[Research purpose] The Theory of Motivated Information Management（TMIM） proposes a process through which uncertainty motivates information management. To examine the applicability of the theory and extend theoretical understanding of the framework, this study meta-analyzed associations among different TMIM variables in existing empirical TMIM studies and tested potential moderating effects.[Research method] Through a systematic literature search, the current study meta-analyzed 44 studies from 40 articles that empirically tested predictions from the TMIM. Moderation analyses were conducted using meta-regression. [Research conclusion] Overall, bivariate associations proposed by the TMIM were supported, although these associations were heterogeneous and moderated by the average age of the sample, issue importance, and study design. The findings of this study have important implications for future TMIM research, and provide a theoretical framework for uncertainty and information management research.
[Research purpose] Investigating the information access and its relevance judgment during essay writing task is beneficial for better understanding user information behavior and further promoting the research in this field. [Research method] This paper adopts diary method to collect users’ self-recorded diary data about their relevance judgment at different stages of authentic work task process given a course essay assignment. It includes predictive and evaluative judgments, as well as the application of judgment criteria and document information elements. These data are analyzed using both qualitative and quotative methods for statistical significance testing. [Research conclusion] The study finds that there is significant correlation between predictive and evaluative judgments, and the former is significantly affected by the work task stage. In addition, judgment criteria and document elements as well as their combination patterns are diverse in predictive judgment. The type and number of judgment criteria and document elements applied are significantly different at different work-task stages. This research can provide inspiration for user information behavior theory and practice of information organization and retrieval system in the context of course essay writing.
[研究目的]明确国家安全情报产品的规范标准,制定科学、合理、可操作的评价体系,能够审视并发现国家安全情报工作中的不足和缺陷,并根据实际情况对情报产品的规范标准做出调整,更好地服务于国家安全.[研究方法]美国是当前国际上为数不多的,出台了国家安全情报产品规范性指导文件的国家,在归纳整理后形成评价指标.以层次分析法、模糊综合评价法作为评价方法,以产品层次理论为评价理论,尝试构建出具有一般意义的,能够广泛使用的国家安全情报产品规范的评价体系.[研究结论]经过评价体系的构建和实证分析后发现,美国国家安全情报产品在信息源披露以及用户反馈等方面表现不佳,启示我们在相关立法中应加强国家安全情报产品规范的实施细则并建立产品评估的反馈机制.
[研究目的]现代信息技术正在为知识服务发展开辟新赛道,元宇宙与隐性知识的完美契合,使其成为技术赋能隐性知识服务的最佳模式.[研究方法]首先分析隐性知识服务元宇宙构建须遵循的原则,然后融合复叠知识场、服务场和交互场构建服务元宇宙概念模型.该模型集资源空间、服务空间、技术空间和管理空间为一体,能够为用户提供具身体验、虚拟实验、真人图书等多维度、多感官的隐性知识服务.最后,从知识需求意识、元宇宙场景和基于伪体验的认知建构阐释隐性知识服务元宇宙的相关机理.[研究结论]隐性知识服务元宇宙的构建方案能够促使组织构建起更加完备、智能的隐性知识服务网络,提高组织隐性知识服务效能.
[研究目的]舆情情感分析作为开源情报的重要方面,以少样本学习的形式进行能极大提高情报获取效率.针对在低资源场景下信息不足而导致的情感检测模型性能不佳问题,提出一种融合大语言模型知识的对比提示情感分析方法.[研究方法]首先利用大语言模型的通用知识实现训练样本的扩充,然后使用无标注数据进行对比嵌入训练以优化文本编码器的语义表征能力,最后利用提示学习机制,在无标注数据上进行自预测的迭代扩充训练以使模型进一步使适应特定任务.[研究结论]在公开数据集上的实验结果表明,提出的模型在同等标注数据量的条件下表现优于基线方法.
