Papers with classification experiments
Classifying the Informative Behaviour of Emoji in Microblogs (L18-1)
Copied to clipboard
| Challenge: | Emoji are pictographs used in microblogs as emotion markers, but can also represent a wider range of concepts. |
| Approach: | They analyze a corpus of tweets pairs and classify emoji with respect to redundancy . they propose to further investigate the informative behaviour of e-mails using eoji . |
| Outcome: | The proposed model achieved an F-score of 0.7 for emoji use in 2475 tweets pairs. |
The Subject Annotations of the Danish Parliament Corpus (2009-2017) - Evaluated with Automatic Multi-label Classification (2022.lrec-1)
Copied to clipboard
| Challenge: | The interest in analysing and automatically processing large amounts of political data has increased in the past decades. |
| Approach: | They address the semi-automatic annotation of subjects in the Danish Parliament Corpus (2009-2017) v.2 and describe multi-label classification experiments to verify the consistency of the subject annotation. |
| Outcome: | The proposed method improves on the baseline classifier, which is a majority classifier. |