Merge pull request #115 from meedan/CV2-5247-support-yake-keyword-ext…

…raction-for-chinese CV2-5247-support-yake-keyword-extraction-for-chinese
meedan · Oct 22, 2024 · 5d6d98b · 5d6d98b
2 parents be8027a + 6bb58c2
commit 5d6d98b
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/lib/model/yake_keywords.py b/lib/model/yake_keywords.py
@@ -38,7 +38,7 @@ def normalize_special_characters(self, text):
         return text
 
     def run_chinese_segmentation_with_jieba(self, text):
-        return " ".join(list(jieba.cut_for_search(text)))
+        return " ".join(list(jieba.cut(text)))
 
     def run_yake(self, text: str,
                  language: str,

diff --git a/test/lib/model/test_yake_keywords.py b/test/lib/model/test_yake_keywords.py
@@ -55,7 +55,7 @@ def test_normalize_special_characters(self):
 
     def test_run_chinese_segmentation_with_jieba(self):
         test_text = '''哈里斯同意与特朗普再进行一次美大选辩论'''
-        expected = "哈里 里斯 哈里斯 同意 与 特朗普 再 进行 一次 美 大选 辩论"
+        expected = "哈里斯 同意 与 特朗普 再 进行 一次 美 大选 辩论"
         self.assertEqual(self.yake_model.run_chinese_segmentation_with_jieba(test_text), expected)
 
     def test_run_yake_real_with_chinese(self):