Grammar teaching is an indispensable part of English Language Teaching (ELT) in an EFL context like China. Compared with the classroom teaching innova tion in listening, speaking, reading and writing, grammar teaching...Grammar teaching is an indispensable part of English Language Teaching (ELT) in an EFL context like China. Compared with the classroom teaching innova tion in listening, speaking, reading and writing, grammar teaching has often bee n forgotten. The newly published Senior High School English Teaching Curriculum Standard points out that the general aim of the English curriculum is to enable students to clarify the aim of English learning, to develop ability in autonomou s learning and cooperative learning, to establish efficient learning strategy, a nd to develop integrated skills in using the language. Knowledge, skills, affect , learning strategies, and culture awareness have attracted more attention in te aching and learning. The publication of this new standard leads the ELT in China to a new direction. Correspondingly, grammar teaching should not just maintain its rule-listing routine. Instead, some new attempts should be involved. Gramma r teaching through tasks in situational contexts, which pursues the appropriate p ractical use of grammar, has been tested in the writer’s class. This paper is m a inly about the author’s understanding of grammar teaching and her real classroo m practice. A small-scale survey was done with the author’s colleagues in the s am e school to investigate their attitudes and opinions. In general, teachers show an affirmative attitude. Meanwhile they have some worries as well since good att ainments in exams are expected from both students and parents. However, the auth or has a strong belief that teaching grammar through situational tasks cannot on ly benefit students’ linguistic competence, communicative competence and langua ge proficiency, but also students’ achievements in exams.展开更多
利用电子病历数据进行疾病预测是时下的研究热点,医学事件是电子病历的重要组成部分。由于电子病历数据具有异质、高维的特性,且对时间的依赖性比较强,获得良好的医学事件表示存在一定困难,因此文章提出一种基于形变-时控长短期记忆网络...利用电子病历数据进行疾病预测是时下的研究热点,医学事件是电子病历的重要组成部分。由于电子病历数据具有异质、高维的特性,且对时间的依赖性比较强,获得良好的医学事件表示存在一定困难,因此文章提出一种基于形变-时控长短期记忆网络(Mogrifier-Time Long Short Term Memory Network, MT-LSTM)的医学事件表示学习方法。通过在MIMIC-Ⅲ数据集上进行多项对比实验,结果表明MT-LSTM模型可获得优良的医学事件表示,有助于疾病预测任务的进行,证明了该方法的有效性。展开更多
文摘Grammar teaching is an indispensable part of English Language Teaching (ELT) in an EFL context like China. Compared with the classroom teaching innova tion in listening, speaking, reading and writing, grammar teaching has often bee n forgotten. The newly published Senior High School English Teaching Curriculum Standard points out that the general aim of the English curriculum is to enable students to clarify the aim of English learning, to develop ability in autonomou s learning and cooperative learning, to establish efficient learning strategy, a nd to develop integrated skills in using the language. Knowledge, skills, affect , learning strategies, and culture awareness have attracted more attention in te aching and learning. The publication of this new standard leads the ELT in China to a new direction. Correspondingly, grammar teaching should not just maintain its rule-listing routine. Instead, some new attempts should be involved. Gramma r teaching through tasks in situational contexts, which pursues the appropriate p ractical use of grammar, has been tested in the writer’s class. This paper is m a inly about the author’s understanding of grammar teaching and her real classroo m practice. A small-scale survey was done with the author’s colleagues in the s am e school to investigate their attitudes and opinions. In general, teachers show an affirmative attitude. Meanwhile they have some worries as well since good att ainments in exams are expected from both students and parents. However, the auth or has a strong belief that teaching grammar through situational tasks cannot on ly benefit students’ linguistic competence, communicative competence and langua ge proficiency, but also students’ achievements in exams.
文摘利用电子病历数据进行疾病预测是时下的研究热点,医学事件是电子病历的重要组成部分。由于电子病历数据具有异质、高维的特性,且对时间的依赖性比较强,获得良好的医学事件表示存在一定困难,因此文章提出一种基于形变-时控长短期记忆网络(Mogrifier-Time Long Short Term Memory Network, MT-LSTM)的医学事件表示学习方法。通过在MIMIC-Ⅲ数据集上进行多项对比实验,结果表明MT-LSTM模型可获得优良的医学事件表示,有助于疾病预测任务的进行,证明了该方法的有效性。
