From text to political positions : text analysis across disciplines /
This chapter explores how three methods of political text analysis can complement each other to differentiate parties in detail. A word-frequency method and corpus linguistic techniques are joined by critical discourse analysis in an attempt to assess the ideological relation between election manife...
Saved in:
| Online Access: |
Full text (MCPHS users only) |
|---|---|
| Other Authors: | , , |
| Format: | Electronic eBook |
| Language: | English |
| Published: |
Amsterdam :
John Benjamins Publishing Co.,
2014
|
| Series: | Discourse approaches to politics, society, and culture.
|
| Subjects: | |
| Local Note: | ProQuest Ebook Central |
Table of Contents:
- From Text to Political Positions; Editorial page ; Title page ; LCC data ; Table of contents; Foreword; Reference; Acknowledgements; 1. Positions of parties and political cleavages between parties in texts; Political language and content analysis ; Meta-language about political language ; Three tools for analysing political texts ; Does a concept occur? First-order agenda setting and entity recognition ; The ontological problem: (named) entity recognition ; Co-occurrence of concepts? Conditional probabilities and associative framing ; Semantic network analysis.
- Manual coding using the NET-method Automation using semantic rules on top of an ontology, POS-tags, syntax dependency trees; Summary ; References ; Part I. Computational methods for political text analysis; Introduction; 2. Comparing the position of Canadian political parties using French and English manifestos; Word-based parallel content analysis ; Methodology ; Canadian expert surveys ; Wordscores ; Wordfish ; Conclusion ; References; Appendix ; 3. Leveraging textual sentiment analysis with social network modeling; 1. Introduction ; 2. Data ; 2.1 Election data; 2.2 Sentiment annotations.
- 3. Related work 4. Overview of the classification framework ; 4.1 Shallow document classification ; 4.2 Deep entity-level sentiment scoring ; 4.3 Social network modeling ; 4.4 Overview of algorithms ; 5. Experiments ; 5.1 Experimental conditions ; 5.2 Evaluation measures ; 5.3 Discussion ; 5.4 Significance of results ; 5.5 Future work ; 6. Conclusion ; References; 4. Issue framing and language use in the Swedish blogosphere; Introduction; The case of Sweden: Issue framing and the 'outsider' concept ; Methodological considerations ; Random indexing.
- Language use by the Social Democratic and the Conservative Moderate Party in relation to 'outsiders'The Conservative Moderate Party ; The Social Democratic Party ; From quality to quantity in party related documents ; Random Indexing of words related to 'outsider' in the Swedish blogosphere 2008-2010 ; Summary and conclusions ; References; Appendix ; 5. Text to ideology or text to party status?; 1. Introduction ; 2. Background: The Canadian party system and Parliament ; 3. First set of experiments: Classifying by party ; 3.1 Data ; 3.2 Method ; 3.3 Results ; 3.4 Discussion.
- 4. Second set of experiments: Classifying by party status 4.1 Data ; 4.2 Method and results ; 4.3 Discussion ; 5. Classification based on the emotional content of speeches ; 5.1 Method and data ; 5.2 Results ; 6. Third set of experiments: European Parliamentary data ; 6.1 Data ; 6.2 Method; 6.3 Results ; 6.4 Discussion ; 7. Conclusion ; References; 6. Sentiment analysis in parliamentary proceedings; 1. Introduction ; 2. Background ; 3. Data ; 4. Assessing subjectivity and orientation ; 4.1 Classification level ; 4.2 Gold standard corpus ; 4.3 Automatically determining subjectivity.
