这本书探讨了设计面向语义的自然语言（NL）处理系统的关键问题。其中一个关键特征是将现有的万维网转化为新一代语义网（SW-2）的原始策略，以及实现该策略的基本形式化工具。拟建SW-2的主要特点是NL处理能力良好。
通过引入描述概念结构上10个相互关联的部分运算的数学模型，获得了描述自然语言文本（句子和任意复杂语篇）结构意义的广义概念框架。定义了一类新的形式语言，称为标准知识语言（SK语言）。读者将获得这些语言的知识，并学习使用它们构建语义表示的方法。
此外，还构建了一个广泛适用的语言数据库数学模型。利用原始形式概念描述了一种实用的、结构化强的自然语言文本语义句法分析算法；输入文本可以是英语、俄语和德语的句子。
通过广泛使用示例和插图来澄清复杂的材料并演示实际应用、许多历史和书目注释、章末练习和词汇表，本书可以作为研究生级教材，也可以作为研究人员和实践者处理涉及自然语言文本语义的各种问题的良好参考，本体论、语义Web、电子科学中的语义数据集成，以及多代理系统中的内容语言，尤其是在电子商务和电子健康领域。
Semantics-Oriented Natural Language Processing: Mathematical Models and Algorithms
This book examines key issues in designing semantics-oriented natural language (NL) processing systems. One of the key features is an original strategy for transforming the existing World Wide Web into a new generation Semantic Web (SW-2) and the basic formal tools for its realization, which are proposed. The principal distinguishing feature of the proposed SW-2 is the well-developed ability of NL processing.
A broad conceptual framework for describing structured meanings of NL-texts (sentences and arbitrarily complex discourses) is obtained by introducing a mathematical model describing 10 interrelated partial operations on conceptual structures. A new class of formal languages called standard knowledge languages (SK-languages) is defined. Readers will gain knowledge of these languages and learn a way of building semantic representations using them.
Additionally, a broadly applicable mathematical model of a linguistic database is constructed. A useful for practice and strongly structured multi-lingual algorithm of semantic-syntactic analysis of NL-texts is described by means of original formal concepts; the input texts can be sentences in English, Russian, and German.
With extensive use of examples and illustrations to clarify complex material and demonstrate practical applications, many historical and bibliographical notes, end-of-chapter exercises, and glossaries, this book can serve as a graduate-level textbook, as well as a good reference for researchers and practitioners who deal with the various problems involving semantics of natural language texts, ontologies, Semantic Web, semantic data integration in e-science, and content languages in multi-agent systems, in particular, in e-commerce and e-health.
OR
