Catalogue search • Linguistik portal • Fachinformationsdienst (FID)

1	Automatic prediction of text aesthetics and interestingness
	Ganguly, Debasis; Jones, Gareth J.F.; Leveling, Johannes
	In: Ganguly, Debasis orcid:0000-0003-0050-7138 , Leveling, Johannes orcid:0000-0003-0603-4191 and Jones, Gareth J.F. orcid:0000-0002-4033-9135 (2014) Automatic prediction of text aesthetics and interestingness. In: 25th International Conference on Computational Linguistics (COLING 2014), 23-29 Aug 2014, Dublin, Ireland. (2014)
	Abstract: This paper investigates the problem of automated text aesthetics prediction. The availability of user generated content and ratings, e.g. Flickr, has induced research in aesthetics prediction for non-text domains, particularly for photographic images. This problem, however, has yet not been explored for the text domain. Due to the very subjective nature of text aesthetics, it is dicult to compile human annotated data by methods such as crowd sourcing with a fair degree of inter-annotator agreement. The availability of the Kindle \popular highlights" data has motivated us to compile a dataset comprised of human annotated aesthetically pleasing and interesting text passages. We then undertake a supervised classication approach to predict text aesthetics by constructing real-valued feature vectors from each text passage. In particular, the features that we use for this classification task are word length, repetitions, polarity, part-of-speech, semantic distances; and topic generality and diversity. A traditional binary classication approach is not effective in this case because non-highlighted passages surrounding the highlighted ones do not necessarily represent the other extreme of unpleasant quality text. Due to the absence of real negative class samples, we employ the MC algorithm, in which training can be initiated with instances only from the positive class. On each successive iteration the algorithm selects new strong negative samples from the unlabeled class and retrains itself. The results show that the mapping convergence (MC) algorithm with a Gaussian and a linear kernel used for the mapping and convergence phases, respectively, yields the best results, achieving satisfactory accuracy, precision and recall values of about 74%, 42% and 54% respectively.
	Keyword: Computational linguistics; Information retrieval
	URL: http://doras.dcu.ie/20379/
	BASE
	Hide details

2	Overview of the ShARe/CLEF eHealth Evaluation Lab 2013
	Velupillai, Sumithra; Leveling, Johannes; Pradhan, Sameer...
	In: Suominen, Hanna, Salanterä, Sanna, Velupillai, Sumithra, Chapman, Wendy, Savova, Guergana, Elhadad, Noemie, Pradhan, Sameer, South, Brett R., Mowery, Danielle L., Jones, Gareth J.F. orcid:0000-0003-2923-8365 , Leveling, Johannes orcid:0000-0003-0603-4191 , Kelly, Liadh orcid:0000-0003-1131-5238 , Goeuriot, Lorraine orcid:0000-0001-7491-1980 , Martinez, David and Zuccon, Guido (2013) Overview of the ShARe/CLEF eHealth Evaluation Lab 2013. In: 4th International Conference of the CLEF Initiative (CLEF 2013), 23-26 Sept 2013, Valencia, Spain. ISBN 978-3-642-40802-1 (2013)
	BASE
	Show details

3	Making results fit into 40 characters: a study in document rewriting
	Jones, Gareth J.F.; Leveling, Johannes
	In: Leveling, Johannes orcid:0000-0003-0603-4191 and Jones, Gareth J.F. orcid:0000-0003-2923-8365 (2012) Making results fit into 40 characters: a study in document rewriting. In: The 35th Annual ACM SIGIR 2012 Conference, 12-16 Aug 2012, Portland, Oregon. (2012)
	BASE
	Show details

Search in the Catalogues and Directories