From be73b6600e9980bb0902f584efc114148a0adec2 Mon Sep 17 00:00:00 2001
From: "Mark Edward M. Gonzales" <gonzales.markedward@gmail.com>
Date: Tue, 5 Sep 2023 20:10:42 +0800
Subject: [PATCH] Make similarity cutoff stricter

---
 callbacks/text_mining/util.py | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/callbacks/text_mining/util.py b/callbacks/text_mining/util.py
index 30fff44c..86c6c0cd 100644
--- a/callbacks/text_mining/util.py
+++ b/callbacks/text_mining/util.py
@@ -10,8 +10,7 @@
 
 
 COLNAMES = ['Gene', 'PMID', 'Title', 'Sentence', 'Score']
-SIMILARITY_CUTOFF = 75
-MAX_NUM_RESULTS = 100
+SIMILARITY_CUTOFF = 85
 
 
 def sanitize_text(text):
@@ -177,8 +176,6 @@ def text_mining_query_search(query_string):
                         df.loc[len(df.index)] = [Entity, PMID,
                                                  Title, Sentence, similarity.score]
 
-                        if df.shape[0] == MAX_NUM_RESULTS:
-                            break
             except:
                 pass