Sign inSign up
Automatic Discovery of Patterns in News Content thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Automatic Discovery of Patterns in News Content

Published on Apr 25, 20127053 Views

Nello Cristianini

Workshops 2012 - Cumberland Lodge

Related categories

Data MiningMachine LearningText MiningPsychologyInformation SocietyInnovation Journalism

Chapter list

Automatic Discovery of Patterns in News Content00:00
An Interesting Fact00:17
Feedback01:55
Cultivation Theory03:04
News Content Analysis04:14
News Coding04:31
The MediaPatterns Project04:55
The Problem with Large Projects...05:43
Getting the Data06:29
The NOAM infrastructure06:34
The Data (1)06:48
Analysis of the Data07:16
Our Questions07:46
Question 1: What Is In the News?07:59
From Articles to Stories08:35
People in the News08:55
More in detail...09:08
People in the News09:15
M:F Ratio09:16
Gender Bias in the Media09:55
Detecting Topics10:14
M:F by Topic10:23
Validation12:03
Observations12:49
Question1: Which stories are covered by which outlets?13:01
Mapping the EU Mediasphere13:24
The Data (2)13:45
Outlets Covering Same Stories14:09
Linking Countries (1)15:12
Linking Countries (2)15:26
Explaining the Relations16:14
Embedding: MDS on content similarity17:04
Question 2: What Readers Want?19:56
What Readers Want? (1)20:11
What Readers Want? (2)21:03
Ranking SVM21:25
Weight vector21:42
Average relative distances between outlets23:11
Appeal vs. non-public-affairs bias23:46
Question 3: Writing Style and Narrative Patterns24:56
Writing Style25:10
Readability25:57
Linguistic Subjectivity26:20
Outlet Similarity by Style26:23
Topic Similarity by Style27:00
Outlet Similarity by Topic Distribution27:30
Appeal vs. Linguistic Subjectivity28:01
Validations28:25
Relation: style vs. demographics28:55
Narrative analysis29:16
Cluster (1)30:10
NY Times Corpus, Year 2002 – crime stories31:28
Networks of political support31:43
Cluster (2)31:57
Topology32:30
Cluster (3)33:20
Video (1)33:43
Video (2)34:14
Question 4: Measuring Public Mood35:11
Time Series for Anger in Twitter Content35:59
Time Series for Fear in Twitter Content36:41
Rate of Mood Change by Day using the Difference in 50-day Mean36:51
The Face of Britain...37:20
Animation of Mood Changes37:46
Demos39:24
electionwatch.enm.bris.ac.uk39:46
foundintranslation.enm.bris.ac.uk39:53
Meme watch39:57
celebwatch.enm.bris.ac.uk39:59
Flu detector40:03
Conclusions (1)40:06
Conclusions (2)41:02
Thanks To41:28
MediaPatterns.enm.bristol.ac.uk41:39