Weakly Supervised Cross-Lingual Named Entity Recognition via Effective Annotation and Representation Projection

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1470–1480 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1135 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1470–1480 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1135 Weakly Supervised Cross-Lingual Named Entity Recognition via Effective Annotation and Representation Projection Jian Ni and Georgiana Dinu and Radu Florian IBM T. J. Watson Research Center 1101 Kitchawan Road, Yorktown Heights, NY 10598, USA {nij, gdinu, raduf}@us.ibm.com Abstract The state-of-the-art named entity recog- nition (NER) systems are supervised ma- chine learning models that require large amounts of manually annotated data to achieve high accuracy. However, anno- tating NER data by human is expensive and time-consuming, and can be quite dif- ficult for a new language. In this pa- per, we present two weakly supervised ap- proaches for cross-lingual NER with no human annotation in a target language. The first approach is to create automati- cally labeled NER data for a target lan- guage via annotation projection on compa- rable corpora, where we develop a heuris- tic scheme that effectively selects good- quality projection-labeled data from noisy data. The second approach is to project distributed representations of words (word embeddings) from a target language to a source language, so that the source- language NER system can be applied to the target language without re-training. We also design two co-decoding schemes that effectively combine the outputs of the two projection-based approaches. We evaluate the performance of the proposed approaches on both in-house and open NER data for several target languages. The results show that the combined sys- tems outperform three other weakly super- vised approaches on the CoNLL data. 1 Introduction Named entity recognition (NER) is a fundamen- tal information extraction task that automatically detects named entities in text and classifies them into pre-defined entity types such as PERSON, ORGANIZATION, GPE (GeoPolitical Entities), EVENT, LOCATION, TIME, DATE, etc. NER provides essential inputs for many information ex- traction applications, including relation extraction, entity linking, question answering and text min- ing. Building fast and accurate NER systems is a crucial step towards enabling large-scale auto- mated information extraction and knowledge dis- covery on the huge volumes of electronic docu- ments existing today. The state-of-the-art NER systems are super- vised machine learning models (Nadeau and Sekine, 2007), including maximum entropy Markov models (MEMMs) (McCallum et al., 2000), conditional random fields (CRFs) (Lafferty et al., 2001) and neural networks (Collobert et al., 2011; Lample et al., 2016). To achieve high ac- curacy, a NER system needs to be trained with a large amount of manually annotated data, and is often supplied with language-specific resources (e.g., gazetteers, word clusters, etc.). Annotating NER data by human is rather expensive and time- consuming, and can be quite difficult for a new language. This creates a big challenge in building NER systems of multiple languages for supporting multilingual information extraction applications. The difficulty of …

Content truncated to 500 words