Sentiment-Analysis/result_twitter~ at master · xiaowh7/Sentiment-Analysis · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
TASK B
Comparing predictions ../code/ans.txt to gold standard ../dataset/test-gold-B.csv.
File: ../code/ans.txt -- no errors found
Prediction file: 3267 lines
                 3267 unique
Breakdown by class:
positive 	986
neutral 	1915
negative 	366
File: ../dataset/test-gold-B.csv -- no errors found


Gold standard: 3267 lines
               3267 unique
Breakdown by class:
positive 	1331
neutral 	1442
negative 	494


Confusion table:
gs \ pred| positive| negative|  neutral
---------------------------------------
 positive|      818|       54|      459
 negative|       27|      257|      210
  neutral|      141|       55|     1246


Scores:
class	                       prec	              recall     fscore
positive 	   (818/986) 0.8296	   (818/1331) 0.6146	 0.7061
negative 	   (257/366) 0.7022	    (257/494) 0.5202	 0.5977
neutral 	 (1246/1915) 0.6507	  (1246/1442) 0.8641	 0.7423
-----------------------------------------------------------------------
average(pos and neg)                                             0.6519