NLP meets Computational Social (Media) Science thumbnail
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

NLP meets Computational Social (Media) Science

Published on Jun 06, 20171345 Views

Language is the ultimate social medium: We don't just communicate to convey information, but also to entertain, to gossip, to console, and much more. Social media data is one of the purest expressions

Related categories

Chapter list

Creation and Use of Social Media Resources00:00
NLP meets Computational Social (Media) Science00:10
Social Media00:56
Natural Language Processing - 102:07
Natural Language Processing - 203:28
Who buys what? - 105:02
Who buys what? - 206:04
Sexism in Politics06:34
Censorship07:20
Neighborhoods07:46
Language Change08:34
Measuring Language Variation09:06
Visualizing Language Variation10:01
Personality10:59
The Transparent User11:36
Workshop on NLP and Computational Social Sciences12:31
That’s it, NLP can help social science and business12:56
…no, really…13:04
Ok, so maybe there’s one thing13:09
The Assumption13:21
Language Distributions14:01
Shannon Game - 114:47
Shannon Game - 215:30
Language16:02
Demographics16:29
The Problem17:25
The Consequences18:44
More Data?19:15
Improving NLP with Demographic Information20:06
Example 1: Text Classification20:16
Data - 120:26
Embeddings21:28
Systems22:15
Results for Age (avg)22:45
Results for Gender (avg)23:53
Example 2:24:56
Mental Health Risk Prediction26:01
Comorbidity and Correlation28:36
Motivation from “The Karate Kid”30:14
Data - 231:17
Results ROC32:54
Results: TPR@FPR=0.133:28
So then all is good, right?35:20
…right?35:32
Challenges35:35
Ethics and NLP35:42
Exclusion - 137:19
Exclusion - 238:44
Overgeneralization39:19
The Signal and the Noise39:52
Hate Speech Detection40:38
Exposure40:51
Dual Use42:17
What can we do?42:56
Ethics44:46
Wrapping up45:01
Conclusion45:06
Open questions45:55
Thanks!46:39
Questions?46:41