Summary of Automatic Emotion Modelling in Written Stories, by Lukas Christ et al.
Automatic Emotion Modelling in Written Stories
by Lukas Christ, Shahin Amiriparian, Manuel Milling, Ilhan Aslan, Björn W. Schuller
First submitted to arxiv on: 21 Dec 2022
Categories
- Main: Computation and Language (cs.CL)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The abstract presents a new approach to automatically modeling emotional trajectories in stories, which can evoke emotions and influence the affective states of the audience. The authors propose several novel Transformer-based methods for predicting valence and arousal signals over the course of written stories. They also explore strategies for fine-tuning a pretrained ELECTRA model and study the benefits of considering a sentence’s context when inferring its emotionality. The best configuration achieves a Concordance Correlation Coefficient (CCC) of .7338 for valence and .6302 for arousal on the test set, demonstrating the suitability of the proposed approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Stories are an important way to communicate with others and can make us feel happy or sad. Researchers have been trying to figure out how to use computers to understand the emotions in stories. One problem is that there isn’t a good way to measure how emotional a story is, so the authors created new labels for an existing dataset of children’s stories. They then used these labels to train a computer model to predict how emotional each part of a story is. The best model they tested was able to accurately identify the emotions in a story about 73% of the time. |
Keywords
» Artificial intelligence » Fine tuning » Transformer
