Detection of stance and sentiment modifiers in political blogs
Skeppstedt, Maria ; Simaki, Vasiliki LU ; Paradis, Carita LU
and Kerren, Andreas
(2017)
In Lecture Notes in Artificial Intelligence
10458.
p.302-311
- Abstract
- The automatic detection of seven types of modifiers was studied: Certainty, Uncertainty, Hypotheticality, Prediction, Recommendation, Concession/Contrast and Source. A classifier aimed at detecting local cue words that signal the categories was the most successful method for five of the categories. For Prediction and Hypotheticality, however, better results were obtained with a classifier trained on tokens and bi-grams present in the entire sentence. Unsupervised cluster features were shown useful for the categories Source and Uncertainty, when a subset of the training data available was used. However, when all of the 2,095 sentences that had been actively selected and manually annotated were used as training data, the cluster features had... (More)
- The automatic detection of seven types of modifiers was studied: Certainty, Uncertainty, Hypotheticality, Prediction, Recommendation, Concession/Contrast and Source. A classifier aimed at detecting local cue words that signal the categories was the most successful method for five of the categories. For Prediction and Hypotheticality, however, better results were obtained with a classifier trained on tokens and bi-grams present in the entire sentence. Unsupervised cluster features were shown useful for the categories Source and Uncertainty, when a subset of the training data available was used. However, when all of the 2,095 sentences that had been actively selected and manually annotated were used as training data, the cluster features had a very limited effect. Some of the classification errors made by the models would be possible to avoid by extending the training data set, while other features and feature representations, as well as the incorporation of pragmatic knowledge, would be required for other error types. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/b171ca83-1af6-43b1-8563-5f88b82fed02
- author
- Skeppstedt, Maria
; Simaki, Vasiliki
LU
; Paradis, Carita
LU
and Kerren, Andreas
- organization
- publishing date
- 2017
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- stance modifiers, sentiment modifiers, active learning, unsupervised features, resource-aware natural language processing
- host publication
- Speech and computer : 19th International Conference, SPECOM 2017, Hatfield, UK, September 12-16, 2017, Proceedings - 19th International Conference, SPECOM 2017, Hatfield, UK, September 12-16, 2017, Proceedings
- series title
- Lecture Notes in Artificial Intelligence
- editor
- Karpov, Alexey ; Potapova, Rodmonga and Mporas, Iosif
- volume
- 10458
- pages
- 302 - 311
- publisher
- Springer International Publishing
- external identifiers
-
- scopus:85029498983
- ISBN
- 978-3-319-66428-6
- 978-3-319-66429-3
- DOI
- 10.1007/978-3-319-66429-3_29
- project
- StaViCTA - Advances in the description and explanation of stance in discourse using visual and computational text analytics
- language
- English
- LU publication?
- yes
- id
- b171ca83-1af6-43b1-8563-5f88b82fed02
- date added to LUP
- 2017-06-02 19:44:08
- date last changed
- 2024-02-12 20:10:30
@inproceedings{b171ca83-1af6-43b1-8563-5f88b82fed02,
abstract = {{The automatic detection of seven types of modifiers was studied: Certainty, Uncertainty, Hypotheticality, Prediction, Recommendation, Concession/Contrast and Source. A classifier aimed at detecting local cue words that signal the categories was the most successful method for five of the categories. For Prediction and Hypotheticality, however, better results were obtained with a classifier trained on tokens and bi-grams present in the entire sentence. Unsupervised cluster features were shown useful for the categories Source and Uncertainty, when a subset of the training data available was used. However, when all of the 2,095 sentences that had been actively selected and manually annotated were used as training data, the cluster features had a very limited effect. Some of the classification errors made by the models would be possible to avoid by extending the training data set, while other features and feature representations, as well as the incorporation of pragmatic knowledge, would be required for other error types.}},
author = {{Skeppstedt, Maria and Simaki, Vasiliki and Paradis, Carita and Kerren, Andreas}},
booktitle = {{Speech and computer : 19th International Conference, SPECOM 2017, Hatfield, UK, September 12-16, 2017, Proceedings}},
editor = {{Karpov, Alexey and Potapova, Rodmonga and Mporas, Iosif}},
isbn = {{978-3-319-66428-6}},
keywords = {{stance modifiers; sentiment modifiers; active learning; unsupervised features; resource-aware natural language processing}},
language = {{eng}},
pages = {{302--311}},
publisher = {{Springer International Publishing}},
series = {{Lecture Notes in Artificial Intelligence}},
title = {{Detection of stance and sentiment modifiers in political blogs}},
url = {{http://dx.doi.org/10.1007/978-3-319-66429-3_29}},
doi = {{10.1007/978-3-319-66429-3_29}},
volume = {{10458}},
year = {{2017}},
}