Abstractive Summarizers Become Emotional on News Summarization
Emotions are central to understanding contemporary journalism; however, they are overlooked in automatic news summarization. Actually, summaries are an entry point to the source article that could favor some emotions to captivate the reader. Nevertheless, the emotional content of summarization corpo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/2/713 |
| _version_ | 1827369886260658176 |
|---|---|
| author | Vicent Ahuir José-Ángel González Lluís-F. Hurtado Encarna Segarra |
| author_facet | Vicent Ahuir José-Ángel González Lluís-F. Hurtado Encarna Segarra |
| author_sort | Vicent Ahuir |
| collection | DOAJ |
| description | Emotions are central to understanding contemporary journalism; however, they are overlooked in automatic news summarization. Actually, summaries are an entry point to the source article that could favor some emotions to captivate the reader. Nevertheless, the emotional content of summarization corpora and the emotional behavior of summarization models are still unexplored. In this work, we explore the usage of established methodologies to study the emotional content of summarization corpora and the emotional behavior of summarization models. Using these methodologies, we study the emotional content of two widely used summarization corpora: <span style="font-variant: small-caps;">Cnn/Dailymail</span> and <span style="font-variant: small-caps;">Xsum</span>, and the capabilities of three state-of-the-art transformer-based abstractive systems for eliciting emotions in the generated summaries: <span style="font-variant: small-caps;">Bart</span>, <span style="font-variant: small-caps;">Pegasus</span>, and <span style="font-variant: small-caps;">T5</span>. The main significant findings are as follows: (i) emotions are persistent in the two summarization corpora, (ii) summarizers approach moderately well the emotions of the reference summaries, and (iii) more than 75% of the emotions introduced by novel words in generated summaries are present in the reference ones. The combined use of these methodologies has allowed us to conduct a satisfactory study of the emotional content in news summarization. |
| first_indexed | 2024-03-08T09:59:11Z |
| format | Article |
| id | doaj.art-c45dfd8d677840fa9fe67dd557bc9411 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-08T09:59:11Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-c45dfd8d677840fa9fe67dd557bc94112024-01-29T13:43:53ZengMDPI AGApplied Sciences2076-34172024-01-0114271310.3390/app14020713Abstractive Summarizers Become Emotional on News SummarizationVicent Ahuir0José-Ángel González1Lluís-F. Hurtado2Encarna Segarra3Valencian Research Institute for Artificial Intelligence (VRAIN), Universitat Politècnica de València, 46022 Valencia, SpainSymanto Symanto Research, C/Reina 12, 46011 Valencia, SpainValencian Research Institute for Artificial Intelligence (VRAIN), Universitat Politècnica de València, 46022 Valencia, SpainValencian Research Institute for Artificial Intelligence (VRAIN), Universitat Politècnica de València, 46022 Valencia, SpainEmotions are central to understanding contemporary journalism; however, they are overlooked in automatic news summarization. Actually, summaries are an entry point to the source article that could favor some emotions to captivate the reader. Nevertheless, the emotional content of summarization corpora and the emotional behavior of summarization models are still unexplored. In this work, we explore the usage of established methodologies to study the emotional content of summarization corpora and the emotional behavior of summarization models. Using these methodologies, we study the emotional content of two widely used summarization corpora: <span style="font-variant: small-caps;">Cnn/Dailymail</span> and <span style="font-variant: small-caps;">Xsum</span>, and the capabilities of three state-of-the-art transformer-based abstractive systems for eliciting emotions in the generated summaries: <span style="font-variant: small-caps;">Bart</span>, <span style="font-variant: small-caps;">Pegasus</span>, and <span style="font-variant: small-caps;">T5</span>. The main significant findings are as follows: (i) emotions are persistent in the two summarization corpora, (ii) summarizers approach moderately well the emotions of the reference summaries, and (iii) more than 75% of the emotions introduced by novel words in generated summaries are present in the reference ones. The combined use of these methodologies has allowed us to conduct a satisfactory study of the emotional content in news summarization.https://www.mdpi.com/2076-3417/14/2/713news summarizationabstractive summarizationemotional contentemotional behavior |
| spellingShingle | Vicent Ahuir José-Ángel González Lluís-F. Hurtado Encarna Segarra Abstractive Summarizers Become Emotional on News Summarization Applied Sciences news summarization abstractive summarization emotional content emotional behavior |
| title | Abstractive Summarizers Become Emotional on News Summarization |
| title_full | Abstractive Summarizers Become Emotional on News Summarization |
| title_fullStr | Abstractive Summarizers Become Emotional on News Summarization |
| title_full_unstemmed | Abstractive Summarizers Become Emotional on News Summarization |
| title_short | Abstractive Summarizers Become Emotional on News Summarization |
| title_sort | abstractive summarizers become emotional on news summarization |
| topic | news summarization abstractive summarization emotional content emotional behavior |
| url | https://www.mdpi.com/2076-3417/14/2/713 |
| work_keys_str_mv | AT vicentahuir abstractivesummarizersbecomeemotionalonnewssummarization AT joseangelgonzalez abstractivesummarizersbecomeemotionalonnewssummarization AT lluisfhurtado abstractivesummarizersbecomeemotionalonnewssummarization AT encarnasegarra abstractivesummarizersbecomeemotionalonnewssummarization |