diff --git a/logparser/utils/evaluator.py b/logparser/utils/evaluator.py
index 9e56060e..1595d919 100644
--- a/logparser/utils/evaluator.py
+++ b/logparser/utils/evaluator.py
@@ -28,9 +28,9 @@ def evaluate(groundtruth, parsedresult):
     df_groundtruth = pd.read_csv(groundtruth)
     df_parsedlog = pd.read_csv(parsedresult)
     # Remove invalid groundtruth event Ids
-    null_logids = df_groundtruth[~df_groundtruth['EventId'].isnull()].index
-    df_groundtruth = df_groundtruth.loc[null_logids]
-    df_parsedlog = df_parsedlog.loc[null_logids]
+    non_empty_log_ids = df_groundtruth[~df_groundtruth['EventId'].isnull()].index
+    df_groundtruth = df_groundtruth.loc[non_empty_log_ids]
+    df_parsedlog = df_parsedlog.loc[non_empty_log_ids]
     (precision, recall, f_measure, accuracy) = get_accuracy(df_groundtruth['EventId'], df_parsedlog['EventId'])
     print('Precision: %.4f, Recall: %.4f, F1_measure: %.4f, Parsing_Accuracy: %.4f'%(precision, recall, f_measure, accuracy))
     return f_measure, accuracy