Semantic Labeling of Chinese Special Sentence Patterns Based on Automation Technology

2013 ◽  
Vol 427-429 ◽  
pp. 1649-1652
Author(s):  
Bo Chen ◽  
Chen Lv ◽  
Dong Hong Ji

Parsing Chinese is a key issue in NLP. Many controversies arise from Chinese special sentence patterns. This paper puts forward a novel model Feature Structure theory to resolve the semantic labeling of Chinese special sentence patterns. We analyze the difficulties in annotating these sentences, and compare Feature Structure with dependency structure. Feature Structure represents more semantic information and more semantic relations. Feature Graph is a recursive undirected graph, allows nesting and multiple correlations.

2005 ◽  
Vol 04 (02) ◽  
pp. 133-138
Author(s):  
D. Manjula ◽  
T. V. Geetha

The traditional Boolean word-based approach to information retrieval (IR) considers only words for indexing. Irrelevant information is retrieved because of non-inclusion of semantic information like word senses and word context. In this work, the importance of representing the documents along another semantic dimension in addition to sense context information is considered. The incorporation of semantic relations as an additional dimension gives a better insight into the interpretation of the document. The micro-contexts generated from the documents are also used in indexing. The retrieval performance is measured in terms of precision and recall. The results tabulated show better performance.


2015 ◽  
Vol 2015 ◽  
pp. 1-9 ◽  
Author(s):  
Lin Guo ◽  
Wanli Zuo ◽  
Tao Peng ◽  
Lin Yue

The diversities of large-scale semistructured data make the extraction of implicit semantic information have enormous difficulties. This paper proposes an automatic and unsupervised method of text categorization, in which tree-shape structures are used to represent semantic knowledge and to explore implicit information by mining hidden structures without cumbersome lexical analysis. Mining implicit frequent structures in trees can discover both direct and indirect semantic relations, which largely enhances the accuracy of matching and classifying texts. The experimental results show that the proposed algorithm remarkably reduces the time and effort spent in training and classifying, which outperforms established competitors in correctness and effectiveness.


2016 ◽  
Vol 7 (1) ◽  
pp. 3-34 ◽  
Author(s):  
Valentina Piunno

Abstract This paper focuses on a specific type of Multiword Expressions, particularly widespread in Italian as well as in other Romance languages: Multiword Modifiers, i.e. prepositional phrases functioning as modifiers of a noun (Multiword Adjectives) and of a verb (Multiword Adverbs). Exploiting both syntactic and semantic analysis, this paper explores the hypothesis that Multiword Modifiers are formed on the basis of regular syntactic templates, which can structure and organize the semantic information associated with words. In this perspective, after a brief presentation of Multiword Lexical Units and the class of Multiword Modifiers, the methodology and the general theoretical framework of this study will be explained. The last section is devoted to the analysis of some semantic relations frequently fulfilled by Multiword Modifiers of Italian, French and Spanish. This investigation aims at demonstrating that all Romance languages considered make a regular use of this kind of analytical resource in adjectival or adverbial function, showing similar patterns and syntactic templates.


2018 ◽  
Vol 18 (1) ◽  
pp. 152-170
Author(s):  
Arkadiusz Janz ◽  
Paweł Kędzia ◽  
Maciej Piasecki

Abstract This paper presents a supervised approach to the recognition of Cross-document Structure Theory (CST) relations in Polish texts. Its core is a graph-based representation constructed for sentences. Graphs are built on the basis of lexicalised syntactic-semantic relations extracted from text. Similarity between sentences is calculated as similarity between their graphs, and the values are used as features to train the classifiers. Several different configurations of graphs, as well as graph similarity methods were analysed for this task. The approach was evaluated on a large open corpus annotated manually with 17 types of selected CST relations. The configuration of experiments was similar to those known from SEMEVAL and we obtained very promising results.


Author(s):  
DongLai Ge ◽  
Junhui Li ◽  
Muhua Zhu ◽  
Shoushan Li

Sequence-to-sequence (seq2seq) approaches formalize Abstract Meaning Representation (AMR) parsing as a translation task from a source sentence to a target AMR graph. However, previous studies generally model a source sentence as a word sequence but ignore the inherent syntactic and semantic information in the sentence. In this paper, we propose two effective approaches to explicitly modeling source syntax and semantics into neural seq2seq AMR parsing. The first approach linearizes source syntactic and semantic structure into a mixed sequence of words, syntactic labels, and semantic labels, while in the second approach we propose a syntactic and semantic structure-aware encoding scheme through a self-attentive model to explicitly capture syntactic and semantic relations between words. Experimental results on an English benchmark dataset show that our two approaches achieve significant improvement of 3.1% and 3.4% F1 scores over a strong seq2seq baseline.


Author(s):  
О.С. Филичева

Постановка задачи. В статье рассматриваются механизмы порождения и понимания текста с опорой на положения теории риторической структуры (ТРС), разработанной У. Манном и С. Томпсон. Данная теория основана на идее структурно-функционального направления в лингвистике о том, что связь между формой и содержанием языковой единицы объясняется структурой зависимостей. Зависимость одной единицы языка от другой в новообразованном единстве трактуется как функция, которую выполняет языковая единица в речи, или дискурсе. ТРС представлена как модель анализа семантических отношений между ядерными (главными) и сателлитовыми (зависимыми) дискурсивными единицами, образующими когерентный текст. Результаты. Описаны разные подходы к определению дискурса и его структурных компонентов. Проанализирована ядерно-сателлитовая модель структуры дискурсивных единиц, предложенная У. Манном и С. Томпсон. Авторы ТРС подчеркивают, что в основе когерентности текста лежат риторические отношения, которые связывают единицы дискурса и способствуют адекватному кодированию и декодированию информации. Это объясняется тем, что ядерно-сателлитовая организация дискурсивных единиц распределяет фон и фокус смыслового восприятия текста адресатом. Выявлены правила графической репрезентации риторических отношений, а также принципы классификации асимметричных и симметричных, моноядерных и многоядерных риторических отношений. Выводы. Ядерность и иерархичность признаются ведущими принципами структурной организации дискурса в ТРС. Иерархичность прослеживается не только в смысловой и интенциональной выделенности ядра по отношению к сателлиту, но и в построении риторических отношений между дискурсивными единицами. Таким образом, ядерно-сателлитовая модель дискурса является эффективным функциональным способом производства и интерпретации текстов, применяемым для решения различных коммуникативных задач. Statement of the problem. The article gives an explanation of the text generation and interpretation according to the Rhetorical structure theory (RST), developed by W. Mann and S. Thompson. This theory is based on structural and functional approaches in linguistics, which present the relations between the form and the content of a language unit as the structure of dependencies. The dependency of one language unit on another in the newly formed unity is understood as a function performed by the language unit in speech or discourse. RST provides as a model for analyzing semantic relations between nucleus (main) and satellite (dependent) discursive units that form the coherence of the text. Results. Different approaches to defining discourse and its structural components are described. The nucleus-satellite model of the discourse structure proposed by W. Mann and S. Thompson is analyzed. The authors of RST emphasize that the coherence of the text is based on rhetorical relations that connect the units of discourse and contribute to the adequate coding and decoding of information. The nucleus-satellite organization of discursive units distributes the addressee`s focus of text perception. The principles of graphic representation of rhetorical relations are revealed, as well as the principles of classification of asymmetrical and symmetrical, mononucleus and multinucleus rhetorical relations. Conclusion. Nuclearity and hierarchy are recognized as the leading ways of discourse structural organization in RST. Hierarchy can be traced not only in the semantic and intentional allocation of the nucleus, but also in the structural organization of rhetorical relations among discursive units. Thus, a nucleus-satellite model of discourse serves as an effective functional principle to generate and interpret texts, which used to solve various communication tasks.


2021 ◽  
Vol 21 (1) ◽  
pp. 103-118
Author(s):  
Qusai Y. Shambour ◽  
Nidal M. Turab ◽  
Omar Y. Adwan

Abstract Electronic commerce has been growing gradually over the last decade as a new driver of the retail industry. In fact, the growth of e-Commerce has caused a significant rise in the number of choices of products and services offered on the Internet. This is where recommender systems come into play by providing meaningful recommendations to consumers based on their needs and interests effectively. However, recommender systems are still vulnerable to the scenarios of sparse rating data and cold start users and items. To develop an effective e-Commerce recommender system that addresses these limitations, we propose a Trust-Semantic enhanced Multi-Criteria CF (TSeMCCF) approach that exploits the trust relations and multi-criteria ratings of users, and the semantic relations of items within the CF framework to achieve effective results when sufficient rating data are not available. The experimental results have shown that the proposed approach outperforms other benchmark recommendation approaches with regard to recommendation accuracy and coverage.


Sign in / Sign up

Export Citation Format

Share Document