The Course of Emotion in Three Centuries of German Text A - - PowerPoint PPT Presentation

the course of emotion in three centuries of german text
SMART_READER_LITE
LIVE PREVIEW

The Course of Emotion in Three Centuries of German Text A - - PowerPoint PPT Presentation

August 9, 2017, Montreal, Canada Digital Humanities 2017 The Course of Emotion in Three Centuries of German Text A Methodological Framework Sven Buechel 1 , Johannes Hellrich 1,2 and Udo Hahn 1 1 Jena University Language & Information


slide-1
SLIDE 1

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 1

The Course of Emotion in Three Centuries of German Text

A Methodological Framework

Sven Buechel1, Johannes Hellrich1,2 and Udo Hahn1

2 Graduate School

’The Romantic Model’

http://www.modellromantik.uni-jena.de

Friedrich Schiller University Jena, Jena, Germany

1 Jena University Language & Information Engineering

(JULIE) Lab

http://www.julielab.de

slide-2
SLIDE 2

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 2

Motivation: Emotion in the Humanities

http://www.br.de/telekolleg/faecher/deutsch/literatur/goethe-weimarer-klassik-100.html https://de.wikipedia.org/wiki/Laokoon-Gruppe https://de.wikipedia.org/wiki/Der_Schrei

slide-3
SLIDE 3

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 3

Resource Problem in Sentiment Analysis

Contemporary Resources Historical Resources

?

slide-4
SLIDE 4

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 4

Methodological Framework

lexicon

lexicon

adapt + expand apply for emotion analysis modern historically adapted historical text

slide-5
SLIDE 5

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 5

Methods

slide-6
SLIDE 6

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 6

Representing Emotion: VAD

−1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0 −1.0 −0.5 0.0 0.5 1.0

  • Anger

Surprise Disgust Fear Sadness Joy

Valence

(displeasure—pleasure)

Dominance

(in control—being controlled)

slide-7
SLIDE 7

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 7

Emotion Lexicons

  • Lexical resources
  • Store emotion values of individual words
  • Many lexicons available from psychology

Lemma Valence Arousal Dominance sunshine 8.1 5.3 5.4 terrorism 1.6 7.4 2.7 calm 6.9 1.7 7.4 (Warriner et al., 2013) (1-to-9 scales)

slide-8
SLIDE 8

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 8

Lexicon Expansion

mouse filth rat computer keyboard

  • Automatic expansion based on Seed
  • Turney-Littman-Algorithm:
slide-9
SLIDE 9

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 9

Measuring Similarity: Word Embeddings

  • Dominant approach to similarity
  • Dense, low-dimensional vectors

as word representation

  • Most common: word2vec

(SGNS) (Mikolov et al., 2013)

  • SVDPPMI supposedly better

suited for humanities (Levy et al.,

2015, Hellrich & Hahn, 2016)

Ø Johannes’ talk for details! (Friday, 9am, LP-26)

read listen novel poem

  • pera

song

slide-10
SLIDE 10

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 10

Classical Expansion Methodology

contemp. corpus

contemp. embeddings expansion

contemp. expanded lexicon contemp. lexicon

slide-11
SLIDE 11

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 11

Modifications to Expansion Methodology

historical embeddings

expansion

expanded+ adapted Lexicon

contemp. lexicon

historical. corpus

slide-12
SLIDE 12

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 12

Intuition of Temporal Adaptation (Before)

mouse filth rat computer keyboard

slide-13
SLIDE 13

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 13

Intuition of Temporal Adaptation (After)

mouse filth rat computer keyboard

slide-14
SLIDE 14

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 14

Experiments I

(Lexicon Induction)

slide-15
SLIDE 15

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 15

Methodological Framework

lexicon

lexicon

adapt + expand apply for emotion analysis modern historically adapted historical text

slide-16
SLIDE 16

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 16

Resources

  • Seed Lexicon:

– ANGST (Schmidtke et al. (2014) – A thousand German word-emotion pairs – Well established acquisition methodology

  • Target corpus:

– Deutsches Textarchiv (DTA) [“German Text Archive”] – Balanced and manually transcribed – Rich meta-data (e.g., genre and subgenre)

slide-17
SLIDE 17

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 17

Target Corpus: DTA

Im DTA verfügbare Werke

nach Genre und Dekade Belletristik Gebrauchsliteratur Wissenschaft 1601ff. 1611ff. 1621ff. 1631ff. 1641ff. 1651ff. 1661ff. 1671ff. 1681ff. 1691ff. 1701ff. 1711ff. 1721ff. 1731ff. 1741ff. 1751ff. 1761ff. 1771ff. 1781ff. 1791ff. 1801ff. 1811ff. 1821ff. 1831ff. 1841ff. 1851ff. 1861ff. 1871ff. 1881ff. 1891ff. 1901ff.

50 100 150

Werke

Documents 1600s 1700s 1800s 1900s Academia Functional Belles lettres

  • 1st third shows different genre distribution
  • Individual decades comprise too little text

Ø Aggregate 30-years slices Ø Select 1690-1899 (~ 1k documents, 7 slices)

http://www.deutsches-textarchiv.de/doku/textauswahl

slide-18
SLIDE 18

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 18

Target Corpus: DTA

Im DTA verfügbare Werke

nach Genre und Dekade Belletristik Gebrauchsliteratur Wissenschaft 1601ff. 1611ff. 1621ff. 1631ff. 1641ff. 1651ff. 1661ff. 1671ff. 1681ff. 1691ff. 1701ff. 1711ff. 1721ff. 1731ff. 1741ff. 1751ff. 1761ff. 1771ff. 1781ff. 1791ff. 1801ff. 1811ff. 1821ff. 1831ff. 1841ff. 1851ff. 1861ff. 1871ff. 1881ff. 1891ff. 1901ff.

50 100 150

Werke

Documents 1600s 1700s 1800s 1900s Academia Functional Belles lettres

  • 1st third shows different genre distribution
  • Individual decades comprise too little text

Ø Aggregate 30-years slices Ø Select 1690-1899 (~ 1k documents, 7 slices)

http://www.deutsches-textarchiv.de/doku/textauswahl

slide-19
SLIDE 19

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 19

Affective Course of Sünde [Sin]

  • Strong increase in Valence since onset of Enlightenment
  • Backing up our findings with corpus linguistics…

1705 1735 1765 1795 1825 1855 1885 −4 −2 2 4 Year Standard Deviations

  • Valence

Arousal Dominance

slide-20
SLIDE 20

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 20

Top 10 Collocations for Sünde [Sin]

Rank Lemma and Translation 1690s 1890s 1 todt- death (prefix use) Lamm lamb 2 Erzürnung infuriation hinwegnehmen to take away 3 läßlich minor (clerical) Verzeihung forgiveness 4 beichten to confess Ausschweifung excess 5 Nachlaß abatement Gotte god 6 Grobheit crudeness Schande shame 7 verschweigen to conceal Reue repentance 8 beweinen to beweep Ärgernis nuisance 9 pichen to pitch Laster vice 10 beichten to confess aufrichtig sincere

slide-21
SLIDE 21

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 21

Experiments II

(Application to Historical Text)

slide-22
SLIDE 22

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 22

Methodological Framework

lexicon

lexicon

adapt + expand apply for emotion analysis modern historically adapted historical text

slide-23
SLIDE 23

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 23

Sentence-Level Emotion Prediction: JEMAS

Available:

https://github.com/JULIELab/JEmAS (Buechel & Hahn, ECAI 2016) (sunshine, <8, 3, 5>) (terrorism, <2, 7, 3>) (calm, <7, 2, 7>)

full text documents linguistic normalization BOW representation

word emotion lexicon

VAD score calculation

slide-24
SLIDE 24

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 24

Distinction of DTA Main Classes

Belles Lettres Academic

−8 −4 2 4 −4 −2 2 Valence Arousal

−4 −2 2 −4 2 4 Arousal Dominance −8 −4 2 4 −4 2 4 Valence Dominance

slide-25
SLIDE 25

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 25

Distinction between Principal Literary Forms

−8 −4 2 4 −4 −2 2 Valence Arousal

Lyric Narratives Drama

−4 −2 2 −4 2 4 Arousal Dominance

−8 −4 2 4 −4 2 4 Valence Dominance

slide-26
SLIDE 26

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 26

Distinction of Academic Subclasses

−8 −4 2 4 −4 −2 2 Valence Arousal

−8 −4 2 4 −4 2 4 Valence Dominance −4 −2 2 −4 2 4 Arousal Dominance

Law Philosophy Mathematics Technology Physics

slide-27
SLIDE 27

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 27

Development of Literary Forms (1690-1719)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-28
SLIDE 28

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 28

Development of Literary Forms (1720-1749)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-29
SLIDE 29

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 29

Development of Literary Forms (1750-1779)

  • Lyric

Narrative Drama

Valence Valence Dominance

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-30
SLIDE 30

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 30

Development of Literary Forms (1780-1809)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-31
SLIDE 31

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 31

Development of Literary Forms (1810-1839)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-32
SLIDE 32

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 32

Development of Literary Forms (1840-1869)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

3

Dominance

3

slide-33
SLIDE 33

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 33

Development of Literary Genres (1870-1899)

  • Lyric

Narrative Drama

−4 2 −3 −1 1

Valence Arousal

−4 2 −3 −1 1 3

Valence Dominance

−3 −1 1 −3 −1 1 3

Dominance Arousal

slide-34
SLIDE 34

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 34

Conclusion

slide-35
SLIDE 35

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 35

Conclusion

  • Methodological framework for quantifying emotion in non-

contemporary text

  • Combination of temporally-specific lexicon expansion and

lexicon-based emotion analysis

  • Application to German historical corpus as proof of concept
  • Qualitative validation on word and text level

Ø Genres show distinct and dynamic emotional patterns! Ø Resources available:

https://github.com/JULIELab/HistEmo

slide-36
SLIDE 36

Digital Humanities 2017 August 9, 2017, Montreal, Canada Sven Buechel, Johannes Hellrich and Udo Hahn The Course of Emotion in Three Centuries of German Text 36

The Course of Emotion in Three Centuries of German Text

A Methodological Framework

Sven Buechel1, Johannes Hellrich1,2 and Udo Hahn1

2 Graduate School

’The Romantic Model’

http://www.modellromantik.uni-jena.de

Friedrich Schiller University Jena, Jena, Germany

1 Jena University Language & Information Engineering

(JULIE) Lab

http://www.julielab.de