Evaluating KGR10 Polish word embeddings in the recognition of temporal expressions using BiLSTM-CRF
This work addresses the specific challenge of temporal expression recognition for the Polish language, representing an incremental advancement in natural language processing for this domain.
The authors tackled the problem of recognizing temporal expressions in Polish by evaluating new word embeddings from the KGR10 corpus, achieving results that demonstrate their effectiveness in a BiLSTM-CRF model.
The article introduces a new set of Polish word embeddings, built using KGR10 corpus, which contains more than 4 billion words. These embeddings are evaluated in the problem of recognition of temporal expressions (timexes) for the Polish language. We described the process of KGR10 corpus creation and a new approach to the recognition problem using Bidirectional Long-Short Term Memory (BiLSTM) network with additional CRF layer, where specific embeddings are essential. We presented experiments and conclusions drawn from them.