ABSTRACT Community question‐answering (CQA) enables responders to select questions, and respond to the questions by answering, commenting or voting. Accordingly, questions with different cues (i.e. complexity, specificity, emotional expressiveness, politeness, popularity, rewards) tend to attract different responders. However, the research is limited regarding the types of responders based on the questions they responded to. The gap inhibits us to form a complete understanding of how questions bridge askers with responders. Moreover, how different types of responders contribute to maintaining the ecosystem of the CQA has not been studied adequately. Accordingly, we conducted an online survey to organize responders by the cues of questions. Cluster analysis was used to group responders into three types: (1) “leaders” respond to complex and popular questions, attracting many followers in CQA; (2) “socializers” answer less complex and specific questions with emotion‐laden words; and (3) “specialists” respond to complex questions with high specificity but seldom use the social functions of CQAs. Finally, contributions and limitations are discussed.
ABSTRACTThe purpose of this study is to explore connections between immigrant information behaviour and concepts of information poverty. The researchers used the scoping method to examine the immigrant populations, operational definitions, and research methods associated with information poverty. Key findings from this systematic review suggest an overreliance on non‐empirical linkages to information poverty and the need for substantiated knowledge that captures specific immigrant contexts, vulnerabilities, and interventions.
This study examined Israeli researchers’ trends, intentions and awareness regarding scientific publications in open access (OA) journals and repositories. A survey was distributed to 202 Israeli researchers. According to the findings, most researchers used a combination of closed (traditional/toll access) journals with OA gold/green routes. Researchers showed low awareness of OA key concepts, which was significantly associated with the tendency not to publish in OA. In addition, researchers who reported that there was a formal policy concerning publishing articles in high-impact journals in their department tended to publish less in OA. The most common reasons that caused researchers not to publish in OA routes were lack of funding and lack of awareness. Conclusions present a need for an OA's declared policy as well as financial and informational support.
ABSTRACTCollection development policies for university special collections and archives guide archival appraisal, which is the process by which archivists determine the value of an item. Much has been written about appraisal theory, but little is known about archivists' actual appraisal practice and how collection policy guides such practice. The object of this paper is to explore the appraisal practices of archivists by examining collection development policies. As a preliminary examination, this study analyzed the collection development policies of seven Texas tier‐one institutions. The most common component of the policies was a statement on the types of items collected or excluded. The study also found that the language used for these formats varied considerably with no standardization. The policies themselves also varied widely in detail.
ABSTRACTIn the new era, the analysis of academic journal evaluation methods and the comprehensive comparison of the advantages, disadvantages, and stability of different methods can help to provide some reference for academic journal evaluation. In this study, single model evaluation was carried out for 6 weighting methods without comprehensive evaluation value, and fuzzy comprehensive evaluation is performed on the results passing the nonparametric test. Based on the evaluation, BP neural network is introduced, and BP neural network evaluation model is established. The results show that the fuzzy Borda evaluation can integrate the evaluation value and evaluation order of single models, and has higher accuracy compared with single evaluation models. The prediction rate of the network model based on the gradient descent optimization algorithm can reach more than 80%, and the weights obtained from the continuous self‐learning of the neural network training set can reduce the subjectivity and mutual interference between indicators.
