Question answering
Question answering models are able to find or generate an answer to a question from a given text. In extractive question answering the model extracts (i.e., verbatim copies) an answer found in a given input text. On the other hand, in generative question answering a model generates the whole answer itself. Models can also be designed in a way that they solve a classification task - that is, they learn to choose the most appropriate answer among several given answers. This tool supports extractive question answering. The model finds the beginning and end of the excerpt of the given text, where an answer to the question is most likely to be found. It then returns the found text snippets as possible answers to the question. We return the three most probable answers as a result.
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.
Tool available at: https://github.com/RSDO-DS3/SloQA