File size: 11,471 Bytes
cba1186 |
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 |
"""
Professional Product Search Engine for Trek Chatbot
Implements intelligent product matching with fuzzy search and NLP techniques
"""
import re
from difflib import SequenceMatcher
from typing import List, Tuple, Dict, Optional
import unicodedata
class ProductSearchEngine:
"""Advanced product search with intelligent matching"""
def __init__(self, products: List[Tuple]):
"""
Initialize with products list
products: List of tuples (short_name, product_info, full_name)
"""
self.products = products
self.product_index = self._build_index()
def _build_index(self) -> Dict:
"""Build search index for faster lookups"""
index = {
'by_name': {},
'by_words': {},
'by_category': {},
'by_model': {},
'normalized': {}
}
for product in self.products:
short_name = product[0]
full_name = product[2]
# Normalize and store
normalized_full = self._normalize_text(full_name)
normalized_short = self._normalize_text(short_name)
# Store by full name
index['by_name'][normalized_full] = product
index['normalized'][normalized_full] = full_name
# Extract and index words
words = normalized_full.split()
for word in words:
if len(word) > 2: # Skip very short words
if word not in index['by_words']:
index['by_words'][word] = []
index['by_words'][word].append(product)
# Extract model numbers and categories
model_match = re.search(r'\b(\d+\.?\d*)\b', full_name)
if model_match:
model_num = model_match.group(1)
if model_num not in index['by_model']:
index['by_model'][model_num] = []
index['by_model'][model_num].append(product)
# Category extraction (first word often represents category)
if words:
category = words[0]
if category not in index['by_category']:
index['by_category'][category] = []
index['by_category'][category].append(product)
return index
def _normalize_text(self, text: str) -> str:
"""Normalize text for better matching"""
if not text:
return ""
# Convert to lowercase
text = text.lower()
# Remove Turkish characters
replacements = {
'ı': 'i', 'İ': 'i', 'ş': 's', 'Ş': 's',
'ğ': 'g', 'Ğ': 'g', 'ü': 'u', 'Ü': 'u',
'ö': 'o', 'Ö': 'o', 'ç': 'c', 'Ç': 'c'
}
for tr_char, eng_char in replacements.items():
text = text.replace(tr_char, eng_char)
# Remove special characters but keep spaces and numbers
text = re.sub(r'[^\w\s\d\.]', ' ', text)
# Normalize whitespace
text = ' '.join(text.split())
return text
def _calculate_similarity(self, str1: str, str2: str) -> float:
"""Calculate similarity between two strings"""
return SequenceMatcher(None, str1, str2).ratio()
def search(self, query: str, threshold: float = 0.6) -> List[Tuple[float, Tuple]]:
"""
Search for products matching the query
Returns list of (score, product) tuples sorted by relevance
"""
query_normalized = self._normalize_text(query)
query_words = query_normalized.split()
results = {}
# 1. Exact match
if query_normalized in self.product_index['by_name']:
product = self.product_index['by_name'][query_normalized]
results[id(product)] = (1.0, product)
# 2. Model number search
model_match = re.search(r'\b(\d+\.?\d*)\b', query)
if model_match:
model_num = model_match.group(1)
if model_num in self.product_index['by_model']:
for product in self.product_index['by_model'][model_num]:
if id(product) not in results:
# Check if model number is in correct context
score = 0.9 if model_num in product[2].lower() else 0.7
results[id(product)] = (score, product)
# 3. Word-based search with scoring
word_matches = {}
for word in query_words:
if len(word) > 2 and word in self.product_index['by_words']:
for product in self.product_index['by_words'][word]:
if id(product) not in word_matches:
word_matches[id(product)] = {'count': 0, 'product': product}
word_matches[id(product)]['count'] += 1
# Calculate word match scores
for product_id, match_info in word_matches.items():
product = match_info['product']
matched_count = match_info['count']
total_query_words = len([w for w in query_words if len(w) > 2])
if total_query_words > 0:
word_score = matched_count / total_query_words
# Boost score if all important words match
if matched_count == total_query_words:
word_score = min(word_score * 1.2, 0.95)
# Check word order for better scoring
product_text = self._normalize_text(product[2])
if query_normalized in product_text:
word_score = min(word_score * 1.3, 0.98)
if id(product) not in results or results[id(product)][0] < word_score:
results[id(product)] = (word_score, product)
# 4. Fuzzy matching for all products
for product in self.products:
product_normalized = self._normalize_text(product[2])
similarity = self._calculate_similarity(query_normalized, product_normalized)
# Substring matching
if query_normalized in product_normalized:
similarity = max(similarity, 0.8)
# Check if product contains all query words (in any order)
if all(word in product_normalized for word in query_words if len(word) > 2):
similarity = max(similarity, 0.75)
if similarity >= threshold:
if id(product) not in results or results[id(product)][0] < similarity:
results[id(product)] = (similarity, product)
# 5. Category-based fallback
if not results and query_words:
category = query_words[0]
if category in self.product_index['by_category']:
for product in self.product_index['by_category'][category]:
results[id(product)] = (0.5, product)
# Convert to list and sort by score
result_list = list(results.values())
result_list.sort(key=lambda x: x[0], reverse=True)
return result_list
def find_best_match(self, query: str) -> Optional[Tuple]:
"""Find the single best matching product"""
results = self.search(query)
if results and results[0][0] >= 0.6:
return results[0][1]
return None
def find_similar_products(self, product_name: str, limit: int = 5) -> List[Tuple]:
"""Find products similar to the given product name"""
results = self.search(product_name)
similar = []
# Skip the first result if it's an exact match
start_idx = 1 if results and results[0][0] > 0.95 else 0
for score, product in results[start_idx:start_idx + limit]:
if score >= 0.5:
similar.append(product)
return similar
def extract_product_context(self, query: str) -> Dict:
"""Extract context from query (size, color, type, etc.)"""
context = {
'sizes': [],
'colors': [],
'types': [],
'features': [],
'price_range': None
}
# Size detection
size_patterns = [
r'\b(xs|s|m|l|xl|xxl|2xl|3xl)\b',
r'\b(\d{2})\b(?=\s*beden|\s*numara|$)', # 44, 46, etc.
r'\b(small|medium|large)\b'
]
for pattern in size_patterns:
matches = re.findall(pattern, query.lower())
context['sizes'].extend(matches)
# Color detection
colors = ['siyah', 'beyaz', 'mavi', 'kirmizi', 'yesil', 'gri', 'turuncu',
'black', 'white', 'blue', 'red', 'green', 'grey', 'gray', 'orange']
for color in colors:
if color in query.lower():
context['colors'].append(color)
# Type detection
types = ['erkek', 'kadin', 'cocuk', 'yol', 'dag', 'sehir', 'elektrikli',
'karbon', 'aluminyum', 'gravel', 'hybrid']
for type_word in types:
if type_word in query.lower():
context['types'].append(type_word)
# Feature detection
features = ['disk fren', 'shimano', 'sram', 'karbon', 'aluminyum',
'hidrolik', 'mekanik', '29 jant', '27.5 jant']
for feature in features:
if feature in query.lower():
context['features'].append(feature)
# Price range detection
price_match = re.search(r'(\d+)\.?(\d*)\s*(bin|tl)', query.lower())
if price_match:
price = float(price_match.group(1) + ('.' + price_match.group(2) if price_match.group(2) else ''))
if 'bin' in price_match.group(3):
price *= 1000
context['price_range'] = price
return context
def generate_suggestions(self, failed_query: str) -> List[str]:
"""Generate suggestions for failed searches"""
suggestions = []
query_normalized = self._normalize_text(failed_query)
query_words = query_normalized.split()
# Find products with partial matches
partial_matches = set()
for word in query_words:
if len(word) > 3:
for product_word in self.product_index['by_words']:
if word in product_word or product_word in word:
partial_matches.add(product_word)
# Generate suggestions from partial matches
for match in list(partial_matches)[:5]:
if match in self.product_index['by_words']:
products = self.product_index['by_words'][match]
if products:
suggestions.append(products[0][2])
# Add category suggestions
for category in list(self.product_index['by_category'].keys())[:3]:
if any(word in category for word in query_words):
category_products = self.product_index['by_category'][category]
if category_products:
suggestions.append(category_products[0][2])
return list(set(suggestions))[:5] # Return unique suggestions |