Automatically labeled data generation for classification of reputation defence strategies thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Automatically labeled data generation for classification of reputation defence strategies

Published on May 30, 2018584 Views

Reputation defence is a form of persuasive tactic that is used in various social settings especially in political situations. Detection of reputation defence strategy is a novel task that could help i

Related categories

Chapter list

ParlaCLARIN Workshop: Creating and Using Parliamentary Corpora, Miyazaki 201800:00
Automatically Labeled Data Generation for Classification of Reputation Defence Strategies00:11
Graeme Hirst00:12
Reputation00:14
Example00:53
Reputation threat and defence in political speech02:05
Why detecting reputation defence strategies?04:07
Benoit’s reputation defence strategies04:39
Denial approaches05:17
Denial Example - 105:25
Denial Example - 205:25
Excuse approaches06:48
Excuse example07:04
Justification approaches08:24
Justification example08:59
Concession approaches10:00
Concession example10:12
Data11:07
Canadian Parliamentary exchanges11:13
Methods12:49
Method12:50
Observed word pairs13:14
Automatic pattern extraction13:54
Example patterns14:34
Evaluation15:01
Evaluation of the automatically generated labels15:09
Comparison with human annotations15:19
Human annotations15:58
Example disagreement between annotators16:46
Results – Agreement with human annotations17:53
Contribution to classification19:37
Features from answers20:19
Relations between questions and answers20:44
Framework21:00
Results (word pairs)21:09
Results (patterns)22:46
Conclusion23:14