We present an update to UDPipe 1.0 (Straka et al., 2016), a trainable pipeline which performs sentence segmentation, tokenization, POS tagging, lemmatization and dependency parsing. We provide models for all 50 languages of UD 2.0, and furthermore, the pipeline can be trained easily using data in CoNLL-U format.
We propose two neural network architectures for nested named entity recognition (NER), a setting in which named entities may overlap and also be labeled with more than one label. We encode the nested labels using a linearized scheme. In our first proposed approach, the nested labels are modeled as multilabels corresponding to the Cartesian product of the nested labels in a standard LSTM-CRF architecture. In the second one, the nested NER is viewed as a sequence-to-sequence problem, in which the input sequence consists of the tokens and output sequence of the labels, using hard attention on the word whose label is being predicted. The proposed methods outperform the nested NER state of the art on four corpora: ACE-2004, ACE-2005, GENIA and Czech CNEC. We also enrich our architectures with the recently published contextual embeddings: ELMo, BERT and Flair, reaching further improvements for the four nested entity corpora. In addition, we report flat NER stateof-the-art results for CoNLL-2002 Dutch and Spanish and for CoNLL-2003 English.
We present two recently released opensource taggers: NameTag is a free software for named entity recognition (NER) which achieves state-of-the-art performance on Czech; MorphoDiTa (Morphological Dictionary and Tagger) performs morphological analysis (with lemmatization), morphological generation, tagging and tokenization with state-of-the-art results for Czech and a throughput around 10-200K words per second. The taggers can be trained for any language for which annotated data exist, but they are specifically designed to be efficient for inflective languages, Both tools are free software under LGPL license and are distributed along with trained linguistic models which are free for non-commercial use under the CC BY-NC-SA license. The releases include standalone tools, C++ libraries with Java, Python and Perl bindings and web services.
We propose a new architecture for diacritics restoration based on contextualized embeddings, namely BERT, and we evaluate it on 12 languages with diacritics. Furthermore, we conduct a detailed error analysis on Czech, a morphologically rich language with a high level of diacritization. Notably, we manually annotate all mispredictions, showing that roughly 44% of them are actually not errors, but either plausible variants (19%), or the system corrections of erroneous data (25%). Finally, we categorize the real errors in detail. We release the code at https://github.com/ufal/bert-diacritics-restoration.
Contextualized embeddings, which capture appropriate word meaning depending on context, have recently been proposed. We evaluate two methods for precomputing such embeddings, BERT and Flair, on four Czech text processing tasks: part-of-speech (POS) tagging, lemmatization, dependency parsing and named entity recognition (NER). The first three tasks, POS tagging, lemmatization and dependency parsing, are evaluated on two corpora: the Prague Dependency Treebank 3.5 and the Universal Dependencies 2.3. The named entity recognition (NER) is evaluated on the Czech Named Entity Corpus 1.1 and 2.0. We report state-of-the-art results for the above mentioned tasks and corpora.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.