Who is the root in a syntactic dependency structure?

This paper describes an analysis method which uses heuristic knowledge to find local syntactic structures of Chinese sentences. We call it a preprocessing, because we use it before we do global syntactic structure analysis (1) of the input sentence. Our purpose is to guide the global analysis through the search space, to avoid unnecessary computation.To realize this, we use a set of special words that appear in commonly used patterns in Chinese. We call them They enable us to pick out fragments that might figure in the syntactic structure of the sentence. Knowledge concerning the use of characteristic words enables us to rate alternative fragments, according to pattern statistics, fragment length, distance between characteristic words, and so on. The preprocessing system proposes to the global analysis level a most likely partial structure. In case this choice is rejected, backtracking looks for a second choice, and so on.For our system, we use 200 characteristic words. Their rules are written by 101 automata. We tested them against 120 sentences taken from a Chinese physics text book. For this limited set, correct partial structures were proposed as first choice for 94% of sentences. Allowing a 2nd choice, the score is 98%, with a 3rd choice, the score is 100%.

In order to support the processes of reading digital research articles, we propose a summarisation method focused on using the localised topic information of documents. The objective is to produce summaries containing topics which are detailed in a document rather than those representing a general overview. We obtain the summaries by ranking sentences based on key-terms cooccurring throughout the document structure, and key-terms within the syntactic structure of sentences. Our idea of using the concept of syntactic structures is that authors usually have their own writing styles, represented by grammatical structures. In this paper we present a description of the method along with an evaluation, comparing our method to other existing summarisation methods.

A long-standing assumption in the syntactic literature is that coordination can only target constituents. This assumption has been a subject of much debate, with many authors questioning its validity. This article enters this debate by reconsidering a constraint on left-sharing in coordination which Osborne & Gross (2017) have recently introduced, namelyleft node blocking. To account for this constraint, Osborne & Gross proposethePrinciple ofFullClusivitywhich states that coordination cannot cut into a constituent. They couch their analysis in a Dependency Grammar, assuming that coordination does not have to conjoin constituents and that syntactic structures should be construed as flat. Given that the empirical ground on which the LNB is based is not firm, I seek to experimentally investigate it by conducting a large-scale experiment. The results of the investigation reveal that LNB is wrong; left-sharing is as permissive as right-sharing. The results of the investigation have the immediate consequence that the assumptions on which LNB is based are wrong as well, namely that syntactic structures should be construed as flat. I spell out an analysis couched in terms of left-to-right syntax to account for major cases of left-sharing in coordination.

Separating hierarchical relations and word order in language production: is proximity concord syntactic or linear?

For most clustering methods, not only the number of clusters must be set in advance, but also various hyperparameters such as initial centroids, number of nearest neighbours, the minimum number of points, neighbourhood radius, and cutoff distance all require pre-specification. As one of the most promising unsupervised learning methods in machine intelligence, existing clustering methods cannot simultaneously handle datasets with arbitrary shapes, different densities, distinct sizes, and overlapping. Background outliers and high dimensionality make clustering problems more challenging. In this paper, we propose a novel universal clustering methodology, called G2-SCANN, which yields the best clustering performance for all 30 synthetic and real datasets without any hyperparameter tuning if the exact number of clusters is known. Firstly, the shortest path length (SPL) in complex network or graph-based geodesic distance is used to give a locally backbone-structured description of graph vertex similarity. Accordingly, SPL-weighted local degree (SLD) is defined as vertex attributes of a SPL-weighted graph expressed by G2-SPL adjacency matrix with ε-natural neighbourhood. Secondly, the process of calculating SLD for every data point in a bottom-up way directly leads to division from a complete graph constituted by all data points to a group of SLD trees. This brings the interpretability and the elimination of lone trees. Thirdly, contrastive learning of largest SLD values for finding root vertices of each divisive tree is conducted and top-down category message is then transmitted from the root vertices to all the leaf ones of a SLD tree. It eventually produces tree-like clusters. Totally, the proposed G2-SCANN method leverages both local neighbouring similarity of data points and global information about data distribution and makes it perform better than other methods.

The cortical dynamics in building syntactic structures of sentences: An MEG study in a minimal-pair paradigm

Recognition Textual Entailment is a fundamental research direction in natural language processing. However, due to the complexity of sentence structure and the multiple parts of speech of words, it affects the model's deep understanding of sentences. From a linguistic perspective, different parts of speech have different understandings of sentences. We believe that a complete sentence consists of two parts: sentence structure and syntactic structure. In response to this issue, this article proposes a recognition textual entailment model that integrates attention mechanism and part of speech tagging. NLTK (Natural Language Toolkit) is used to annotate the part of speech of sentences, and the part of speech information is interactively fused with the sentence features extracted by the Conformer model. After part of speech tagging, the model can better grasp the syntactic structure of sentences. Experiments have shown that the above method combines sentence structure and syntactic structure, which can effectively enhance the model's understanding of deep semantics and improve its accuracy compared to classical models.

Natural Language Processing is the multidisciplinary area of Artificial Intelligence, Machine Learning and Computational Linguistic for processing human language automatically. It involves understanding and processing of human language. The way through which we share our contents or feelings have always great importance in understanding and processing of language. Parsing is the most suited approach in identifying and scanning what the available sentences expressed? Parsing is the process in which syntactic structure of sentence is identified using grammatical tags. The syntactically correct sentence structure is achieved by assigning grammatical labels to its constituents using lexicon and syntactic rules. Phrase and Dependency are two main structure formalisms for parsing natural language sentences. The growing use of web 2.0 has produced novel research challenges as people from different geographical areas are using this channel and sharing contents in their native languages. Urdu is one of such free word order native language which is widely shared over social media sites but identification and summarization of Urdu sentences is challenging task. In this review paper we present an overview to recent work in parsing of fixed order (i.e. English) and free word order languages (i.e Urdu) in order to reveal the most suited method for Urdu Language Parsing. This survey explored that dependency parsing is more appropriate for Urdu and other free word order languages and parsers of English language are not useful in parsing Urdu sentence due to its morphological, syntactical and grammatical differences.

The paper deals with methodological issues of teaching senior students of language universities to analyze the syntactic structure of sentences in modern Chinese at the phrase level. The technology of teaching Russian students to analyze the syntactic structure of sentences in modern Chinese is presented in the following stages: 1) theoretical explanation of the classification of phrases in modern Chinese and methods of mastering this phenomenon by students; 2) introduction to five different ways of syntactic analysis of sentences at the phrase level; 3) presentation of features of the analysis "from small to large"; 4) improvement of skills of independent analysis of offers by students using the specified method. Much attention is paid to the order of establishment of relations between lexical items in phrases, combined with each other semantically. At the same time, in the developed technology, it is recommended to analyze the group of subject and predicate separately, performing an analysis from right to left. When faced with more difficult cases - long complicated sentences, the authors recommend analyzing the sentence in both directions, both from left to right and from right to left. Russian translation is recommended when determining the type of a phrase in Chinese. It is recommended that you translate the phrase into Russian, set its type in Russian, and then make a conclusion about the type of this phrase based on the typology used in Chinese. The existing limitations of using this method are discussed. As the main method of analyzing the syntactic structure of Chinese sentences, the analysis "from small to large" is chosen as the most convenient in implementation and allows parsing long complicated sentences.

The paper offers a new way of presenting the structure of a sentence. None of the two widely known methods of representation the syntactic structure of a sentence can be of any avail when applied to the Lithuanian language. Neither the tree, based on the phrase structure principle, nor the tree, suggested by the dependency grammar, do reflect all the syntactic information, which a Lithuanian sentence contains. The paper points out the differences between the Lithuanian language and other languages as well as presents the reasons why a Lithuanian sentence should be represented by a graph. The paper presents a generalized structure of a simple sentence in the Lithuanian language, namely, such a structure, which would embrace all the possible instances of a Lithuanian simple sentence. Every sentence of the text would have to activate only one path in the generalized structure.

Introduction One of recent generative theory's leading ideas is that syntax is a projection of the lexicon . The primary goal of this book is to explore this hypothesis and to propose an explicit theory of the mapping between the lexicon and morphosyntactic structure. I will argue that this hypothesis is correct if by ‘lexicon’ we understand predicate argument structure , which is an integral part of the lexical entry of every verb and, more generally, of every predicator in the mental lexicon. My main hypothesis is that a sentence's core syntactic structure (vP) is the direct projection of V 's argument structure. More specifically, argument structure has its own syntax, i.e., it has hierarchical internal structure which is operated on by argument-structure specific rules. This entails that vP is fully determined by the homologous structure of the head verb's final derived argument structure . In other words, in the argument-structure based theory of morphosyntax presented in this book, the grammatical (syntactic) relations of a sentence's arguments are fully determined by the internal organization of V 's diathesis. It is in this sense that V heads its clause. This theory requires that we pay careful attention to whether the rules responsible for a sentence's derivation operate on argument structure ( V 's diathesis) or on the syntactic structure it projects: many operations that were thought to be syntactic will be shown to be diathetic. For example, wh -movement, which does not involve a change of grammatical relations or case, is patently a syntactic rule.

Reviewed by: Clefts and their relatives Marcel Den Dikken Clefts and their relatives. By Matthew Reeve. (Linguistik aktuell/Linguistics today 185.) Amsterdam: John Benjamins, 2012. Pp. xiii, 223. ISBN 9789027255686. $149 (Hb). Devising a deft account of clefts has defied linguists for decades. Past analyses have done little to make these constructions fit comfortably into the generative mold. On the contrary, they have typically excelled only at obfuscation and mystification: it as an expletive, the copula as a curious bit of matter with no obvious structural position or function in a ‘monoclausal’ structure with two finite verbs, an obligatorily extraposed relative clause of a peculiar type—and these are just the most eye-catching quirks. Clefts and their relatives (a revised version of Matthew Reeve’s University College London doctoral dissertation) not only provides much-needed clarity as to how the various ingredients of the cleft find their proper place in the syntactic structure, but it also presents an interesting argument to the effect that clefts are not quite alone in the world of syntax in featuring the kinds of structural relations that they do. The title of the book reflects this dual achievement thanks to the fact that both the technical and common or garden senses of the word ‘relatives’ apply. The book is a splendidly written1 demonstration of how rigorous application of syntactic diagnostics and adoption of a number of fairly simple analytical tools pave the way for the construction of an original and explicit structural account of the syntax of a recalcitrant construction. It not only offers a better fit with the properties of clefts documented in the previous literature but is also packed with remarkable empirical discoveries within and beyond the realm of the construction under discussion, and with exciting connections between seemingly unrelated data. The main emphasis of the discussion is on the syntax of clefts, but the semantics of these constructions is addressed as well. Throughout, the book contributes to our understanding of so-called specificational sentences. The semantics of the proposal is largely straightforward; the syntax is rather mechanical (the exploitation of the functional category Eq(uative)P in the syntax of specificational sentences is a case in point) and in some respects remarkably old-fashioned (exploiting rightward movement and right-adjunction, for instance, and relying heavily on the ways θ-relations are established, with Higginbotham’s θ-binding playing a central role). But my characterizing the syntactic proposal as mechanical should not be read as a fatal critique: as Bugs Bunny retorts in the classic cartoon Hair-raising Hare, just after having been smooched by a wind-up girl bunny, ‘Well, so it’s mechanical!’—and indeed, as long as the mechanics is sound, there is nothing wrong. In what follows, I put some of the mechanics to the test. The four main ingredients of an English-type it-cleft are the subject pronoun it, the copula, the focused constituent in postcopular position, and the subordinate clause following the focused constituent. R’s central argument throughout the book is that the subordinate clause is a restrictive relative that, while semantically restricting the reference of the pronoun in subject position, syntactically belongs to the focused constituent but is obligatorily extraposed from that constituent. This statement encapsulates two striking mismatches, one between syntax and semantics (the relative clause syntactically belongs to the focused constituent but semantically restricts the subject pronoun) and the other within the syntax (the relative clause cannot form a surface constituent with what it is construed with in syntax). Resolving these mismatches turns out to be worth the effort: it clarifies the mystery of clefts in a multitude of ways, and leads to the establishment of a close syntactic and semantic relationship with a little-known construction that at first blush seems to bear no close family resemblance to the cleft (the only-relative construction illustrated by I only saw John that I like) and a partial relationship with a well-known construction that looks much more like the it-cleft on the surface (the it-extraposition construction instantiated by It is unfortunate that he said this). I save the discoveries about the relative clause of the [End Page 918] cleft and...

Extracting entity relations from unstructured medical texts is a fundamental task in the field of medical information extraction. In relation extraction, dependency trees contain rich structural information that helps capture the long-range relations between entities. However, many models cannot effectively use dependency information or learn sentence information adequately. In this paper, we propose a relation extraction model based on syntactic dependency structure information. First, the model learns sentence sequence information by Bi-LSTM. Then, the model learns syntactic dependency structure information through graph convolutional networks. Meanwhile, in order to remove irrelevant information from the dependencies, the model adopts a new pruning strategy. Finally, the model adds a multi-head attention mechanism to focus on the entity information in the sentence from multiple aspects. We evaluate the proposed model on a Chinese medical entity relation extraction dataset. Experimental results show that our model can learn dependency relation information better and has higher performance than other baseline models.

Based on her conception that the subject is the least droppable NP in a clause, and that object position is generated only because the thematic structure of the predicator requires it, Rothstein (1995, 2001, 2006) proposes a structural theory of predication whereby predication is a syntactic relation which is independent of theta–role assignment holding between a predicate and its subject. Based on this theory, she goes as far as to claim that “there can be no such thing as an object pleonastic”. This article provides evidence from Mandarin and some Chinese dialects, showing that there is a third person singular pronoun that can occur as expletive in object positions. It is argued on empirical grounds that a structural theory of predication such as Rothstein’s, which rests upon parsimonious English data, still faces the challenge faced by the Principle of Projection.

This chapter focuses on the structure of the sentence. First, we introduce the fundamental concepts. The prototypical Spanish sentence is structured by the verbal predicate with which a number of arguments is associated by their semantic and syntactic function. To this core structure of the sentence, independent adjuncts and other elements add modifying information. We then briefly discuss some approaches to Spanish sentence structure that, in addition to their theoretical contribution, bring other important aspects into focus. On the one hand, generativists defend the existence of a set of principles, parameters and abstract syntactic structures, to some extent related to the cognitive system, that generate the sentences we use when we communicate. On the other hand, cognitive-functional linguistics emphasizes the interaction between cognitive processes, our experience of usage and the formation of linguistic structures. In the chapter, we pay specific attention to the constructionist model of the sentence. Its theoretical foundation is the cognitive-functional approach, and, in particular, the usage-based model of grammar. We thus describe the distribution of verbs in some of the fundamental clausal constructions in Spanish by using corpus data from the BDS database.