@inproceedings{2fcea982c94e484c8ed5a4371a4a99b4,
title = "Training set similarity based parameter selection for statistical machine translation",
abstract = "Log-linear model based statistical machine translation systems (SMT) are usually composed of multiple feature functions. Each feature function is assigned a weight as a model parameter. In this paper, we consider that different input source sentences may have discrepant needs for model parameters. To adapt the model to different inputs, we propose a model parameters selection method for log-linear model based SMT systems. The method is mainly based on the characteristics of different feature functions themselves without any assumption on unseen test sets. Experimental results on two language pairs (Zh-En and Ug-Zh) show that our method leads to the improvements up to 2.4 and 2.2 BLEU score respectively, and it also shows the good interpretability of our proposed method.",
keywords = "Log-linear model, Parameter selection, Statistical machine translation",
author = "Xuewen Shi and Heyan Huang and Ping Jian and Tang, {Yi Kun}",
note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG, part of Springer Nature 2018.; 2nd Asia Pacific Web and Web-Age Information Management Joint Conference on Web and Big Data, APWeb-WAIM 2018 ; Conference date: 23-07-2018 Through 25-07-2018",
year = "2018",
doi = "10.1007/978-3-319-96890-2_6",
language = "English",
isbn = "9783319968896",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "63--71",
editor = "Jianliang Xu and Yoshiharu Ishikawa and Yi Cai",
booktitle = "Web and Big Data - Second International Joint Conference, APWeb-WAIM 2018, Proceedings",
address = "Germany",
}