中文电子病历的分词及实体识别研究

[目的/意义]健康医疗大数据是我国重要的基础性战略资源，本研究对中文电子病历分词与实体识别的探讨与实证较好地完成了医疗数据的信息抽取任务，对今后医疗大数据在语义层面的应用发展具有重要意义。[方法/过程]本研究首先融合权威词表、官方标准、健康网站数据及其他医学补充词库构建了词语数量级达到10万的医学词表；然后对电子病历的字段进行分词，对比了jieba工具、导入词典后的jieba、无监督学习及AC自动机4种模型的分词效果；最后，以自动分词和人工标注结果为语料，实现基于条件随机场的电子病历实体识别研究，并比较不同实体类别以及不同文本特征下的实体识别效果，选出最优模板。[结果/结论]分词结果显示，AC自动机的效果最好，F值可达82%；实体识别结果表明，"检查"和"疾病"实体的识别效果最好，而"症状"的识别效果不太理想。     

时间词“明日”“明天”的多维度考察

在共时层面上,时间词"明日""明天"在句法组合以及语用表达方面都有所不同,因此,不能认定两者是完全等同的;从历时层面看,两者也呈现出一个此消彼长的演变过程,同时这种状况也在方言调查中得到证实。     

Exploring mechanisms of rapid automatized naming to arithmetic skills in Chinese primary schoolers

Rapid automatized naming (RAN) is associated with children's arithmetic performance, which is multifactorial by nature and consists of arithmetic fluency, arithmetic procedure, and word problems. Yet, it is unclear whether RAN can predict all key aspects of arithmetic skills independently and what cognitive mechanisms may explain this relation. Using a sample of Chinese primary school students (N = 170), results showed that while RAN was associated with all key aspects of arithmetic skills, the relations of RAN–arithmetic procedure and RAN–word problems may be dependent on that of RAN–arithmetic fluency. Supporting the processing speed hypothesis, results further suggested that processing speed may underlie the relation between RAN and arithmetic performance. Findings made unique contribution to understanding the effect of RAN on arithmetic performance and cognitive mechanisms that may underpin this association in the context of Chinese culture.     

How can children with mild literacy difficulties be supported at the transition to secondary school? A small‐scale quasi‐experimental study

There is evidence that pupils with weak literacy skills struggle on transition to secondary school. Many experience a drop in attainment in the summer break between the two. A British government‐funded programme of rigorously designed research on boosting literacy at transition had (by 2015) found only four of 15 interventions evaluated had positive effects. This small‐scale quasi‐experimental study investigated the effectiveness of support for pupils with mild literacy difficulties on transition to secondary school. Thirty‐two pupils in three schools were involved; half received the programme. Pairs were matched on reading, spelling, age and gender. Intervention was designed around the individual needs of each pupil, focusing variously on language skills, writing, reading and spelling. The group receiving the programme made modest gains in spelling, reading efficiency and single word reading. The comparison group lost ground, relatively, in all three areas. The results suggest a promising line for more rigorous investigation.     

普通话教学中朗读环节易见的口语语感问题及解决方法

在普通话教学中,朗读是学生最容易出现口语语感问题的环节,最常见的口语语感问题是停连不当、重音不准确、节奏把握不当。学生无法做到完全的情景再现,对内在语把握不准,情感运动不到位和过于关注细节是造成上述问题的主要原因。可以通过体验式感受、情景式教学、表演式表达以及暗示教学法帮助学生解决朗读中的口语语感问题。     

《应县木塔辽代秘藏》与《龙龛手镜》俗字比较研究

将《应县木塔辽代秘藏》残卷与《龙龛手镜》相印证,纠正了《龙龛手镜》中一部分俗字的错误,并补充了《龙龛手镜》未收的一些俗字。     

金山词霸在科技英语翻译中的应用技巧

金山词霸是科技英语翻译的主要工具之一，被广大科技英语翻译工作者广泛应用，本文从屏幕取词、词典查询等方面介绍了金山词霸在英语翻译中的应用技巧，以使翻译更直接、方便、快捷。     

揣摩语境选取语义

英语词语的意义由其所处的语境来决定 ,尤其是不同的学科有其独特的意义。译者在翻译时须予以重视 ,认真揣摩原文的语境 ,正确选取语义。同时 ,译者对所译学科的术语有所掌握或了解无疑有助于译文的通畅达意。此外 ,译者不可忽视历史文化背景决定语义     

吴闽语进行体和持续体的语法化序列分析

莆仙方言的进行体和持续体标记来源于处所词,这是吴语和闽语的共同特征.但莆仙方言的这两种体标记来源于处所代词"即、许",而不是方位后缀或动介词,这又不同于吴语与闽语的其他方言.本文先描写莆仙方言的进行体和持续体特征,然后结合吴闽语其他八种方言的资料,分析进行体和持续体标记的语法化序列,从而观察其演变发展的过程.     

黄冈方言的后加成分"着"

在黄冈方言中,“着”可附在时间词或动词短语之后,构成具有动词性的“着”字结构,相当于“等到……以后再……”或“等到……再说”。成都方言“哆”字结构与黄冈方言“着”字结构性质相同,二之间可能存在一定的渊源关系。“哆”、“着”都不是“再说”衍生出来的;关于“着”的来源,杨永龙（2002）所说较为可信。