Automatic Detection and Characterization of Propaganda Techniques from Diplomats (DIPROMATS)¶
The DIPROMATS task (webpage)presented at IberLEF 2023 focused on indentifying tweets that have a propaganda technique.
The following code can generate an instance of the system used in the competition.
>>> from EvoMSA.competitions import Comp2023
>>> D = # Training set
>>> comp2023 = Comp2023(lang='es')
>>> ins = comp2023.stack_3_bows(D)
Configuration |
Performance |
p-value |
---|---|---|
|
0.6551 |
1.0000 |
|
0.6544 |
0.4180 |
|
0.6515 |
0.2200 |
|
0.6514 |
0.2500 |
|
0.6488 |
0.1120 |
|
0.6486 |
0.1360 |
|
0.6486 |
0.1360 |
|
0.6486 |
0.1520 |
|
0.6485 |
0.1480 |
|
0.6484 |
0.1180 |
|
0.6290 |
0.0080 |
|
0.6136 |
0.0000 |
|
0.6123 |
0.0000 |
The following code can generate an instance of the system used in the competition.
>>> from EvoMSA.competitions import Comp2023
>>> D = # Training set
>>> tailored = 'IberLEF2023_DIPROMATS_task1'
>>> comp2023 = Comp2023(lang='en', tailored=tailored)
>>> ins = comp2023.stack_3_bow_tailored_all_keywords(D)
Configuration |
Performance |
p-value |
---|---|---|
|
0.6498 |
1.0000 |
|
0.6489 |
0.2260 |
|
0.6471 |
0.1280 |
|
0.6448 |
0.0440 |
|
0.6446 |
0.0140 |
|
0.6443 |
0.0240 |
|
0.6386 |
0.0080 |
|
0.6381 |
0.0000 |
|
0.6377 |
0.0040 |
|
0.6327 |
0.0000 |
|
0.6043 |
0.0000 |
|
0.5961 |
0.0000 |
|
0.5922 |
0.0000 |