Although modern named entity recognition (NER) systems show impressive performance on standard datasets, they perform poorly when presented with noisy data. In particular, capitalization is a strong signal for entities in many languages, and even state of the art models overfit to this feature, with drastically lower performance on uncapitalized text. In this work, we address the problem of robustness of NER systems in data with noisy or uncertain casing, using a pretraining objective that predicts casing in text, or a truecaser, leveraging unlabeled data. The pretrained truecaser is combined with a standard BiLSTM-CRF model for NER by appending output distributions to character embeddings. In experiments over several datasets of varying domain and casing quality, we show that our new model improves performance in uncased text, even adding value to uncased BERT embeddings. Our method achieves a new state of the art on the WNUT17 shared task dataset.

Supervised machine learning assumes the availability of fully-labeled data, but in many cases, such as low-resource languages, the only data available is partially annotated.We study the problem of Named Entity Recognition (NER) with partially annotated training data in which a fraction of the named entities are labeled, and all other tokens, entities or otherwise, are labeled as non-entity by default.In order to train on this noisy dataset, we need to distinguish between the true and false negatives.To this end, we introduce a constraintdriven iterative algorithm that learns to detect false negatives in the noisy set and downweigh them, resulting in a weighted training set.With this set, we train a weighted NER model.We evaluate our algorithm with weighted variants of neural and non-neural NER models on data in 8 languages from several language and script families, showing strong ability to learn from partial data.Finally, to show real-world efficacy, we evaluate on a Bengali NER corpus annotated by non-speakers, outperforming the prior state-of-the-art by over 5 points F1.

Stephen Mayhew, Tatiana Tsygankova, Dan Roth. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Character-level patterns have been widely used as features in English Named Entity Recognition (NER) systems.However, to date there has been no direct investigation of the inherent differences between name and nonname tokens in text, nor whether this property holds across multiple languages.This paper analyzes the capabilities of corpus-agnostic Character-level Language Models (CLMs) in the binary task of distinguishing name tokens from non-name tokens.We demonstrate that CLMs provide a simple and powerful model for capturing these differences, identifying named entity tokens in a diverse set of languages at close to the performance of full NER systems.Moreover, by adding very simple CLM-based features we can significantly improve the performance of an off-the-shelf NER system for multiple languages.1

Recent work in NLP has attempted to deal with low-resource languages but still assumed a resource level that is not present for most languages, e.g., the availability of Wikipedia in the target language. We propose a simple method for cross-lingual named entity recognition (NER) that works well in settings with very minimal resources. Our approach makes use of a lexicon to "translate" annotated data available in one or several high resource language(s) into the target language, and learns a standard monolingual NER model there. Further, when Wikipedia is available in the target language, our method can enhance Wikipedia based methods to yield state-of-the-art NER results; we evaluate on 7 diverse languages, improving the state-of-the-art by an average of 5.5% F1 points. With the minimal resources required, this is an extremely portable cross-lingual NER approach, as illustrated using a truly low-resource language, Uyghur.

Named Entity Recognition (NER) models for language L are typically trained using annotated data in that language.We study cross-lingual NER, where a model for NER in L is trained on another, source, language (or multiple source languages).We introduce a language independent method for NER, building on cross-lingual wikification, a technique that grounds words and phrases in non-English text into English Wikipedia entries.Thus, mentions in any language can be described using a set of categories and FreeBase types, yielding, as we show, strong language-independent features.With this insight, we propose an NER model that can be applied to all languages in Wikipedia.When trained on English, our model outperforms comparable approaches on the standard CoNLL datasets (Spanish, German, and Dutch) and also performs very well on lowresource languages (e.g., Turkish, Tagalog, Yoruba, Bengali, and Tamil) that have significantly smaller Wikipedia.Moreover, our method allows us to train on multiple source languages, typically improving NER results on the target languages.Finally, we show that our languageindependent features can be used also to enhance monolingual NER systems, yielding improved results for all 9 languages.