diff --git a/autorag/utils/preprocess.py b/autorag/utils/preprocess.py
index d4fa3b838..1dd7f2f26 100644
--- a/autorag/utils/preprocess.py
+++ b/autorag/utils/preprocess.py
@@ -3,6 +3,8 @@
 import numpy as np
 import pandas as pd
 
+from autorag.utils.util import normalize_unicode
+
 
 def validate_qa_dataset(df: pd.DataFrame):
     columns = ['qid', 'query', 'retrieval_gt', 'generation_gt']
@@ -49,6 +51,8 @@ def cast_generation_gt(gt):
         "query must be string type."
     df['retrieval_gt'] = df['retrieval_gt'].apply(cast_retrieval_gt)
     df['generation_gt'] = df['generation_gt'].apply(cast_generation_gt)
+    df['query'] = df['query'].apply(normalize_unicode)
+    df['generation_gt'] = df['generation_gt'].apply(lambda x: list(map(normalize_unicode, x)))
     return df
 
 
@@ -80,6 +84,19 @@ def make_prev_next_id_metadata(x, id_type: str):
     df['metadata'] = df['metadata'].apply(lambda x: make_prev_next_id_metadata(x, 'prev_id'))
     df['metadata'] = df['metadata'].apply(lambda x: make_prev_next_id_metadata(x, 'next_id'))
 
+    df['contents'] = df['contents'].apply(normalize_unicode)
+
+    def normalize_unicode_metadata(metadata: dict):
+        result = {}
+        for key, value in metadata.items():
+            if isinstance(value, str):
+                result[key] = normalize_unicode(value)
+            else:
+                result[key] = value
+        return result
+
+    df['metadata'] = df['metadata'].apply(normalize_unicode_metadata)
+
     # check every metadata have a prev_id, next_id key
     assert all('prev_id' in metadata for metadata in df['metadata']), "Every metadata must have a prev_id key."
     assert all('next_id' in metadata for metadata in df['metadata']), "Every metadata must have a next_id key."
diff --git a/autorag/utils/util.py b/autorag/utils/util.py
index 6f067d804..0a4077fbf 100644
--- a/autorag/utils/util.py
+++ b/autorag/utils/util.py
@@ -7,6 +7,7 @@
 import os
 import re
 import string
+import unicodedata
 from copy import deepcopy
 from typing import List, Callable, Dict, Optional, Any, Collection
 
@@ -360,3 +361,7 @@ def filter_dict_keys(dict_, keys: List[str]):
 def split_dataframe(df, chunk_size):
     num_chunks = len(df) // chunk_size + 1 if len(df) % chunk_size != 0 else len(df) // chunk_size
     return list(map(lambda x: df[x * chunk_size:(x + 1) * chunk_size], range(num_chunks)))
+
+
+def normalize_unicode(text: str) -> str:
+    return unicodedata.normalize('NFC', text)
diff --git a/tests/autorag/utils/test_util.py b/tests/autorag/utils/test_util.py
index 7429a5429..f977145e9 100644
--- a/tests/autorag/utils/test_util.py
+++ b/tests/autorag/utils/test_util.py
@@ -13,7 +13,7 @@
 from autorag.utils import fetch_contents
 from autorag.utils.util import load_summary_file, result_to_dataframe, \
     make_combinations, explode, replace_value_in_dict, normalize_string, convert_string_to_tuple_in_dict, process_batch, \
-    convert_env_in_dict, openai_truncate_by_token, convert_datetime_string, split_dataframe
+    convert_env_in_dict, openai_truncate_by_token, convert_datetime_string, split_dataframe, normalize_unicode
 from tests.mock import MockLLM
 
 root_dir = pathlib.PurePath(os.path.dirname(os.path.realpath(__file__))).parent.parent
@@ -331,3 +331,18 @@ def test_split_dataframe():
     assert len(df_list_2[0]) == 3
     assert len(df_list_2[-1]) == 1
     assert pd.DataFrame({'a': list(range(3)), 'b': list(range(10, 13))}).equals(df_list_2[0])
+
+
+def test_normalize_unicode():
+    str1 = "전국보행자전용도로표준데이터"
+    str2 = "전국보행자전용도로표준데이터"
+    assert len(str1) == 14
+    assert len(str2) == 34
+    assert str1 != str2
+
+    new_str1 = normalize_unicode(str1)
+    new_str2 = normalize_unicode(str2)
+
+    assert len(new_str1) == 14
+    assert len(new_str2) == 14
+    assert new_str1 == new_str2