Taggus: an automated pipeline for the extraction of characters’ social networks from portuguese fiction literature

The Automatic identification of characters and their interactions from literary fiction is, arguably, a complex task that requires pipelines that leverage multiple Natural Language Processing (NLP) methods, such as Named Entity Recognition (NER) and Part-of-speech (POS) tagging. However, these methods are not optimized for retrieving Social Networks of Characters. Indeed, the currently available methods tend to underperform, especially in less-represented languages, due to a lack of manually annotated data for training. Here, we propose a pipeline, which we call Taggus, to extract social networks from literary fiction works in Portuguese without requiring a training phase. Our results show that compared to readily available State-of-the-Art tools—off-the-shelf NER tools and Large Language Models (ChatGPT)—the resulting pipeline, which uses POS tagging and a combination of heuristics, achieves satisfying results with an average F1-Score of $$92.1\%$$ in the task of identifying characters and solving for co-reference and $$74.0\%$$ in interaction detection. These represent, respectively, increases of $$115.7\%$$ and $$38.1\%$$ over the results achieved by the readily available State-of-the-Art tools. Further steps to improve results are outlined, including methods for detecting relationships among characters. Limitations on the size and scope of our testing samples are acknowledged and in the exclusive focus on interactions as the relationship type. The Taggus pipeline is publicly available to encourage development in this field for the Portuguese language.