Output details
11 - Computer Science and Informatics
University of Edinburgh
Article title
Stable Classification of Text Genres
Type
D - Journal article
Title of journal
Computational Linguistics
Article number
-
Volume number
37
Issue number
2
First page of article
385
ISSN of journal
0891-2017
Year of publication
2011
Number of additional authors
1
Additional information
<22> Originality: Introduces a new criterion of "stability" for genre classification systems (ie, that their performance does not vary with changes in the topic mix within the genre) and makes the new argument that parts-of-speech should be considered individually as features, since they vary in stability.
Significance: The criterion is beginning to be adopted by researchers working on genre classification.
Rigour: Based on 20-years worth of New York Times' articles, with all claims based on statistically significant results.
Interdisciplinary
-
Cross-referral requested
-
Research group
D - Institute for Language, Cognition & Computation
Citation count
1
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
