Pregled bibliografske jedinice broj: 930985
Predicting News Values from Headline Text and Emotions
Predicting News Values from Headline Text and Emotions // Proceedings of the 2017 EMNLP Workshop on Natural Language Processing Meets Journalism
Kopenhagen, Danska, 2017. str. 1-6 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 930985 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Predicting News Values from Headline Text and Emotions
Autori
Di Buono, Maria Pia ; Šnajder, Jan ; Dalbelo Bašić, Bojana ; Glavaš, Goran ; Tutek, Martin ; Milic-Frayling, Nataša
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of the 2017 EMNLP Workshop on Natural Language Processing Meets Journalism
/ - , 2017, 1-6
Skup
2017 EMNLP Workshop on Natural Language Processing Meets Journalism
Mjesto i datum
Kopenhagen, Danska, 07.09.2017. - 11.09.2017
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
machine learning, prediction, text classification, SVN, CNN, clustering, factorial analysis, news values
Sažetak
We present a preliminary study on predicting news values from headline text and emotions. We perform a multivariate analysis on a dataset manually annotated with news values and emotions, discovering interesting correlations among them. We then train two competitive machine learning models – an SVM and a CNN – to predict news values from headline text and emotions as features. We find that, while both models yield a satisfactory performance, some news values are more difficult to detect than others, while some profit more from including emotion information.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
