Abstract

The paper addresses the issue of filling the gaps in the semantic library based on the distributional semantics of the terms of its thesaurus and the ontology relations. The goal of the study is to fully reflect the actual structure of relations between mathematical subject domains. This is done through identifying context-sensitive semantic relations, and with the use of an algorithm that is based on the word2vec feedforward neural networks. The understanding of the query is analyzed after preliminary processing of set of articles and metadata saturation. The proposed procedure helps to improve the work with the full-text index and, as a result, improves the quality of search in the library. Using a full-text index of a digital semantic library as an example, we demonstrate the process of filling gaps by saturating the semantic relations of the ontology of mathematical subject domains.

中文翻译：

---

使用应用本体来饱和语义关系

摘要

本文从词库词条的分布语义和本体关系的角度来解决语义库的空白问题。研究的目标是充分反映数学学科领域之间关系的实际结构。这是通过识别上下文敏感的语义关系，并使用基于word2vec的算法来完成的前馈神经网络。在对文章集和元数据饱和进行初步处理后分析查询的理解。建议的程序有助于改进全文索引的工作，从而提高图书馆的搜索质量。以一个数字语义库的全文索引为例，我们展示了通过饱和数学学科领域本体的语义关系来填补空白的过程。