit", "credit"}
VALID_PAYMENT_METHODS: ClassVar[set] = {"upi", "neft", "imps", "rtgs", "card", "wallet"}
VALID_CATEGORIES: ClassVar[set] = {
"food", "shopping", "transport", "bills", "grocery",
"entertainment", "health", "education", "transfer", "other"
}
def __post_init__(self) -> None:
"""Validate and normalize fields after initialization."""
# Normalize type to lowercase
if self.type:
self.type = self.type.lower()
# Normalize payment method
if self.payment_method:
self.payment_method = self.payment_method.lower()
# Normalize category
if self.category:
self.category = self.category.lower()
def is_valid(self) -> bool:
"""
Check if the entity has minimum required fields.
A valid entity must have at least an amount and transaction type.
Returns:
bool: True if entity has minimum required fields.
Example:
>>> entity = FinancialEntity(amount="100", type="debit")
>>> entity.is_valid()
True
>>> entity = FinancialEntity(amount="100")
>>> entity.is_valid()
False
"""
return bool(self.amount and self.type)
def is_complete(self) -> bool:
"""
Check if the entity has all core fields populated.
A complete entity has amount, type, account, and date.
Returns:
bool: True if entity has all core fields.
"""
return bool(
self.amount and
self.type and
self.account and
self.date
)
def to_dict(self) -> Dict[str, Any]:
"""
Convert entity to dictionary, excluding None values.
Returns:
Dict[str, Any]: Dictionary with non-None fields only.
Example:
>>> entity = FinancialEntity(amount="500", type="debit")
>>> entity.to_dict()
{'amount': '500', 'type': 'debit'}
"""
return {
key: value
for key, value in asdict(self).items()
if value is not None and key != "raw_text"
}
def to_json_string(self) -> str:
"""
Convert entity to JSON string for model training.
Returns:
str: JSON representation of the entity.
"""
import json
return json.dumps(self.to_dict(), indent=2)
def confidence_score(self) -> float:
"""
Calculate confidence score based on populated fields.
Returns:
float: Score between 0.0 and 1.0.
"""
required_fields = ["amount", "type"]
optional_fields = ["account", "date", "reference", "merchant", "category"]
required_score = sum(
1 for f in required_fields
if getattr(self, f) is not None
) / len(required_fields)
optional_score = sum(
1 for f in optional_fields
if getattr(self, f) is not None
) / len(optional_fields)
return required_score * 0.6 + optional_score * 0.4
class EntityExtractor:
"""
Production-grade financial entity extractor using regex patterns.
This extractor uses a comprehensive set of regex patterns to extract
structured financial data from transaction notifications. It supports
multiple Indian banks and payment platforms with high accuracy.
Features:
- Amount extraction (Rs., ₹, INR formats)
- Transaction type detection (debit/credit)
- Account number extraction (masked formats)
- Date parsing (multiple formats)
- Reference number extraction
- Merchant identification
- Payment method detection
- Category classification
Attributes:
AMOUNT_PATTERNS: Compiled regex patterns for amount extraction.
DEBIT_KEYWORDS: Keywords indicating debit transactions.
CREDIT_KEYWORDS: Keywords indicating credit transactions.
MERCHANTS: Merchant name to keyword mapping.
CATEGORIES: Category to merchant mapping.
Example:
>>> extractor = EntityExtractor()
>>> result = extractor.extract(
... "HDFC Bank: Rs.2500.00 debited from A/c **3545 on 05-01-26"
... )
>>> print(result.amount)
'2500.00'
>>> print(result.merchant)
None # Would need VPA for merchant detection
Note:
For best results, pass the complete transaction message including
sender information and subject line when available.
"""
# Amount extraction patterns (ordered by specificity)
AMOUNT_PATTERNS: ClassVar[List[Pattern]] = [
re.compile(r'(?:Rs\.?|INR|₹)\s*([\d,]+(?:\.\d{2})?)', re.IGNORECASE),
re.compile(r'(?:amount|amt)[:\s]*([\d,]+(?:\.\d{2})?)', re.IGNORECASE),
re.compile(r'(?:debited|credited)[:\s]*(?:Rs\.?|INR|₹)?\s*([\d,]+(?:\.\d{2})?)', re.IGNORECASE),
]
# Transaction type keywords
DEBIT_KEYWORDS: ClassVar[set] = {
'debited', 'debit', 'paid', 'sent', 'withdrawn',
'purchase', 'payment', 'transferred out', 'dr'
}
CREDIT_KEYWORDS: ClassVar[set] = {
'credited', 'credit', 'received', 'deposited',
'refund', 'cashback', 'transferred in', 'cr'
}
# Account extraction patterns
ACCOUNT_PATTERNS: ClassVar[List[Pattern]] = [
re.compile(r'(?:a/c|acct?|account)\s*(?:no\.?)?\s*[:\s]*\**[xX]*(\d{4,})', re.IGNORECASE),
re.compile(r'[*xX]+(\d{4})\b'),
re.compile(r'(?:XX|xx)(\d{4})\b'),
]
# Date extraction patterns
DATE_PATTERNS: ClassVar[List[Pattern]] = [
re.compile(r'(\d{2}[-/]\d{2}[-/]\d{2,4})'),
re.compile(r'(\d{2}\s+(?:Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec)\s+\d{4})', re.IGNORECASE),
re.compile(r'(\d{4}[-/]\d{2}[-/]\d{2})'),
]
# Reference number patterns
REFERENCE_PATTERNS: ClassVar[List[Pattern]] = [
re.compile(r'(?:ref(?:erence)?|txn|upi|imps)\s*(?:no\.?|id)?[:\s]*(\d{10,})', re.IGNORECASE),
re.compile(r'(?:transaction)\s*(?:id)?[:\s]*(\d{10,})', re.IGNORECASE),
]
# Merchant identification
MERCHANTS: ClassVar[Dict[str, List[str]]] = {
# Food delivery
'swiggy': ['swiggy', 'swiggy@'],
'zomato': ['zomato', 'zomato@'],
'dominos': ['dominos', 'domino'],
'mcdonalds': ['mcdonald', 'mcd@'],
'kfc': ['kfc@', 'kfc '],
'starbucks': ['starbucks', 'sbux'],
# E-commerce
'amazon': ['amazon', 'amzn', 'amazon@'],
'flipkart': ['flipkart', 'fkrt'],
'myntra': ['myntra'],
'ajio': ['ajio'],
'nykaa': ['nykaa'],
# Transport
'uber': ['uber'],
'ola': ['ola@', 'olacabs'],
'rapido': ['rapido'],
'metro': ['metro', 'dmrc', 'bmrc'],
# Bills & Utilities
'airtel': ['airtel'],
'jio': ['jio@', 'reliancejio'],
'vodafone': ['vodafone', 'vi@'],
'electricity': ['bescom', 'electricity', 'power'],
'gas': ['indane', 'bharat gas', 'hp gas'],
# Grocery
'bigbasket': ['bigbasket', 'bb@'],
'zepto': ['zepto'],
'blinkit': ['blinkit', 'grofers'],
'dmart': ['dmart', 'd-mart'],
}
# Category mapping
CATEGORY_KEYWORDS: ClassVar[Dict[str, List[str]]] = {
'food': ['swiggy', 'zomato', 'dominos', 'mcdonald', 'kfc', 'restaurant', 'cafe', 'food'],
'shopping': ['amazon', 'flipkart', 'myntra', 'ajio', 'nykaa', 'shopping'],
'transport': ['uber', 'ola', 'rapido', 'metro', 'cab', 'taxi', 'fuel', 'petrol'],
'bills': ['airtel', 'jio', 'vodafone', 'electricity', 'water', 'gas', 'broadband'],
'grocery': ['bigbasket', 'zepto', 'blinkit', 'dmart', 'grocery', 'supermarket'],
'entertainment': ['netflix', 'prime', 'hotstar', 'spotify', 'movie', 'theatre'],
'health': ['pharmacy', 'medical', 'hospital', 'doctor', 'health'],
'education': ['school', 'college', 'course', 'udemy', 'education'],
}
# Payment method patterns
PAYMENT_PATTERNS: ClassVar[Dict[str, List[str]]] = {
'upi': ['upi', 'vpa', '@ybl', '@oksbi', '@okicici', '@paytm', '@axisbank', '@icici'],
'neft': ['neft'],
'imps': ['imps'],
'rtgs': ['rtgs'],
'card': ['card', 'visa', 'mastercard', 'rupay', 'credit card', 'debit card'],
'wallet': ['wallet', 'paytm wallet', 'phonepe wallet'],
}
# Bank identification
BANK_PATTERNS: ClassVar[Dict[str, List[str]]] = {
'hdfc': ['hdfc', 'hdfcbank'],
'icici': ['icici'],
'sbi': ['sbi', 'state bank'],
'axis': ['axis'],
'kotak': ['kotak'],
'pnb': ['pnb', 'punjab national'],
'bob': ['bob', 'bank of baroda'],
'canara': ['canara'],
'union': ['union bank'],
'idbi': ['idbi'],
}
def __init__(self, debug: bool = False) -> None:
"""
Initialize the EntityExtractor.
Args:
debug: If True, enables debug logging.
Example:
>>> extractor = EntityExtractor(debug=True)
"""
self.debug = debug
if debug:
logger.setLevel(logging.DEBUG)
logger.info("EntityExtractor initialized")
def extract(self, text: str) -> FinancialEntity:
"""
Extract financial entities from transaction text.
This is the main entry point for entity extraction. It processes
the input text and returns a FinancialEntity object with all
detected fields populated.
Args:
text: The transaction message text to process.
Returns:
FinancialEntity: Extracted entity with populated fields.
Example:
>>> extractor = EntityExtractor()
>>> result = extractor.extract(
... "Rs.2500 debited from A/c 3545 to swiggy@ybl on 05-01-26"
... )
>>> print(result.amount)
'2500'
>>> print(result.merchant)
'swiggy'
Note:
The method never raises exceptions. On failure, it returns
an entity with None fields and logs the error.
"""
if not text or not isinstance(text, str):
logger.warning("Empty or invalid text provided")
return FinancialEntity(raw_text=str(text) if text else "")
text_lower = text.lower()
try:
entity = FinancialEntity(
amount=self._extract_amount(text),
type=self._extract_type(text_lower),
account=self._extract_account(text),
date=self._extract_date(text),
reference=self._extract_reference(text),
merchant=self._extract_merchant(text_lower),
payment_method=self._extract_payment_method(text_lower),
category=self._extract_category(text_lower),
bank=self._extract_bank(text_lower),
balance=self._extract_balance(text),
raw_text=text,
)
logger.debug(f"Extracted entity: {entity.to_dict()}")
return entity
except Exception as e:
logger.error(f"Extraction failed: {e}", exc_info=True)
return FinancialEntity(raw_text=text)
def extract_to_dict(self, text: str) -> Dict[str, Any]:
"""
Extract entities and return as dictionary.
Convenience method that extracts and converts to dict in one call.
Args:
text: The transaction message text.
Returns:
Dict[str, Any]: Dictionary of extracted entities.
"""
return self.extract(text).to_dict()
def _extract_amount(self, text: str) -> Optional[str]:
"""Extract transaction amount from text."""
for pattern in self.AMOUNT_PATTERNS:
match = pattern.search(text)
if match:
amount = match.group(1).replace(',', '')
logger.debug(f"Found amount: {amount}")
return amount
return None
def _extract_type(self, text_lower: str) -> Optional[str]:
"""Determine if transaction is debit or credit."""
# Check debit keywords first (more common)
for keyword in self.DEBIT_KEYWORDS:
if keyword in text_lower:
return 'debit'
# Then check credit keywords
for keyword in self.CREDIT_KEYWORDS:
if keyword in text_lower:
return 'credit'
return None
def _extract_account(self, text: str) -> Optional[str]:
"""Extract account number from text."""
for pattern in self.ACCOUNT_PATTERNS:
match = pattern.search(text)
if match:
account = match.group(1)
# Return last 4 digits only
return account[-4:] if len(account) > 4 else account
return None
def _extract_date(self, text: str) -> Optional[str]:
"""Extract transaction date from text."""
for pattern in self.DATE_PATTERNS:
match = pattern.search(text)
if match:
return match.group(1)
return None
def _extract_reference(self, text: str) -> Optional[str]:
"""Extract reference/transaction number."""
for pattern in self.REFERENCE_PATTERNS:
match = pattern.search(text)
if match:
return match.group(1)
return None
def _extract_merchant(self, text_lower: str) -> Optional[str]:
"""Identify merchant from text."""
for merchant, keywords in self.MERCHANTS.items():
for keyword in keywords:
if keyword in text_lower:
return merchant
return None
def _extract_payment_method(self, text_lower: str) -> Optional[str]:
"""Detect payment method."""
for method, keywords in self.PAYMENT_PATTERNS.items():
for keyword in keywords:
if keyword in text_lower:
return method
return None
def _extract_category(self, text_lower: str) -> Optional[str]:
"""Classify transaction category."""
for category, keywords in self.CATEGORY_KEYWORDS.items():
for keyword in keywords:
if keyword in text_lower:
return category
return None
def _extract_bank(self, text_lower: str) -> Optional[str]:
"""Identify source bank."""
for bank, keywords in self.BANK_PATTERNS.items():
for keyword in keywords:
if keyword in text_lower:
return bank
return None
def _extract_balance(self, text: str) -> Optional[str]:
"""Extract available balance if mentioned."""
patterns = [
re.compile(r'(?:bal(?:ance)?|avl\.?\s*bal)[:\s]*(?:Rs\.?|INR|₹)?\s*([\d,]+(?:\.\d{2})?)', re.IGNORECASE),
]
for pattern in patterns:
match = pattern.search(text)
if match:
return match.group(1).replace(',', '')
return None
# Module-level convenience function
def extract_entities(text: str) -> Dict[str, Any]:
"""
Convenience function to extract entities without instantiating class.
Args:
text: Transaction message text.
Returns:
Dict[str, Any]: Extracted entities as dictionary.
Example:
>>> from src.data.extractor import extract_entities
>>> result = extract_entities("Rs.500 debited from account 1234")
>>> print(result['amount'])
'500'
"""
return EntityExtractor().extract_to_dict(text)
if __name__ == "__main__":
# Self-test when run directly
logging.basicConfig(level=logging.DEBUG)
extractor = EntityExtractor(debug=True)
test_cases = [
"HDFC Bank: Rs.2500.00 debited from A/c **3545 on 05-01-26 to VPA swiggy@ybl. Ref: 123456789012",
"Dear Customer, INR 45000 credited to A/c 7890 on 04-01-2026. Salary from ACME Corp.",
"You paid Rs.599 to Amazon from HDFC Bank a/c XX4567. UPI Ref: 987654321012",
]
for i, test in enumerate(test_cases, 1):
print(f"\n{'='*60}")
print(f"Test {i}: {test[:50]}...")
result = extractor.extract(test)
print(f"Result: {result.to_dict()}")
print(f"Valid: {result.is_valid()}, Confidence: {result.confidence_score():.2%}")
|