English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Neural machine translation (NMT), a new approach to machine translation, hasachieved promising results comparable to those of traditional approaches suchas statistical machine translation (SMT). Despite its recent success, NMTcannot handle a larger vocabulary because training complexity and decodingcomplexity proportionally increase with the number of target words. Thisproblem becomes even more serious when translating patent documents, whichcontain many technical terms that are observed infrequently. In NMTs, wordsthat are out of vocabulary are represented by a single unknown token. In thispaper, we propose a method that enables NMT to translate patent sentencescomprising a large vocabulary of technical terms. We train an NMT system onbilingual data wherein technical terms are replaced with technical term tokens;this allows it to translate most of the source sentences except technicalterms. Further, we use it as a decoder to translate source sentences withtechnical term tokens and replace the tokens with technical term translationsusing SMT. We also use it to rerank the 1,000-best SMT translations on thebasis of the average of the SMT score and that of the NMT rescoring of thetranslated sentences with technical term tokens. Our experiments onJapanese-Chinese patent sentences show that the proposed NMT system achieves asubstantial improvement of up to 3.1 BLEU points and 2.3 RIBES points overtraditional SMT systems and an improvement of approximately 0.6 BLEU points and0.8 RIBES points over an equivalent NMT system without our proposed technique.

Translation of Patent Sentences with a Large Vocabulary of Technical Terms Using Neural Machine Translation