Bengali to English Word Alignment Dataset

The dataset is in XML format and contains manually annotated 2000 Bengali and English parallelly aligned sentences. These parallel sentences were collected from different news articles and encyclopedias. The translation of some of the sentences was improved via Google Translator, and all the punctuation marks from parallel sentences were removed for the tokenization issues. A sample representation of the dataset is given below: Bengali: বাংলাদেশের জলবায়ু তাপমাত্রায় মৃদু English: Climate of Bangladesh is mild in temperature Alignments: 0-1 0-2 1-0 2-5 2-6 3-3 3-4

本数据集采用XML格式，收录经人工标注的2000组孟加拉语-英语平行对齐句对。该平行句对集采集自多类新闻文章与百科全书。部分句对的译文通过谷歌翻译（Google Translator）进行了优化；由于分词（tokenization）处理需求，所有平行句对中的标点符号均已被移除。以下为该数据集的样例展示：孟加拉语：বাংলাদেশের জলবায়ু তাপমাত্রায় মৃদু 英语：孟加拉国的气候气温温和 对齐标注：0-1 0-2 1-0 2-5 2-6 3-3 3-4