Hugging Face's logo

5CD-AI
/
Vietnamese-Sentiment-visobert

Text Classification
Transformers
Safetensors
Vietnamese
xlm-roberta
sentiment-analysis
social-listening
Model card Files Community
1
baohuynhbk14 commited on
Commit
a26dbaa
·
verified ·
1 Parent(s): 2432df7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +134 -113
README.md CHANGED
@@ -46,148 +46,149 @@ The model will give softmax outputs for three labels.
46
  <th>Neu</th>
47
  </tr>
48
  <tr align="center">
49
- <td align="left">All</td>
50
- <td>1000</td>
51
- <td>2000</td>
52
- <td>3000</td>
53
- <td>1000</td>
54
- <td>2000</td>
55
- <td>3000</td>
56
- <td>1000</td>
57
- <td>2000</td>
58
- <td>3000</td>
59
  </tr>
60
  <tr align="center">
61
  <td align="left">SA-VLSP2016</td>
62
- <td>1000</td>
63
- <td>2000</td>
64
- <td>3000</td>
65
- <td>1000</td>
66
- <td>2000</td>
67
- <td>3000</td>
68
- <td>1000</td>
69
- <td>2000</td>
70
- <td>3000</td>
71
  </tr>
72
  <tr align="center">
73
  <td align="left">UIT-VSFC </td>
74
- <td>1000</td>
75
- <td>2000</td>
76
- <td>3000</td>
77
- <td>1000</td>
78
- <td>2000</td>
79
- <td>3000</td>
80
- <td>1000</td>
81
- <td>2000</td>
82
- <td>3000</td>
83
  </tr>
84
  <tr align="center">
85
  <td align="left">UIT-VSMEC</td>
86
- <td>1000</td>
87
- <td>2000</td>
88
- <td>3000</td>
89
- <td>1000</td>
90
- <td>2000</td>
91
- <td>3000</td>
92
- <td>1000</td>
93
- <td>2000</td>
94
- <td>3000</td>
 
95
  </tr>
96
  <tr align="center">
97
  <td align="left">AIVIVN-2019</td>
98
- <td>1000</td>
99
- <td>2000</td>
100
- <td>3000</td>
101
- <td>1000</td>
102
- <td>2000</td>
103
- <td>3000</td>
104
- <td>1000</td>
105
- <td>2000</td>
106
- <td>3000</td>
107
  </tr>
108
  <tr align="center">
109
  <td align="left">UIT-ViCTSD</td>
110
- <td>1000</td>
111
- <td>2000</td>
112
- <td>3000</td>
113
- <td>1000</td>
114
- <td>2000</td>
115
- <td>3000</td>
116
- <td>1000</td>
117
- <td>2000</td>
118
- <td>3000</td>
119
  </tr>
120
  <tr align="center">
121
  <td align="left">UIT-ViHSD</td>
122
- <td>1000</td>
123
- <td>2000</td>
124
- <td>3000</td>
125
- <td>1000</td>
126
- <td>2000</td>
127
- <td>3000</td>
128
- <td>1000</td>
129
- <td>2000</td>
130
- <td>3000</td>
131
  </tr>
132
  <tr align="center">
133
  <td align="left">UIT-ViSFD</td>
134
- <td>1000</td>
135
- <td>2000</td>
136
- <td>3000</td>
137
- <td>1000</td>
138
- <td>2000</td>
139
- <td>3000</td>
140
- <td>1000</td>
141
- <td>2000</td>
142
- <td>3000</td>
143
  </tr>
144
  <tr align="center">
145
  <td align="left">UIT-ViOCD</td>
146
- <td>1000</td>
147
- <td>2000</td>
148
- <td>3000</td>
149
- <td>1000</td>
150
- <td>2000</td>
151
- <td>3000</td>
152
- <td>1000</td>
153
- <td>2000</td>
154
- <td>3000</td>
155
  </tr>
156
  <tr align="center">
157
  <td align="left">Ecommerce-reviews</td>
158
- <td>1000</td>
159
- <td>2000</td>
160
- <td>3000</td>
161
- <td>1000</td>
162
- <td>2000</td>
163
- <td>3000</td>
164
- <td>1000</td>
165
- <td>2000</td>
166
- <td>3000</td>
167
  </tr>
168
  <tr align="center">
169
  <td align="left">VOZ-HSD</td>
170
- <td>1000</td>
171
- <td>2000</td>
172
- <td>3000</td>
173
- <td>1000</td>
174
- <td>2000</td>
175
- <td>3000</td>
176
- <td>1000</td>
177
- <td>2000</td>
178
- <td>3000</td>
179
  </tr>
180
  <tr align="center">
181
  <td align="left">Vietnamese-amazon-polarity</td>
182
- <td>1000</td>
183
- <td>2000</td>
184
- <td>3000</td>
185
- <td>1000</td>
186
- <td>2000</td>
187
- <td>3000</td>
188
- <td>1000</td>
189
- <td>2000</td>
190
- <td>3000</td>
191
  </tr>
192
  </table>
193
 
@@ -196,10 +197,15 @@ The model will give softmax outputs for three labels.
196
  <tr align="center">
197
  <td rowspan=2><b>Model</td>
198
  <td rowspan=2><b>Avg MF1</td>
199
- <td colspan=3><b>Emotion Recognition</td>
200
- <td colspan=3><b>Hate Speech Detection</td>
201
- <td colspan=3><b>Spam Reviews Detection</td>
202
- <td colspan=3><b>Hate Speech Spans Detection</td>
 
 
 
 
 
203
  </tr>
204
  <tr align="center">
205
  <td><b>Acc</td>
@@ -214,6 +220,21 @@ The model will give softmax outputs for three labels.
214
  <td><b>Acc</td>
215
  <td><b>WF1</td>
216
  <td><b>MF1</td>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
217
  </tr>
218
  <tr align="center">
219
  <td align="left">viBERT</td>
 
46
  <th>Neu</th>
47
  </tr>
48
  <tr align="center">
49
+ <td align="left">All-filtered</td>
50
+ <td>62708</td>
51
+ <td>41400</td>
52
+ <td>11593</td>
53
+ <td>-</td>
54
+ <td>-</td>
55
+ <td>-</td>
56
+ <td>5079</td>
57
+ <td>3724</td>
58
+ <td>638</td>
59
  </tr>
60
  <tr align="center">
61
  <td align="left">SA-VLSP2016</td>
62
+ <td>4759</td>
63
+ <td>4798</td>
64
+ <td>4459</td>
65
+ <td>1180</td>
66
+ <td>1190</td>
67
+ <td>1114</td>
68
+ <td>-</td>
69
+ <td>-</td>
70
+ <td>-</td>
71
  </tr>
72
  <tr align="center">
73
  <td align="left">UIT-VSFC </td>
74
+ <td>5325</td>
75
+ <td>5643</td>
76
+ <td>458</td>
77
+ <td>1409</td>
78
+ <td>1590</td>
79
+ <td>167</td>
80
+ <td>-</td>
81
+ <td>-</td>
82
+ <td>-</td>
83
  </tr>
84
  <tr align="center">
85
  <td align="left">UIT-VSMEC</td>
86
+ <td>3219</td>
87
+ <td>1665</td>
88
+ <td>594</td>
89
+ <td>458</td>
90
+ <td>407</td>
91
+ <td>210</td>
92
+ <td>71</td>
93
+ <td>388</td>
94
+ <td>239</td>
95
+ <td>52</td>
96
  </tr>
97
  <tr align="center">
98
  <td align="left">AIVIVN-2019</td>
99
+ <td>6776</td>
100
+ <td>7879</td>
101
+ <td>-</td>
102
+ <td>4770</td>
103
+ <td>5168</td>
104
+ <td>-</td>
105
+ <td>-</td>
106
+ <td>-</td>
107
+ <td>-</td>
108
  </tr>
109
  <tr align="center">
110
  <td align="left">UIT-ViCTSD</td>
111
+ <td>3370</td>
112
+ <td>2615</td>
113
+ <td>933</td>
114
+ <td>3370</td>
115
+ <td>2615</td>
116
+ <td>933</td>
117
+ <td>3370</td>
118
+ <td>2615</td>
119
+ <td>933</td>
120
  </tr>
121
  <tr align="center">
122
  <td align="left">UIT-ViHSD</td>
123
+ <td>4162</td>
124
+ <td>19886</td>
125
+ <td>-</td>
126
+ <td>1132</td>
127
+ <td>5548</td>
128
+ <td>-</td>
129
+ <td>482</td>
130
+ <td>2190</td>
131
+ <td>-</td>
132
  </tr>
133
  <tr align="center">
134
  <td align="left">UIT-ViSFD</td>
135
+ <td>2850</td>
136
+ <td>3670</td>
137
+ <td>1266</td>
138
+ <td>827</td>
139
+ <td>1000</td>
140
+ <td>397</td>
141
+ <td>409</td>
142
+ <td>515</td>
143
+ <td>188</td>
144
  </tr>
145
  <tr align="center">
146
  <td align="left">UIT-ViOCD</td>
147
+ <td>2292</td>
148
+ <td>2095</td>
149
+ <td>-</td>
150
+ <td>279</td>
151
+ <td>270</td>
152
+ <td>-</td>
153
+ <td>283</td>
154
+ <td>265</td>
155
+ <td>-</td>
156
  </tr>
157
  <tr align="center">
158
  <td align="left">Ecommerce-reviews</td>
159
+ <td>20093</td>
160
+ <td>6669</td>
161
+ <td>4698</td>
162
+ <td>-</td>
163
+ <td>-</td>
164
+ <td>-</td>
165
+ <td>-</td>
166
+ <td>-</td>
167
+ <td>-</td>
168
  </tr>
169
  <tr align="center">
170
  <td align="left">VOZ-HSD</td>
171
+ <td>2676</td>
172
+ <td>1213</td>
173
+ <td>1071</td>
174
+ <td>1068</td>
175
+ <td>495</td>
176
+ <td>420</td>
177
+ <td>511</td>
178
+ <td>224</td>
179
+ <td>199</td>
180
  </tr>
181
  <tr align="center">
182
  <td align="left">Vietnamese-amazon-polarity</td>
183
+ <td>2559</td>
184
+ <td>2441</td>
185
+ <td>-</td>
186
+ <td>1017</td>
187
+ <td>983</td>
188
+ <td>-</td>
189
+ <td>523</td>
190
+ <td>477</td>
191
+ <td>-</td>
192
  </tr>
193
  </table>
194
 
 
197
  <tr align="center">
198
  <td rowspan=2><b>Model</td>
199
  <td rowspan=2><b>Avg MF1</td>
200
+ <td colspan=3><b>SA-VLSP2016</td>
201
+ <td colspan=3><b>AIVIVN-2019</td>
202
+ <td colspan=3><b>UIT-VSFC</td>
203
+ <td colspan=3><b>UIT-VSMEC</td>
204
+ <td colspan=3><b>UIT-ViCTSD</td>
205
+ <td colspan=3><b>UIT-ViOCD</td>
206
+ <td colspan=3><b>UIT-ViSFD</td>
207
+ <td colspan=3><b>VOZ-HSD</td>
208
+ <td colspan=3><b>Vi-amazon-polar</td>
209
  </tr>
210
  <tr align="center">
211
  <td><b>Acc</td>
 
220
  <td><b>Acc</td>
221
  <td><b>WF1</td>
222
  <td><b>MF1</td>
223
+ <td><b>Acc</td>
224
+ <td><b>WF1</td>
225
+ <td><b>MF1</td>
226
+ <td><b>Acc</td>
227
+ <td><b>WF1</td>
228
+ <td><b>MF1</td>
229
+ <td><b>Acc</td>
230
+ <td><b>WF1</td>
231
+ <td><b>MF1</td>
232
+ <td><b>Acc</td>
233
+ <td><b>WF1</td>
234
+ <td><b>MF1</td>
235
+ <td><b>Acc</td>
236
+ <td><b>WF1</td>
237
+ <td><b>MF1</td>
238
  </tr>
239
  <tr align="center">
240
  <td align="left">viBERT</td>