added regex pattern matching for suffixes

Author: basak-tepe

src/widid.ipynb

@@ -171,20 +171,24 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "342d2e07",
    "metadata": {
     "lines_to_next_cell": 1
    },
    "outputs": [],
    "source": [
     "def extract_contexts(texts, target_word, window=10):\n",
-    "    \"\"\"Extract short context windows around target word.\"\"\"\n",
+    "    \"\"\"Extract short context windows around target word and its morphological variations.\"\"\"\n",
     "    contexts = []\n",
+    "    # Create regex pattern to match target word and any Turkish suffixes\n",
+    "    pattern = re.compile(rf\"\\b{re.escape(target_word.lower())}\\w*\\b\")\n",
+    "    \n",
     "    for t in texts:\n",
     "        tokens = re.findall(r\"\\w+\", t.lower())\n",
     "        for i, tok in enumerate(tokens):\n",
-    "            if tok == target_word.lower():\n",
+    "            # Use regex to match the word and its variations\n",
+    "            if pattern.match(tok):\n",
     "                start = max(0, i - window)\n",
     "                end = min(len(tokens), i + window + 1)\n",
     "                snippet = \" \".join(tokens[start:end])\n",