Relationship extraction

Relationship extraction is a natural language processing task that attempts to discover semantic relationships between words and/or phrases. In general, a task can be defined as finding a relationship without specific identification of the type of relationship only. It can also be limited to finding relationships among documents or within a single sentence. Within this project we semi-automatically built a corpus and trained a model capable of recognizing 29 types of relationships between mentions within a single sentence. A mention can represent a named entity or a reference to a named entity (e.g., a pronoun). The current model is adapted to the Wikipedia texts, so it performs worse on general texts. In addition to name entity recognition and coreference detection, link extraction is one of the basic and key tasks of information extraction.

Web service should be used for demonstration purposes only, and is limited by the number of requests per time unit and input length. To use the service within your applications, please download results of the projects, available in the Clarin.si repository.