ArgMine

Brief and project description

ArgMine is a research project dedicated to the automatic extraction and analysis of argumentative structures from natural language texts. By combining techniques from Natural Language Processing, Machine Learning, and Argumentation Theory, the project aims to identify claims, premises, and reasoning patterns embedded in written discourse.

The project addresses one of the major challenges in language technologies: understanding not only what opinions are expressed in a text, but also the arguments and evidence that support them. This capability is essential for the development of more transparent and explainable AI systems capable of analysing complex human reasoning.

To support this objective, ArgMine developed computational methods and annotation tools for the creation of the first Portuguese corpus specifically designed for argumentation mining. The project integrates corpus construction, argument annotation, feature engineering, and machine learning experimentation into a unified framework for argument analysis.

ArgMine contributes to advancing research in opinion mining, recommender systems, decision-support systems, and multi-agent environments. Its technologies enable the automatic identification of argumentative discourse in domains such as media analysis, public debate, and scientific communication.

By fostering the intersection of computational linguistics and artificial intelligence, the project strengthens the development of human-centred AI systems capable of understanding, interpreting, and explaining argumentative reasoning in Portuguese.

Period: September 2014 – July 2015