基于Usage Metrics的中文学术论文用户平台偏好和兴趣偏好比较研究

研究中文学术论文用户使用模式有助于分析使用规律,为应用决策提供依据。本文以八个学科被CSSCI或CSCD收录的61本开放获取期刊发表于2014—2015年的学术论文为样本,以期刊官网和信息集成平台上的使用数据为来源,采用Usage Metrics的方法,从用户平台偏好和用户兴趣偏好两方面比较中文学术论文的用户使用模式。研究发现:①期刊官网与信息集成平台的用户平台偏好存在差异。一是从学科视角看,用户整体上倾向于使用期刊官网而不是信息集成平台获取所需论文,而社会科学的信息集成平台篇均下载次数均大于自然科学。二是从期刊视角看,社会科学用户比自然科学用户更倾向于使用信息集成平台获取所需论文;与信息集成平台相比,期刊官网的用户下载数据呈现更为明显的“两级分化”现象;期刊官网下载次数与信息集成平台下载次数整体上的相关性较低。②期刊官网与信息集成平台的用户兴趣偏好存在差异。每本期刊官网和信息集成平台下载次数前20%的学术论文的Jaccard相似系数较低;在所列举的图书馆、情报与文献学中,期刊官网与信息集成平台的用户关注主题不同。图7。表5。参考文献37。     

Health science libraries in Sweden: new directions,expanding roles

Librarians in Sweden are facing huge challenges in meeting the demands of their organisations and users. This article looks at four key areas: coping with open science/open access initiatives; increasing demands from researchers for support doing systematic reviews; understanding user experiences in Swedish health science libraries; and the consequences of expanding roles for recruitment and continuing professional development. With regard to changing roles, there is an increasing shift from the generalist towards the expert role. The authors raise the issue as to how to prepare those new to the profession to the changing environment of health science libraries.     

Trends of infodemiology studies: a scoping review

Introduction

The health care industry is rich in data and information. Web technologies, such as search engines and social media, have provided an opportunity for the management of user generated data in real time in the form of infodemiology studies. The aim of this study was to investigate infodemiology studies conducted during 2002–2016, and compare them based on developed, developing and in transition countries.

Methods

This scoping review was conducted in 2017 with the help of the PRISMA guidelines. PubMed, Scopus, Science Direct, Web of Knowledge, Google Scholar, Wiley and Springer databases were searched between the years 2002 and 2016. Finally, 56 articles were included in the review and analysed.

Results

The initial infodemiology studies pertain to the quality assessment of the hospital's websites. Most of the studies were on developed countries, based on flu, and published in the Journal of Medical Internet Research.

Conclusion

The infodemiology approach provides unmatched opportunities for the management of health data and information generated by the users. Using this potential will provide unique opportunities for the health information need assessment in real time by health librarians and thereby provide evidence based health information to the people.

     

Spatial Discovery of Linked Research Datasets and Documents at a Spatially Enabled Research Library

Current publishing practices in academia tend to result in datasets that are difficult to discover. This is because datasets are not well-integrated across academic domains and they are often not linked to the documents that reference them. For these reasons, discovering datasets across domains can be challenging; for example, discovering archeological observations and biological specimens using the same search is not widely supported, even if both datasets share a similar spatial extent, like Mesoamerica. It is also challenging to retrieve relevant documents that reference datasets; for example, retrieving a series of field reports that reference archeological observations is typically not supported. Our work develops an extensible method for: (1) geographically integrating collections across disciplinary repositories and (2) connecting datasets to related documents. We describe a collection of spatially-referenced researcher datasets, capturing their metadata elements and encoding them as linked open data. We then leverage existing library services to formalize links from datasets to documents. The system described in this work has been deployed, resulting in an experimental open data site for the UCSB campus. Results indicate that this system can be scaled-up with support from an institutional repository in the near future.     

An impulse to exploit: the behavioral turn in data-driven marketing

Industry advocates argue that by tailoring services and commercial solicitations to match media users’ personal interests, data-driven marketing benefits both consumers and businesses. This article shows, however, that advertisers and marketers who are taking up ideas and techniques from behavioral economics tell their clients a very different story about the aims and use of digital marketing and consumer surveillance. Listening in on this discourse demonstrates that some digital marketers conceptualize their own practices as forms of social control, appropriating concepts from behavioral economics to identify consumers’ cognitive and affective biases and target their vulnerabilities. Behavioral economics recognizes that economic decisions are not simply dictated by rational self-interest; rather, such choices depend on cognitive heuristics and habits, and can be manipulated through the design of “choice architecture.” This article discusses implications of the behavioral turn in data-driven marketing for critical advertising scholars, public advocates, and regulators.     

Multimodal data as a means to understand the learning experience

Most work in the design of learning technology uses click-streams as their primary data source for modelling & predicting learning behaviour. In this paper we set out to quantify what, if any, advantages do physiological sensing techniques provide for the design of learning technologies. We conducted a lab study with 251 game sessions and 17 users focusing on skill development (i.e., user's ability to master complex tasks). We collected click-stream data, as well as eye-tracking, electroencephalography (EEG), video, and wristband data during the experiment. Our analysis shows that traditional click-stream models achieve 39% error rate in predicting learning performance (and 18% when we perform feature selection), while for fused multimodal the error drops up to 6%. Our work highlights the limitations of standalone click-stream models, and quantifies the expected benefits of using a variety of multimodal data coming from physiological sensing. Our findings help shape the future of learning technology research by pointing out the substantial benefits of physiological sensing.     

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。     

情报学取向的“数据科学与大数据技术”专业人才培养

[目的/意义]情报学作为计算性较强的人文社科学科，具备数据科学的特点，大数据相关技术能为情报学提供有效手段，因此，设立情报学取向的"数据科学与大数据技术"专业，培养具有大数据思维和扎实的情报学专业基础、掌握大数据处理技术及分析理论、并能将其应用于情报学理论与实践的复合型人才，能够促进情报学学科发展。[方法/过程]对国内外大数据相关专业的培养目标和课程设置模式进行调研，总结学科教育发展特色，结合国内的教学资源情况，设计情报学取向的大数据专业人才培养计划。[结果/结论]情报学与大数据既有联系又有不同，情报学为大数据提供了基础研究方法，促进了多源数据融合，拓宽了大数据的应用领域；大数据为情报学研究提供了更多的可能性、丰富了研究方法，提升了研究效率、扩大了研究范围。情报学取向的大数据与数据科学专业人才的培养，既有其必要性，又有其不可替代性。学科体系需进一步改革创新，设立更多紧密结合大数据与情报学专业的新兴课程，如文献大数据分析、用户大数据研究等，系统地培养学生将情报学领域知识与计算机技术和大数据技术融合创新的能力。