Bibrecord-Based Literature Management with?Interactive Latent Space Learning
Bibrecord-Based Literature Management with?Interactive Latent Space Learning
复制标题
基于书目记录的文献管理与交互式潜在空间学习
DOI：
10.1007/978-3-031-21756-2_13
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Morishima Atsuyuki
中科院分区：
文献类型：
--
作者：
Watanabe Shingo;Ito Hiroyoshi;Matsubara Masaki;Morishima Atsuyuki
每个研究人员都必须进行文献回顾，而从事不同研究主题的研究人员的文件管理需求各不相同。然而，今天有两个重大挑战。首先，传统的方法，如文档文件夹的树形层次结构和基于标签的管理，在巨大的出版物数量面前不再有效。其次，尽管他们的收银台信息对每个人都是可用的，但许多报纸只能通过付费服务才能访问。本研究试图开发一种仅基于书目记录的交互式个人文献管理工具。为了使这种工具成为可能，我们开发了一种原则性的“人在环潜在空间学习”方法，该方法根据每个研究人员的反馈来估计他或她的管理标准，以计算文档在屏幕上的二维空间中的位置。由于一组书目记录形成一个图形，因此我们的模型自然被设计为一个连接图形和空间的基于图形的编解码器模型。对来自人文、科学和工程领域的10名研究人员的实验表明，所提出的框架比典型的图卷积编解码模型具有更好的结果。
Every researcher must conduct a literature review, and the document management needs of researchers working on various research topics vary. However, there are two significant challenges today. First, traditional methods like the tree hierarchy of document folders and tag-based management are no longer effective with the enormous volume of publications. Second, although their bib information is available to everyone, many papers can be accessed only through paid services. This study attempts to develop an interactive tool for personal literature management solely based on their bibliographic records. To make such a tool possible, we developed a principled “human-in-the-loop latent space learning” method that estimates the management criteria of each researcher based on his or her feedback to calculate the positions of documents in a two-dimensional space on the screen. Since a set of bibliographic records forms a graph, our model is naturally designed as a graph-based encoder-decoder model that connects the graph and the space. The experiments with ten researchers from humanities, science, and engineering domains show that the proposed framework gives much superior results to a typical graph convolutional encoder-decoder model.