scholarly journals Keep the Structure: A Latent Shift-Reduce Parser for Semantic Parsing

2021 ◽  
Author(s):  
Yuntao Li ◽  
Bei Chen ◽  
Qian Liu ◽  
Yan Gao ◽  
Jian-Guang Lou ◽  
...  
Keyword(s):  
Natural Language ◽  
Empirical Studies ◽  
Logical Form ◽  
Hierarchical Structures ◽  
Local Alignment ◽  
Semantic Parsing ◽  
Natural Languages ◽  
Different Types ◽  
End To End

Traditional end-to-end semantic parsing models treat a natural language utterance as a holonomic structure. However, hierarchical structures exist in natural languages, which also align with the hierarchical structures of logical forms. In this paper, we propose a latent shift-reduce parser, called LASP, which decomposes both natural language queries and logical form expressions according to their hierarchical structures and finds local alignment between them to enhance semantic parsing. LASP consists of a base parser and a shift-reduce splitter. The splitter dynamically separates an NL query into several spans. The base parser converts the relevant simple spans into logical forms, which are further combined to obtain the final logical form. We conducted empirical studies on two datasets across different domains and different types of logical forms. The results demonstrate that the proposed method significantly improves the performance of semantic parsing, especially on unseen scenarios.

Syntax

2018 ◽  
Author(s):  
Stephen Neale
Keyword(s):  
Twentieth Century ◽  
Natural Language ◽  
Logical Form ◽  
Natural Languages ◽  
Related Notion ◽  
Theories Of Mind ◽  
Contemporary Work ◽  
Linguistic Categories ◽  
Traditional Work ◽  
Syntactic Properties

Syntax (more loosely, ‘grammar’) is the study of the properties of expressions that distinguish them as members of different linguistic categories, and ‘well-formedness’, that is, the ways in which expressions belonging to these categories may be combined to form larger units. Typical syntactic categories include noun, verb and sentence. Syntactic properties have played an important role not only in the study of ‘natural’ languages (such as English or Urdu) but also in the study of logic and computation. For example, in symbolic logic, classes of well-formed formulas are specified without mentioning what formulas (or their parts) mean, or whether they are true or false; similarly, the operations of a computer can be fruitfully specified using only syntactic properties, a fact that has a bearing on the viability of computational theories of mind. The study of the syntax of natural language has taken on significance for philosophy in the twentieth century, partly because of the suspicion, voiced by Russell, Wittgenstein and the logical positivists, that philosophical problems often turned on misunderstandings of syntax (or the closely related notion of ‘logical form’). Moreover, an idea that has been fruitfully developed since the pioneering work of Frege is that a proper understanding of syntax offers an important basis for any understanding of semantics, since the meaning of a complex expression is compositional, that is, built up from the meanings of its parts as determined by syntax. In the mid-twentieth century, philosophical interest in the systematic study of the syntax of natural language was heightened by Noam Chomsky’s work on the nature of syntactic rules and on the innateness of mental structures specific to the acquisition (or growth) of grammatical knowledge. This work formalized traditional work on grammatical categories within an approach to the theory of computability, and also revived proposals of traditional philosophical rationalists that many twentieth-century empiricists had regarded as bankrupt. Chomskian theories of grammar have become the focus of most contemporary work on syntax.

SEMANTIC-PARSING BASED ON SEMANTIC UNITS THEORY — A NEW APPROACH TO NATURAL LANGUAGES PROCESSING

2008 ◽  
Vol 22 (07) ◽  
pp. 1447-1459 ◽  
Author(s):  
XIAOYU GAO ◽  
HU YUE ◽  
L. LI ◽  
QINGSHI GAO
Keyword(s):  
Natural Language ◽  
Semantic Parsing ◽  
Natural Languages ◽  
New Approach ◽  
Sentence Parsing ◽  
Syntax Parsing ◽  
Parsing Tree

The syntax of different natural languages are different, hence the parsing of different natural languages are also different, thus leadings to structures of their parsing-trees being different. The reason that the sentences in different natural languages can be translated to each other is that they have the same meaning. This paper discusses a new sentence parsing, called semantic-parsing, based on semantic units theory. It is a new theory where a sentence of a natural language is not regarded as of words and phrases arranged linearly; rather it is expected to consist of semantic units with or without type-parameters. This is a new parsing approach where the syntax-parsing-tree and semantic-parsing-tree are isomorphic. It is also a new approach in which the structure-trees of the sentences in all different natural languages can correspond.

Thesaurus for Natural-Language-Based Conceptual Design

2009 ◽  
Author(s):  
Eiko Yamamoto ◽  
Toshiharu Taura ◽  
Shota Ohashi ◽  
Masaki Yamamoto
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Engineering Design ◽  
Conceptual Design ◽  
Language Processing ◽  
Hierarchical Structures ◽  
Natural Languages ◽  
Mathematical Equations ◽  
Thinking Processes ◽  
Physical Phenomena

Conceptual design is a process wherein new functions are created through engineering design. In conceptual design, we use natural language since it plays an important role in the expression and operation of a function. Moreover, natural language is used in our day-to-day thinking processes and is expected to keep a fine interface with the designer. However, it is at a disadvantage with regard to the expression of a function, since physical phenomena, which are the essence of a function, are better expressed in the form of mathematical equations than natural languages. In this study, we attempt to develop a method for using natural language for operating a function by harnessing its advantages and overcoming its disadvantage. We focus on the vital process in conceptual design, that is, the function dividing process wherein the required function is decomposed into sub functions that satisfy the required function. We construct a thesaurus by semiautomatic extraction of the hierarchical structures of words from a document by using natural language processing. We show that the constructed thesaurus can be useful in supporting the function dividing process.

scholarly journals Automation of solving planimetry problems written in Ukrainian

2020 ◽  
pp. 071-080
Author(s):  
O.P. Zhezherun ◽  
◽  
O.R. Smysh ◽  
◽  
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Text Processing ◽  
Comprehensive Analysis ◽  
Text Representation ◽  
Natural Languages ◽  
Different Types ◽  
Mathematical Problems ◽  
Further Development

The article focuses on developing a software solution for solving planimetry problems that are written in Ukrainian. We discuss tendencies and available abilities in Ukrainian natural language processing. Presenting a comprehensive analysis of different types of describing a problem, which shows regularities in the formulation and structure of the text representation of problems. Also, we demonstrate the similarities of writing a problem not only in Ukrainian but also in Belarusian, English, and Russian languages. The final result of the paper is a system that uses the morphosyntactic analyzer to process a problem’s text and provide the answer to it. Ukrainian natural language processing is growing rapidly and showing impressive results. Huge possibilities appear as the Gold standard annotated corpus for Ukrainian language was recently developed. The created architecture is flexible, which indicates the possibility of adding both new geometry figures and their properties, as well as the additional logic to the program. The developed system with a little reformatting can be used with other natural languages, such as English, Belarusian or Russian, as the algorithm for text processing is universal due to the globally accepted representations for presenting such types of mathematical problems. Therefore, the further development of the system is possible.

Linguistic Data Model for Natural Languages and Artificial Intelligence. Part 5. Introduction to Logic

Discourse ◽  
2020 ◽  
Vol 6 (3) ◽  
pp. 109-117
Author(s):  
O. M. Polyakov
Keyword(s):  
Artificial Intelligence ◽  
Mathematical Logic ◽  
Natural Language ◽  
Semantic Interpretation ◽  
General View ◽  
World Model ◽  
Semantic Approach ◽  
Natural Languages ◽  
The World ◽  
Linguistic Aspect

Introduction. The article continues the series of publications on the linguistics of relations (hereinafter R–linguistics) and is devoted to an introduction to the logic of natural language in relation to the approach considered in the series. The problem of natural language logic still remains relevant, since this logic differs significantly from traditional mathematical logic. Moreover, with the appearance of artificial intelligence systems, the importance of this problem only increases. The article analyzes logical problems that prevent the application of classical logic methods to natural languages. This is possible because R-linguistics forms the semantics of a language in the form of world model structures in which language sentences are interpreted.Methodology and sources. The results obtained in the previous parts of the series are used as research tools. To develop the necessary mathematical representations in the field of logic and semantics, the formulated concept of the interpretation operator is used.Results and discussion. The problems that arise when studying the logic of natural language in the framework of R–linguistics are analyzed. These issues are discussed in three aspects: the logical aspect itself; the linguistic aspect; the aspect of correlation with reality. A very General approach to language semantics is considered and semantic axioms of the language are formulated. The problems of the language and its logic related to the most General view of semantics are shown.Conclusion. It is shown that the application of mathematical logic, regardless of its type, to the study of natural language logic faces significant problems. This is a consequence of the inconsistency of existing approaches with the world model. But it is the coherence with the world model that allows us to build a new logical approach. Matching with the model means a semantic approach to logic. Even the most General view of semantics allows to formulate important results about the properties of languages that lack meaning. The simplest examples of semantic interpretation of traditional logic demonstrate its semantic problems (primarily related to negation).

Reliability Estimates for IRT-Based Forced-Choice Assessment Scores

2021 ◽  
pp. 109442812199908
Author(s):  
Yin Lin
Keyword(s):  
Impression Management ◽  
Empirical Studies ◽  
Forced Choice ◽  
Reliability Estimation ◽  
Estimation Methods ◽  
High Stakes ◽  
Personnel Decisions ◽  
Assessment Scores ◽  
Reliability Estimates ◽  
Different Types

Forced-choice (FC) assessments of noncognitive psychological constructs (e.g., personality, behavioral tendencies) are popular in high-stakes organizational testing scenarios (e.g., informing hiring decisions) due to their enhanced resistance against response distortions (e.g., faking good, impression management). The measurement precisions of FC assessment scores used to inform personnel decisions are of paramount importance in practice. Different types of reliability estimates are reported for FC assessment scores in current publications, while consensus on best practices appears to be lacking. In order to provide understanding and structure around the reporting of FC reliability, this study systematically examined different types of reliability estimation methods for Thurstonian IRT-based FC assessment scores: their theoretical differences were discussed, and their numerical differences were illustrated through a series of simulations and empirical studies. In doing so, this study provides a practical guide for appraising different reliability estimation methods for IRT-based FC assessment scores.

scholarly journals Large-scale Semantic Parsing without Question-Answer Pairs

2014 ◽  
Vol 2 ◽  
pp. 377-392 ◽  
Author(s):  
Siva Reddy ◽  
Mirella Lapata ◽  
Mark Steedman
Keyword(s):  
Natural Language ◽  
Large Scale ◽  
Graph Matching ◽  
State Of The Art ◽  
The State ◽  
Semantic Parsing ◽  
Matching Problem ◽  
Weak Supervision ◽  
Benchmark Datasets

In this paper we introduce a novel semantic parsing approach to query Freebase in natural language without requiring manual annotations or question-answer pairs. Our key insight is to represent natural language via semantic graphs whose topology shares many commonalities with Freebase. Given this representation, we conceptualize semantic parsing as a graph matching problem. Our model converts sentences to semantic graphs using CCG and subsequently grounds them to Freebase guided by denotations as a form of weak supervision. Evaluation experiments on a subset of the Free917 and WebQuestions benchmark datasets show our semantic parser improves over the state of the art.

A UNIFIED AND SYNCHRONOUS GENERATING SYSTEM FOR MULTIPLE NATURAL LANGUAGES BASED ON CFG AND SEMANTIC LANGUAGE

2012 ◽  
Vol 10 (04) ◽  
pp. 1250033
Author(s):  
LI LI ◽  
HONGLAI LIU ◽  
QINGSHI GAO ◽  
PEIFENG WANG
Keyword(s):  
Natural Language ◽  
Natural Languages ◽  
Generating System

The sentences in several different natural languages can be produced congruously and synchronous by the new generating system USGS = {↔, GI|GI = (TI, N, B-RISU, C-treeI, S, PI, FI), I = 0, 1, 2, …, n}, based on Semantic Language(SL) theory, all are legitimate and reasonable, where, B-RISU is the set of basic-RISU, C-treeI is the set of category-trees, and FI is the set of functions in I-natural language. The characteristic of this new generating system is unified, synchronous and one by one corresponding, based on semantic unit theory and that the number of rules is several millions.

scholarly journals Natural Language Processing by Enhanced Honey Encryption Technique

2019 ◽  
Vol 8 (12S) ◽  
pp. 159-163
Keyword(s):  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Cyber Attacks ◽  
Binary Form ◽  
Brute Force ◽  
Natural Languages ◽  
Cipher Text ◽  
The Right ◽  
Binary Strings

Traditional encryption systems and techniques have always been vulnerable to brute force cyber-attacks. This is due to bytes encoding of characters utf8 also known as ASCII characters. Therefore, an opponent who intercepts a cipher text and attempts to decrypt the signal by applying brute force with a faulty pass key can detect some of the decrypted signals by employing a mixture of symbols that are not uniformly dispersed and contain no meaningful significance. Honey encoding technique is suggested to curb this classical authentication weakness by developing cipher-texts that provide correct and evenly dispersed but untrue plaintexts after decryption with a false key. This technique is only suitable for passkeys and PINs. Its adjustment in order to promote the encoding of the texts of natural languages such as electronic mails, records generated by man, still remained an open-end drawback. Prevailing proposed schemes to expand the encryption of natural language messages schedule exposes fragments of the plaintext embedded with coded data, thus they are more prone to cipher text attacks. In this paper, amending honey encoded system is proposed to promote natural language message encryption. The main aim was to create a framework that would encrypt a signal fully in binary form. As an end result, most binary strings semantically generate the right texts to trick an opponent who tries to decipher an error key in the cipher text. The security of the suggested system is assessed..

Sign in / Sign up

Export Citation Format

Share Document