Change model from 'frequency matrix model' into 'Direct model'

news_classify_tag.py

@@ -27,11 +27,11 @@ class Classifier:
         self.db = _mysql.connect (host=host, port=port, user=user, passwd=password, db=db)
 
         # geting tags
-        self.db.query("SELECT id, name FROM wp_esi_tag")
-        rez = self.db.store_result()
+        self.db.query ("SELECT id, name FROM wp_esi_tag")
+        rez = self.db.store_result ()
         # result =
         tags = list ()
-        for id, description in rez.fetch_row(maxrows=0):
+        for id, description in rez.fetch_row (maxrows=0):
             tags.append ((id, description))
         self.tags = tags
         # print (len(tags), tags)
@@ -43,9 +43,9 @@ class Classifier:
                        FROM wp_esi_news, wp_esi_tag_news WHERE wp_esi_tag_news.news_id=wp_esi_news.id
                           ORDER BY wp_esi_tag_news.tag_id LIMIT 1000 '''
         self.db.query (sql1)
-        result = self.db.store_result()
+        result = self.db.store_result ()
         data = list ()
-        for tag_id, title, description in result.fetch_row(maxrows=0):
+        for tag_id, title, description in result.fetch_row (maxrows=0):
             data.append ((tag_id, title, description))
 
         for tag_id, _ in self.tags:
@@ -93,7 +93,7 @@ class Classifier:
         for i in range (columns):
             for j in range (rows):
                 if trained_matrix[i, j] != 0:
-                    frequency_matrix[i, j] = 1 / trained_matrix[i, j]
+                    frequency_matrix[i, j] = trained_matrix[i, j]
 
         # normalise rowsfrequency matrix
         for j in range (columns):
@@ -128,7 +128,7 @@ class Classifier:
             temp_matrix = self.tag_frequency_matrix[row]
             rez_summ = temp_matrix[self.matrix_test_data.toarray ()[0] > 0].sum ()
             if rez_summ > (persantage / 100):
-                vector_accordance.append ((row, int (rez_summ * 1000) / 10, self.tags[row][1]))
+                vector_accordance.append ((int(self.tags[row][0]), int (rez_summ * 1000) / 10, self.tags[row][1]))
             vector_accordance.sort (key=lambda tup: tup[1], reverse=True)
 
         return vector_accordance
@@ -223,35 +223,36 @@ if __name__ == "__main__":
             FROM (
                     SELECT wp_esi_tag_news.news_id, wp_esi_news.title, wp_esi_news.description
                        FROM wp_esi_news, wp_esi_tag_news WHERE wp_esi_tag_news.news_id=wp_esi_news.id
-                          ORDER BY wp_esi_tag_news.tag_id ) as rez 
+                          ORDER BY wp_esi_tag_news.tag_id ) AS rez 
 			                  ORDER BY rand() LIMIT 15'''
-    tags_classif.db.query(sql1)
-    news_results = tags_classif.db.store_result()
-    news_results = news_results.fetch_row(maxrows=0)
+
+    tags_classif.db.query (sql1)
+    news_results = tags_classif.db.store_result ()
+    news_results = news_results.fetch_row (maxrows=0)
     data = list ()
     for i, (id_news, title, description) in enumerate (news_results):
-        text_for_analis = title.decode('ascii', 'ignore') + '\n' + description.decode('ascii', 'ignore')
+        text_for_analis = title.decode ('ascii', 'ignore') + '\n' + description.decode ('ascii', 'ignore')
         # print ("\n#", str (i))
-        print ("\nNews title: " + title.decode('ascii', 'ignore'))
+        print ("\nNews title: " + title.decode ('ascii', 'ignore'))
         tags_classif.classify (text_for_analis)
         print ("Model calculated Accordance (#tag, %-accordance, tag_description): ", end=" ")
         print (tags_classif.teg_accordance)
-        sql = " select tag_id from wp_esi_tag_news where news_id =" + str(id_news)
+        sql = " SELECT tag_id FROM wp_esi_tag_news WHERE news_id =" + str (id_news)
         # print(sql)
-        results = tags_classif.db.query(sql)
-        tags = tags_classif.db.store_result()
-        tags = tags.fetch_row(maxrows=0)
+        results = tags_classif.db.query (sql)
+        tags = tags_classif.db.store_result ()
+        tags = tags.fetch_row (maxrows=0)
         # print(tags)
         # print(tags_classif.tags)
         print ("User classified tags for present news:")
-        for (tag, ) in tags:
-            tag = int(tag)
+        for (tag,) in tags:
+            tag = int (tag)
             # print(tag)
-            print (str(tag - 1) + " " + tags_classif.tags[tag - 1][1].decode('ascii', 'ignore'))
+            print (str (tag) + " " + tags_classif.tags[tag-1][1].decode ('ascii', 'ignore'))
 
     exit (0)
-    tags_classif.db.query("SELECT title, description FROM wp_esi_news ORDER BY RAND() LIMIT 25")
-    result = tags_classif.db.store_result()
+    tags_classif.db.query ("SELECT title, description FROM wp_esi_news ORDER BY RAND() LIMIT 25")
+    result = tags_classif.db.store_result ()
 
     for i, (title, description) in enumerate (result.fetch_row (maxrows=0)):
         text_for_analis = title + '\n' + description