A Joint Model to Identify and Align Bilingual Named Entities

Abstract

In this article, an integrated model is derived that jointly identifies and aligns bilingual named entities (NEs) between Chinese and English. The model is motivated by the following observations: (1) whether an NE is translated semantically or phonetically depends greatly on its entity type, (2) entities within an aligned pair should share the same type, and (3) the initially detected NEs can act as anchors and provide further information while selecting NE candidates. Based on these observations, this article proposes a translation mode ratio feature (defined as the proportion of NE internal tokens that are semantically translated), enforces an entity type consistency constraint, and utilizes additional new NE likelihoods (based on the initially detected NE anchors).Experiments show that this novel method significantly outperforms the baseline. The typeinsensitive F-score of identified NE pairs increases from 78.4% to 88.0% (12.2% relative improvement) in our Chinese-English NE alignment task, and the type-sensitive F-score increases from 68.4% to 83.0% (21.3% relative improvement). Furthermore, the proposed model demonstrates its robustness when it is tested across different domains. Finally, when semi-supervised learning is conducted to train the adopted English NE recognition model, the proposed model also significantly boosts the English NE recognition type-sensitive F-score.

Towards Effective Entity Extraction of Scientific Documents using Discriminative Linguistic Features

“…NER also plays an important role in machine translation where it is used to automatically translate source texts or speech to a target language [21,22]. The misidentification of NEs not only affects the lexical recognition accuracy, but also the context as a whole, which then affects the translation quality.…”

Tibetan-Chinese Bilingual Sentences Alignment Method based on Multiple Features

“…For Chinese-English sentence alignment, there are some outstanding works, such as [7] and [12]. Wu et al proposed Chinese-Tibetan sentence alignment method based on bilingual dictionary [10,15].…”

Cross-Lingual Annotation Projection for Weakly-Supervised Relation Extraction

“…Recently, annotation projection applications, such as dependency parsing [Hwa et al 2005], named entity recognition [Fu et al 2011;Zitouni and Florian 2008], and semantic role labeling [Pado and Lapata 2009], have been studied. Some researchers have proposed utilizing parallel corpora for bilingual named-entity recognition [Chen et al 2010;Li et al 2012]. To the best of our knowledge, our work is the first to introduce a cross-lingual annotation projection approach to the relation extraction task.…”