Spaces:
Running
Running
File size: 5,912 Bytes
39028c9 | 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 | """
Text preprocessing and tokenization module for technical documents
"""
import re
import logging
from typing import List, Tuple
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import sent_tokenize, word_tokenize
# Download required NLTK resources
try:
nltk.data.find('tokenizers/punkt')
except LookupError:
nltk.download('punkt', quiet=True)
try:
nltk.data.find('corpora/stopwords')
except LookupError:
nltk.download('stopwords', quiet=True)
logger = logging.getLogger(__name__)
class TextPreprocessor:
"""Comprehensive text preprocessing for technical documents."""
def __init__(self, remove_stopwords: bool = False):
"""
Initialize preprocessor.
Args:
remove_stopwords: Whether to remove English stopwords
"""
self.remove_stopwords = remove_stopwords
self.stop_words = set(stopwords.words('english')) if remove_stopwords else set()
def clean_text(self, text: str) -> str:
"""
Clean technical document text.
Args:
text: Raw text to clean
Returns:
Cleaned text
"""
# Remove extra whitespace
text = re.sub(r'\s+', ' ', text)
# Remove URLs
text = re.sub(r'http\S+|www\S+', '', text)
# Remove email addresses
text = re.sub(r'\S+@\S+', '', text)
# Remove special characters but keep punctuation for sentences
text = re.sub(r'[^\w\s.!?,;:\-()]', '', text)
# Remove extra spaces created by above operations
text = re.sub(r'\s+', ' ', text).strip()
return text
def remove_citations(self, text: str) -> str:
"""
Remove citation references from text.
Args:
text: Text with citations
Returns:
Text without citations
"""
# Remove [Author et al., Year] style citations
text = re.sub(r'\[\d+\]|\[[\w\s\.]+,\s*\d{4}\]', '', text)
# Remove (Author Year) style citations
text = re.sub(r'\([\w\s\.]+,?\s*\d{4}\)', '', text)
return text
def remove_equations(self, text: str) -> str:
"""
Remove mathematical equations and formulas.
Args:
text: Text with equations
Returns:
Text without equations
"""
# Remove LaTeX equations
text = re.sub(r'\$\$.*?\$\$', '', text, flags=re.DOTALL)
text = re.sub(r'\$.*?\$', '', text)
return text
def sent_tokenize(self, text: str) -> List[str]:
"""
Tokenize text into sentences.
Args:
text: Input text
Returns:
List of sentences
"""
sentences = sent_tokenize(text)
return [sent.strip() for sent in sentences if sent.strip()]
def word_tokenize(self, text: str) -> List[str]:
"""
Tokenize text into words.
Args:
text: Input text
Returns:
List of words
"""
tokens = word_tokenize(text.lower())
if self.remove_stopwords:
tokens = [t for t in tokens if t not in self.stop_words and t.isalnum()]
return tokens
def preprocess_document(self, text: str, remove_citations: bool = True,
remove_equations: bool = False) -> str:
"""
Complete preprocessing pipeline.
Args:
text: Raw document text
remove_citations: Whether to remove citations
remove_equations: Whether to remove equations
Returns:
Preprocessed text
"""
# Clean text
text = self.clean_text(text)
# Remove citations if requested
if remove_citations:
text = self.remove_citations(text)
# Remove equations if requested
if remove_equations:
text = self.remove_equations(text)
logger.info("Document preprocessing completed")
return text
class TechnicalDocumentParser:
"""Parse technical document structure (sections, abstracts, etc.)."""
@staticmethod
def extract_abstract(text: str) -> Tuple[str, str]:
"""
Extract abstract from document.
Args:
text: Full document text
Returns:
Tuple of (abstract, remaining_text)
"""
abstract_match = re.search(
r'(?:^|\n)(abstract|summary)(.*?)(?:\n(?:introduction|1\.|contents))',
text, re.IGNORECASE | re.DOTALL
)
if abstract_match:
abstract = abstract_match.group(2).strip()
remaining = text[:abstract_match.start()] + text[abstract_match.end():]
return abstract, remaining
return "", text
@staticmethod
def extract_sections(text: str) -> List[Tuple[str, str]]:
"""
Extract document sections.
Args:
text: Document text
Returns:
List of (section_title, section_content) tuples
"""
# Match common section patterns
section_pattern = r'(?:^|\n)((?:\d+\.\s+)?(?:introduction|methodology|results|discussion|conclusion|references|abstract).*?)(?:\n(?:\d+\.\s+)?(?:[A-Z][^.]*?)(?=\n|$))'
sections = []
matches = re.finditer(section_pattern, text, re.IGNORECASE)
for match in matches:
title = match.group(1).strip()
content = match.group(2).strip() if match.lastindex >= 2 else ""
sections.append((title, content))
return sections
|