Natural Language Meaning Research Articles

Against semantic features: the view from derivational affixes

This paper builds a systematic argument against the existence of semantic features, although these would in principle conform with the understanding of features in Chomsky (1995) as instructions to the interfaces, to the Conceptual-Intentional Interface in this case. I first lay out their superfluous character as well as their redundancy in separationist / realisational approaches, and in non-lexicalist models of grammar, more generally. Under the assumption that lexical meaning in natural language is mediated by grammatical structure containing roots, (purely) semantic features would inevitably be restricted to “non-lexical” elements only, i.e. those derivational affixes that encode rich conceptual content. This makes the positing of semantic features methodologically suspect and, ultimately, redundant.Accordingly, the rich content of derivational affixes, which can involve pretty much any nominal concept (as in Acquaviva 2009) from ‘profession’, ’tree’, and ‘place’ to body parts, will be argued not to be encoded in terms of semantic features. On the contrary, this paper makes the case for derivational affixes not belonging to a unitary syntactic category, with some derivational affixes actually being roots interpreted in particular structural contexts, as has been argued already since De Belder (2011). The chapter closes by offering a taxonomy of the elements that grammar manipulates and sketches the division of labour between root structures and formal features.

Machine Translation and Natural Language Processing

Machine translation and natural language processing are powerful tools that allow for the automated translation of text from one language to another. Machine translation systems have evolved from rule-based systems to statistical and neural network-based approaches, with each approach having its own strengths and weaknesses. NLP, on the other hand, focuses on understanding the structure and meaning of natural language and includes tasks such as part-of-speech tagging, named entity recognition, and sentiment analysis. Thegoalofmachinetranslation,a subfield of natural language processing (NLP), is to create platform that translate text or speechautomatically.RBMT provides linguistic accuracy and control, but linguistic resources must be built and maintained meticulously. However, SMT makes use of large-scale parallel corpora to automatically learn translation patterns, making it adaptable to new language pairs and context-dependent translations.However, SMT may struggle with handling grammatical nuances and domain-specific vocabulary. Hybrid approaches combining the best of both RBMT andSMT and newer approaches like neural machine translation (NMT) are being explored to overcome the limitations and improve the quality of translation output.

Development of an error correction algorithm for Kazakh language

This article discusses a method for correcting spelling errors in the Kazakh language using the advantages of morphological analysis and a model based on noisy channels. To achieve this goal, modern problems of automatic processing of Kazakh textual information were analyzed, existing linguistic resources and processing systems of the Kazakh language were systematized, the basic requirements for the development of a system for analyzing Kazakh textual information based on machine learning were determined, and models and algorithms for extracting facts from unstructured and poorly structured text arrays were developed. The search function, an enhanced spelling correction algorithm, was utilized in this work and has the ability to recommend the proper spelling of the input text. The maximum editing distance, whether to include the original word when near matches are not found, and how to handle case sensitivity and exclusion based on regular expressions are all easily adjustable features of this functionality. Because of their adaptability, algorithms can be applied to a wide range of problems, from straightforward spell checks in user interfaces to intricate natural language processing assignments. Because of the way it’s designed, the search function finds possible corrections and verifies the context of words while accounting for user preferences like verbosity and ignore markers. Most modern multilingual natural language processing programs use only the graphical stage of text processing. On the other hand, semantic text analysis or analysis of the meaning of natural language is still an important problem in the theory of artificial intelligence and computational linguistics. But in order to process the grammar and semantics of multilingual information, precreated semantic and grammatical corpora of each natural language are necessary. To solve this problem, several tasks were considered and solved. These tasks included the analysis of research in the field of machine learning methods used in the processing of textual information, the existing problems of formalization and modeling of the Kazakh language, as well as the development and implementation of models, methods and algorithms for morphological and semantic analysis of texts of the Kazakh language.

Sound-meaning associations allow listeners to infer the meaning of foreign language words

An attribute of human language is the seemingly arbitrary association between a word’s form and meaning. We provide evidence that the meaning of foreign words can be partially deduced from phonological form. Monolingual English speakers listened to 45 antonym word pairs in nine foreign languages and judged which English words corresponded to these words’ respective meanings. Despite no proficiency in the foreign language tested, participants’ accuracy was higher than chance in each language. Words that shared meaning across languages were more likely to share phonological form. Accuracy in judging meaning from form was associated with participants’ verbal working memory and with how consistently phonological and semantic features of words covaried across unrelated languages. A follow-up study with native Spanish speakers replicated the results. We conclude that sound maps to meaning in natural languages with some regularity, and sensitivity to form-meaning mappings indexes broader cognitive functions.

Semantic Representations during Language Comprehension Are Affected by Context.

The meaning of words in natural language depends crucially on context. However, most neuroimaging studies of word meaning use isolated words and isolated sentences with little context. Because the brain may process natural language differently from how it processes simplified stimuli, there is a pressing need to determine whether prior results on word meaning generalize to natural language. fMRI was used to record human brain activity while four subjects (two female) read words in four conditions that vary in context: narratives, isolated sentences, blocks of semantically similar words, and isolated words. We then compared the signal-to-noise ratio (SNR) of evoked brain responses, and we used a voxelwise encoding modeling approach to compare the representation of semantic information across the four conditions. We find four consistent effects of varying context. First, stimuli with more context evoke brain responses with higher SNR across bilateral visual, temporal, parietal, and prefrontal cortices compared with stimuli with little context. Second, increasing context increases the representation of semantic information across bilateral temporal, parietal, and prefrontal cortices at the group level. In individual subjects, only natural language stimuli consistently evoke widespread representation of semantic information. Third, context affects voxel semantic tuning. Finally, models estimated using stimuli with little context do not generalize well to natural language. These results show that context has large effects on the quality of neuroimaging data and on the representation of meaning in the brain. Thus, neuroimaging studies that use stimuli with little context may not generalize well to the natural regime.SIGNIFICANCE STATEMENT Context is an important part of understanding the meaning of natural language, but most neuroimaging studies of meaning use isolated words and isolated sentences with little context. Here, we examined whether the results of neuroimaging studies that use out-of-context stimuli generalize to natural language. We find that increasing context improves the quality of neuro-imaging data and changes where and how semantic information is represented in the brain. These results suggest that findings from studies using out-of-context stimuli may not generalize to natural language used in daily life.

Towards a unified representation of linguistic meaning

AbstractNatural language meaning has properties of both cognitive representations and formal/mathematical structures. But it is not clear how they actually relate to one another. The central aim of this article is to show that properties of cognitive representations and formal/mathematical structures of natural language meaning, albeit apparently divergent, can be united, as far as the basic properties of semantic structures are concerned. Thus, this article will formulate the form of unified representations for semantic structures. With this goal, this article takes into account standard formal-semantic representations and also Discourse Representation Theory (DRT) representations on the one hand and semantic representations in different versions of Conceptual/Cognitive Semantics (Jackendoff’s, Langacker’s and Talmy’s approaches to Conceptual/Cognitive Semantics) and representations of Mental Spaces (Fauconnier’s approach) on the other hand. The rationale behind the selection of these approaches is that the representations of semantic structures under these approaches are all amenable to unification. It must be emphasized that showing that the representations of semantic structures under these approaches can be unified does not simply amount to unifying these theories/approaches in toto. Rather, it is to demonstrate that cognitive representations and formal/mathematical structures can be shown to be inter-translatable for at least some accounts of linguistic meaning.

Grammar-aware sentence classification on quantum computers

Natural language processing (NLP) is at the forefront of great advances in contemporary AI, and it is arguably one of the most challenging areas of the field. At the same time, in the area of quantum computing (QC), with the steady growth of quantum hardware and notable improvements towards implementations of quantum algorithms, we are approaching an era when quantum computers perform tasks that cannot be done on classical computers with a reasonable amount of resources. This provides an new range of opportunities for AI, and for NLP specifically. In this work, we work with the Categorical Distributional Compositional (DisCoCat) model of natural language meaning, whose underlying mathematical underpinnings make it amenable to quantum instantiations. Earlier work on fault-tolerant quantum algorithms has already demonstrated potential quantum advantage for NLP, notably employing DisCoCat. In this work, we focus on the capabilities of noisy intermediate-scale quantum (NISQ) hardware and perform the first implementation of an NLP task on a NISQ processor, using the DisCoCat framework. Sentences are instantiated as parameterised quantum circuits; word-meanings are embedded in quantum states using parameterised quantum-circuits and the sentence’s grammatical structure faithfully manifests as a pattern of entangling operations which compose the word-circuits into a sentence-circuit. The circuits’ parameters are trained using a classical optimiser in a supervised NLP task of binary classification. Our novel QNLP model shows concrete promise for scalability as the quality of the quantum hardware improves in the near future and solidifies a novel branch of experimental research at the intersection of QC and AI.

Combining Unsupervised and Supervised Learning for Sample Efficient Continuous Language Grounding

Natural and efficient communication with humans requires artificial agents that are able to understand the meaning of natural language. However, understanding natural language is non-trivial and requires proper grounding mechanisms to create links between words and corresponding perceptual information. Since the introduction of the “Symbol Grounding Problem” in 1990, many different grounding approaches have been proposed that either employed supervised or unsupervised learning mechanisms. The latter have the advantage that no other agent is required to learn the correct groundings, while the former are often more sample-efficient and accurate but require the support of another agent, like a human or another artificial agent. Although combining both paradigms seems natural, it has not achieved much attention. Therefore, this paper proposes a hybrid grounding framework which combines both learning paradigms so that it is able to utilize support from a tutor, if available, while it can still learn when no support is provided. Additionally, the framework has been designed to learn in a continuous and open-ended manner so that no explicit training phase is required. The proposed framework is evaluated through two different grounding scenarios and its unsupervised grounding component is compared to a state-of-the-art unsupervised Bayesian grounding framework, while the benefit of combining both paradigms is evaluated through the analysis of different feedback rates. The obtained results show that the employed unsupervised grounding mechanism outperforms the baseline in terms of accuracy, transparency, and deployability and that combining both paradigms increases both the sample-efficiency as well as the accuracy of purely unsupervised grounding, while it ensures that the framework is still able to learn the correct mappings, when no supervision is available.

Can there be a feature‐placing language?

AbstractThe aim of this article is to argue against the real possibility of languages without subject‐predicate structure, so‐called feature‐placing languages. They were first introduced by Strawson (1959/1990), later given formal expression through Quine's Predicate Functor Logic (Quine, 1960, Quine, 1971/Quine, 1976, Quine, 1992), and further elaboration in (Hawthorne & Cortens, 1995). I argue that, on the presumption that feature‐placing languages are not mere notational variants on first‐order languages, the idea of such languages is incoherent. The argument for this view rests on two highly plausible views about the conditions for the possibility of sentential meaning in natural languages. If accepted, these two views jointly imply that the sentences of feature‐placing languages lack sufficient internal structure to meet the conditions for sentential meaning. Feature‐placing languages cannot then count as genuine languages.

Context mining and graph queries on giant biomedical knowledge graphs

Contextual information is widely considered for NLP and knowledge discovery in life sciences since it highly influences the exact meaning of natural language. The scientific challenge is not only to extract such context data, but also to store this data for further query and discovery approaches. Classical approaches use RDF triple stores, which have serious limitations. Here, we propose a multiple step knowledge graph approach using labeled property graphs based on polyglot persistence systems to utilize context data for context mining, graph queries, knowledge discovery and extraction. We introduce the graph-theoretic foundation for a general context concept within semantic networks and show a proof of concept based on biomedical literature and text mining. Our test system contains a knowledge graph derived from the entirety of PubMed and SCAIView data and is enriched with text mining data and domain-specific language data using Biological Expression Language. Here, context is a more general concept than annotations. This dense graph has more than 71M nodes and 850M relationships. We discuss the impact of this novel approach with 27 real-world use cases represented by graph queries. Storing and querying a giant knowledge graph as a labeled property graph is still a technological challenge. Here, we demonstrate how our data model is able to support the understanding and interpretation of biomedical data. We present several real-world use cases that utilize our massive, generated knowledge graph derived from PubMed data and enriched with additional contextual data. Finally, we show a working example in context of biologically relevant information using SCAIView.

Investigating the Effects of Task Type and Linguistic Background on Accuracy in Automated Speech Recognition Systems: Implications for Use in Language Assessment of Young Learners

ABSTRACT As a branch of artificial intelligence, automated speech recognition (ASR) technology is increasingly used to detect speech, process it to text, and derive the meaning of natural language for various learning and assessment purposes. ASR inaccuracy may pose serious threats to valid score interpretations and fair score use for all when it is exacerbated by test takers’ characteristics, such as language background and accent, and assessment task type. The present study investigated the extent to which speech-to-text accuracy rates of three major ASR systems vary across different oral tasks and students’ language background variables. Results indicate that task types and students’ language backgrounds have statistically significant main and interaction effects on ASR accuracy. The paper discusses the implications of the study results for applying ASR to computerized assessment design and automated scoring.

A novel hybrid methodology for computing semantic similarity between sentences through various word senses

In the area of natural language processing, measuring sentence similarity is an essential problem. Searching for semantic meaning in natural language is a related issue. The task of measuring sentence similarity is to find semantic symmetry in two sentences, not matter how they are arranged. It is important to measure the similarity of sentences accurately. To compute the similarity between sentences, existing methods have been constructed from approaches for large texts. Since these methods work in very high-dimensional spaces, they are inefficient, require human input, and are not flexible enough for some applications. In this study, we propose a hybrid method (HydMethod) which considers not only semantic information including lexical databases, word embeddings, and corpus statistics, but also implied word order information. With lexical databases, our method models human common sense knowledge, and that knowledge can then be adapted to be used in different domains with the incorporation of corpus statistics. Therefore, the methodology is applicable across several domains. As part of our experiments, we used two standard datasets - Pilot Short Text Semantic Similarity Benchmark and MS paraphrase - in order to demonstrate the efficacy of our proposed method. As a result, the proposed method outperforms the existing approaches when tested on these two datasets, giving the highest correlation value for both word and sentence similarity. Moreover, it achieves a maximum of 32% higher increase than only using word vector or WorldNet based methodology. With Rubenstein and Goodenough word & sentence pairs, our algorithm's similarity measure shows a high Pearson correlation coefficient of 0.8953.

The Avoidance of Circularity

One of the earliest and most basic lessons that a legislative drafter learns is that circularity in a definition must be avoided at all costs because it literally renders the definition meaningless. Having absorbed that basic lesson, the neophyte drafter quickly learns that a definition may sometimes appear circular superficially, without actually being so: for example, a word that carries a sufficient natural language meaning to stand alone without definition, may be the subject of a definition that uses the word itself, but merely as a peg for a qualification or limitation (“‘vehicle’ includes any vehicle whether or not…”). Constructions using that kind of apparent circularity apart, there is never any excuse for real circularity in a definition, and every drafter is taught to check her or his work with particular attention to the possibility of a circularity having crept in unwittingly. When the courts spot circularity they are rightly...

Meaning and Alternatives

Alternatives and competition in language are pervasive at all levels of linguistic analysis. More specifically, alternatives have been argued to play a prominent role in an ever-growing class of phenomena in the investigation of natural language meaning. In this article, we focus on scalar implicatures, as they are arguably the most paradigmatic case of an alternative-based phenomenon. We first review the main challenge for theories of alternatives, the so-called symmetry problem, and we briefly discuss how it has shaped the different approaches to alternatives. We then turn to two more recent challenges concerning scalar diversity and the inferences of sentences with multiple scalars. Finally, we describe several related alternative-based phenomena and recent conceptual approaches to alternatives. As we discuss, while important progress has been made, much more work is needed both on the theoretical side and on understanding the empirical landscape better.

The geometry vs. the algebra of meaning

In this paper Gärdenfors’s geometric approach to meaning in natural language is compared to Jackendoff's algebraic one, and this is done against the backdrop of formal semantics. Ultimately, the paper tries to show that Jackendoff's framework is to be preferred to all others. The paper proceeds as follows. In Section 2, the common theoretical commitments of Gärdenfors and Jackendoff are outlined, and it is attempted to argue briefly that they are on the right track. In Section 3, the basics of the two frameworks to be compared are laid out, and it is assessed how they deal with some central issues in semantic theory, namely reference and truth, lexical decomposition, and compositionality. In Section 4, we get into the nitty-gritty of how Gärdenfors and Jackendoff actually proceed in semantic analysis, using an example of a noun and a verb (embedded in a sentence). In Section 5, the merits of Gärdenfors's empiricism when it comes to word learning and concept acquisition are assessed and compared to the moderate nativism of Jackendoff, and it is argued that Jackendoff's nativism is to be preferred. In the sixth section, the semantic internalism common to both frameworks is commented on.

The Significance of Linguistics in the Study of Philosophy

Since the era of Greece and Rome in the 4-2 century BC, until this Postmodern one, language has been one of the most central and core issues of philosophical studies. Language and Philosophy both focus on issues related to structure and meaning in natural language, as discussed in the philosophy of language and other disciplines, among others; philosophical theories about meaning and truth, presuppositions, implicatures, speech acts, etc. This article discusses several case studies that illustrate the relationship between the philosophy of language through three branches of linguistics; syntax (Stanley, 2000), semantics (von Fintel, 2001), and pragmatics (Potts, 2005). The results of the study reveal a significance and interdependence between philosophy and language. Philosophy requires language as a means of communicating ideas and also as an object of study in philosophy. Meanwhile, language also badly needs philosophy as a means or method to analyze systematically to get solutions to solving linguistic problems.Keywords: linguistics, philosophy, syntax, semantics, and pragmatics

Efficient natural language classification algorithm for detecting duplicate unsupervised features

This paper focuses on capturing the meaning of Natural Language Understanding (NLU) text features to detect the duplicate unsupervised features. The NLU features are compared with lexical approaches to prove the suitable classification technique. The transfer-learning approach is utilized to train the extraction of features on the Semantic Textual Similarity (STS) task. All features are evaluated with two types of datasets that belong to Bosch bug and Wikipedia article reports. This study aims to structure the recent research efforts by comparing NLU concepts for featuring semantics of text and applying it to IR. The main contribution of this paper is a comparative study of semantic similarity measurements. The experimental results demonstrate the Term Frequency–Inverse Document Frequency (TF-IDF) feature results on both datasets with reasonable vocabulary size. It indicates that the Bidirectional Long Short Term Memory (BiLSTM) can learn the structure of a sentence to improve the classification.

Tools for intellectual support of project decisions based on analyzing uncertain factors of a different nature

Today, the management of complex innovative projects to create new aircraft is carried out in the conditions of uncertainty, which is due to the lack of “high-quality” (complete, accurate, consistent, etc.) data on their internal and external environment. In this situation, to support the adoption of project decisions, it is advisable to use an approach based on the identification and assessment of NON-factors that have a negative meaning in natural language and deny one of the basic properties of formal systems. The authors propose a procedure for analyzing NON-factors in project risk management, based on the assessment of uncertain factors of internal and external project environment using various methods of data mining (particularly, neural fuzzy classifier, fuzzy pyramidal networks, fuzzy logic methods, fuzzy Kalman filtering algorithms). Using the proposed approach will increase the validity of decisions on the viability of such innovative projects and a set of measures for risk management.

Bridging the Gap between the Meaning of Natural Language and Mental Content

Since Frege, mental content (conscious content) has been distinguished from the meaning of natural language and not regarded mental content as the meaning of language expression. This anti-psychological view cuts off the connection between the meaning of language and mental content, giving rise to the failures in solving the problems of mind. Instead of thinking about linguistic meaning and mental content separately, philosophers of mind put more emphasis on mental content and even equate the meaning of linguistic expressions with it. Therefore, the issues that what the relationship between linguistic meaning and mental content is and how they are connected are not well answered so far. The author argues in this paper that: 1) People attach meaning to symbols through communication activities when they have mental contents, thus forming the language, which is also a process of achieving psychological certainty in turn; 2) the relationship between linguistic meaning and mental content is similar to the interdependent relationship between monetary value and the use value of commodities. Without the latter, the former will lose its source, and the former also affects the latter. Philosophers of mind pay little attention to the interplay between language and mental content. Based on the above arguments and previous theories, this paper conceives a general model of the generation of linguistic meaning.

Tardis & Tame: an Essay on Natural Language Meaning and Metaphysics

This paper theoretically approaches the relationship between the instances of time and space as far as natural language conveys their manifestation, focusing on meaning, majorly represented by the subdiscipline of Semantics, along with some insights on Syntax and Pragmatics. The outset-designed ontology is mainly composed by categories of TAME (tense, aspect, mood and evidentiality/eventology) instantiated by linguistic phenomena that yelds anchoring, displacement and aboutness properties. To achieve this wide range of linguistic manifestations the scope takes over the verbal semantics of Brazilian Portuguese (henceforth BrP), pursuing to manage the lexical nature of entries in this language and its metaphysical counterpart in meaning. This sort of approach intends to illustrate the proper balance for the formal device of a semantic component regarding these language parameters on TAME and their principles correlation on human language by means of this paper intends to coin as TARDIS. This paper presents three sections such as: a) theoretical, introducing the properties of each category of TAME, throughout history of Linguistics and Semantics; b) methodological, characterizing the lexical/metaphysical dualism for Formal Semantic approaches and their correlation to time and space and some other non-logical privileged concepts entertained by semantic knowledge; c) epistemological and analytical, considering BrP as parameters for TARDIS’ properties conveyed by natural language meaning throughout different categories of TAME and its correlation to some principles which seem to regard human cognition, focusing mainly on modality.