Natural Language Systems Research Articles

Steering the conversation: A linguistic exploration of natural language interactions with a digital assistant during simulated driving

Given the proliferation of ‘intelligent’ and ‘socially-aware’ digital assistants embodying everyday mobile technology – and the undeniable logic that utilising voice-activated controls and interfaces in cars reduces the visual and manual distraction of interacting with in-vehicle devices – it appears inevitable that next generation vehicles will be embodied by digital assistants and utilise spoken language as a method of interaction. From a design perspective, defining the language and interaction style that a digital driving assistant should adopt is contingent on the role that they play within the social fabric and context in which they are situated. We therefore conducted a qualitative, Wizard-of-Oz study to explore how drivers might interact linguistically with a natural language digital driving assistant. Twenty-five participants drove for 10 min in a medium-fidelity driving simulator while interacting with a state-of-the-art, high-functioning, conversational digital driving assistant. All exchanges were transcribed and analysed using recognised linguistic techniques, such as discourse and conversation analysis, normally reserved for interpersonal investigation. Language usage patterns demonstrate that interactions with the digital assistant were fundamentally social in nature, with participants affording the assistant equal social status and high-level cognitive processing capability. For example, participants were polite, actively controlled turn-taking during the conversation, and used back-channelling, fillers and hesitation, as they might in human communication. Furthermore, participants expected the digital assistant to understand and process complex requests mitigated with hedging words and expressions, and peppered with vague language and deictic references requiring shared contextual information and mutual understanding. Findings are presented in six themes which emerged during the analysis – formulating responses; turn-taking; back-channelling, fillers and hesitation; vague language; mitigating requests and politeness and praise. The results can be used to inform the design of future in-vehicle natural language systems, in particular to help manage the tension between designing for an engaging dialogue (important for technology acceptance) and designing for an effective dialogue (important to minimise distraction in a driving context).

자연언어 모음체계 보편성 평가 모형 구축 연구

We see from the vowel systems of the world that vocalic phonology generally consists of the three basic vowels, /a i u/ and some other vowels in an ordered fashion. Nevertheless, it is obvious that vowel system of one language differs from another in number of vowels it contains and its inventory. It is well known that vowels are basically classified by the three conventional parameters for vowel description, namely height, position in a front-to-back dimension and lip-rounding. One might say based on this that different vowel systems are different in the three primary parameters. However, a simple description on whether or not a specific vowel, say the high front rounded vowel /y/ is present or absent in a system cannot judge whether the system is universal or very isolated. The question that then arises is which types of vowel system are more universal, and what determines whether the system is universal or not. This paper aims to develop and provide an evaluation model or standards for the vowel systems of natural languages. The model in this paper consists of six standards, two from the property of vocalic typology in relation with number of vowels, and four from the three conventional properties for timbre of a vowel sound. The two standards, [Number of vowels] and [Vowel inventory] fall into the former, and the standards, [Lowness], [Equilibrium and Symmetry], [Frontness] and [Lip roundness] into the latter.

Studying Associations Between Heart Failure Self-Management and Rehospitalizations Using Natural Language Processing.

This study developed an innovative natural language processing algorithm to automatically identify heart failure (HF) patients with ineffective self-management status (in the domains of diet, physical activity, medication adherence, and adherence to clinician appointments) from narrative discharge summary notes. We also analyzed the association between self-management status and preventable 30-day hospital readmissions. Our natural language system achieved relatively high accuracy ( F-measure = 86.3%; precision = 95%; recall = 79.2%) on a testing sample of 300 notes annotated by two human reviewers. In a sample of 8,901 HF patients admitted to our healthcare system, 14.4% ( n = 1,282) had documentation of ineffective HF self-management. Adjusted regression analyses indicated that presence of any skill-related self-management deficit (odds ratio [OR] = 1.3, 95% confidence interval [CI] = [1.1, 1.6]) and non-specific ineffective self-management (OR = 1.5, 95% CI = [1.2, 2]) was significantly associated with readmissions. We have demonstrated the feasibility of identifying ineffective HF self-management from electronic discharge summaries with natural language processing.

GVR predicts 3.9% per annum growth in world carbon black demand to 2022

The National Science Foundation is an independent federal agency created the by National Science Foundation Act of 1950. Its aim is to promote and advance scientific progress in the United States. The idea of such a foundation was an outgrowth of the important contributions made by science and technology during World War II. From those first days, NSF has had a unique place in the federal government: It is responsible for the overall health of science across all disciplines. In contrast, other agencies support research focused on specific missions.NSF funds research in all fields of science and engineering. It does this through grants and contracts to more than 2,000 colleges, universities, and other research institutions in all parts of the United States. The Foundation accounts for about 28 percent of federal support to academic institutions for basic research.NSF receives more than 27,000 proposals each year for research and graduate fellowships and makes more than 12,000 awards. These go to universities, colleges, academic consortia, nonprofit institutions, and small businesses. The agency operates no laboratories itself but does support National Research Centers, certain oceanographic vessels, and Antarctic research stations. The Foundation also aids cooperative research between universities and industry and U.S. participation in international scientific efforts.NSF is structured much like a university, with grant-making divisions for the various disciplines and fields of science and engineering. The Foundation's staff is helped by advisors, primarily from the scientific community, who serve on formal committees or as ad hoc reviewers of research proposals. This advisory system, which focusses on both program direction and specific proposals, involves more than 50,000 scientists and engineers a year, NSF staff members who are experts in a certain field or area make final award decisions; applicants get verbatim unsigned copies of peer reviews and can appeal those decisions.The Knowledge and Database Systems Program of the NSF consists of the following five subject areas: 1. Knowledge Engineering; 2. Formal Models; 3. Compiled Knowledge Systems; 4. Natural Language Systems; 5. Cognitive Systems. For each of these areas a listing of concise project descriptions is given for Fiscal Year 1986.

Multilayer network of language: A unified framework for structural analysis of linguistic subsystems

Abstract Recently, the focus of complex networks’ research has shifted from the analysis of isolated properties of a system toward a more realistic modeling of multiple phenomena — multilayer networks. Motivated by the prosperity of multilayer approach in social, transport or trade systems, we introduce the multilayer networks for language. The multilayer network of language is a unified framework for modeling linguistic subsystems and their structural properties enabling the exploration of their mutual interactions. Various aspects of natural language systems can be represented as complex networks, whose vertices depict linguistic units, while links model their relations. The multilayer network of language is defined by three aspects: the network construction principle, the linguistic subsystem and the language of interest. More precisely, we construct a word-level (syntax and co-occurrence) and a subword-level (syllables and graphemes) network layers, from four variations of original text (in the modeled language). The analysis and comparison of layers at the word and subword-levels are employed in order to determine the mechanism of the structural influences between linguistic units and subsystems. The obtained results suggest that there are substantial differences between the networks’ structures of different language subsystems, which are hidden during the exploration of an isolated layer. The word-level layers share structural properties regardless of the language (e.g. Croatian or English), while the syllabic subword-level expresses more language dependent structural properties. The preserved weighted overlap quantifies the similarity of word-level layers in weighted and directed networks. Moreover, the analysis of motifs reveals a close topological structure of the syntactic and syllabic layers for both languages. The findings corroborate that the multilayer network framework is a powerful, consistent and systematic approach to model several linguistic subsystems simultaneously and hence to provide a more unified view on language.

Review: hindi question answering system using machine learning approach As an upshot of Natural Language Interface to Database (NLIDB), Question Answering System is relatively an Information Retrieval system which is suppose to reflect the user with the correct or closest results to the query being asked to the system in natural language. Information Retrieval and Information Extraction plays vital role

Review: hindi question answering system using machine learning approach As an upshot of Natural Language Interface to Database (NLIDB), Question Answering System is relatively an Information Retrieval system which is suppose to reflect the user with the correct or closest results to the query being asked to the system in natural language. Information Retrieval and Information Extraction plays vital role

Spatial Preposition Use in Indoor Scene Descriptions

GIScience 2016 Short Paper Proceedings Spatial Preposition Use in Indoor Scene Descriptions S. A. Doore, M.K. Beard, N.A. Giudice Spatial Informatics, School of Computing and Information Science, University of Maine Email: {stacy.doore@maine.edu, beard@spatial,maine.edu, nicholas.giudice@maine.edu} Abstract In order to provide accurate automated scene description and navigation directions for indoor space, human beings need intelligent systems to provide an effective cognitive model. Information provided by the structure and use of spatial prepositions is critical to the development of accurate and effective cognitive models. Unfortunately, the use and choice of spatial prepositions in natural language is extremely varied, presenting difficulties for natural language systems attempting to provide descriptions of indoor scenes and wayfinding directions. The goal of the present study is to better understand how humans use spatial prepositions to communicate spatial relationships within virtual environment (VE) indoor scenes. A series of experiments investigates spatial preposition use and the influence scale, topology, orientation and distance within indoor scene descriptions and preliminary results are reported. 1. Introduction Humans perceive and represent information differently for indoor spaces than they do for outdoor spaces (Guidice, Walton and Worboys 2010). Indoor spaces usually lack established distance and direction metrics, global landmarks, explicit route-networks, and cover a range of spatial scales from small rooms to large airports (Winter 2012). Our interest is in creating an indoor spatial description system to assist navigation in indoor environments. This research investigates natural language (NL) structures for describing objects and structural features within vista scale (Montello 1993) virtual environment (VE) indoor scene descriptions. We examine how spatial preposition choice may vary in different contextual settings and which spatial prepositions might yield more effective spatial representations with an increased ability to support spatial behaviors (e.g., object location and navigation). This paper describes early work investigating how spatial factors such as room size, scene elements, and object/structure relationships impact spatial preposition use and semantics for indoor scene descriptions. Previous research has found these same variables of topology, scale, orientation and distance impact spatial preposition use in geographic and table top spatial settings. The current study focuses specifically on vista scale VE scenes (i.e., 10’by 12’ and 20’ by 30’) that one might encounter in common built environments (e.g., home, business, or school size rooms). 2. Motivation Consider two spatial expressions describing the same indoor scene: Speaker 1: “The bookcase is against the wall, beyond the table, just to your left.” Speaker 2: “The bookcase is to the left of the window, directly in front of you.” In both cases, the speakers use the bookcase as the focus of the spatial description, however, the statements differ in how the bookcase is spatially situated in the scene. The first speaker uses the wall as the primary spatial landmark, whereas, the second speaker uses the window.

Jane J. Robinson

Jane Robinson, a pioneering computational linguist, made major contributions to machine translation, natural language, and speech systems research programs at the RAND Corporation, IBM, and in the AI Center at SRI International. She served as ACL president in 1982. Jane became a computational linguist accidentally. She had a Ph.D. in history from UCLA, but could not obtain a faculty position in that field because those were reserved for men. Instead, she took positions teaching English, first at UCLA and then at California State College, Los Angeles. While at LA State, where she was tasked with teaching engineers how towrite, Jane noticed an announcement for a talk on Chomsky’s transformational grammar. She went to the talk thinking this work on grammar might help her teach better. Although its subject matter did not match her expectations, the talk marked a turning point in her career. In the late 1950s, Jane became a consultant to the RANDCorporation groupworking on machine translation under Dave Hays (ACL president, 1964). From the beginning, Jane was concerned with identifying connections between different traditions in formal grammars and their corresponding detailed linguistic realizations. Her 1965 International Conference on Computational Linguistics (COLING) paper, “Endocentric constructions and the Cocke parsing logic” (Robinson 1965), is a beautiful example of connecting specific linguistic phenomena to parsing strategies in a way that preserves the nature of the linguistic phenomena, endocentric constructions. While at RAND, Jane became colleague and friend to many in the machine translation and emerging computational linguistics world, including Susumo Kuno (ACL president 1967), Martin Kay (ACL president, 1969), Joyce Friedman (ACL president, 1971), and Karen Sparck Jones (ACL president, 1994). In the late 1960s, Jane moved to the Automata Theory and Computability Group at the IBM Thomas J. Watson Research Center in Yorktown Heights, NY. She used her knowledge of formal work on grammars and parsing to draw correspondences between Dependency Grammars and Phrase Structure Grammars. Although Jane came

A Rich-Club Organization in Brain Ischemia Protein Interaction Network.

Ischemic stroke involves multiple pathophysiological mechanisms with complex interactions. Efforts to decipher those mechanisms and understand the evolution of cerebral injury is key for developing successful interventions. In an innovative approach, we use literature mining, natural language processing and systems biology tools to construct, annotate and curate a brain ischemia interactome. The curated interactome includes proteins that are deregulated after cerebral ischemia in human and experimental stroke. Network analysis of the interactome revealed a rich-club organization indicating the presence of a densely interconnected hub structure of prominent contributors to disease pathogenesis. Functional annotation of the interactome uncovered prominent pathways and highlighted the critical role of the complement and coagulation cascade in the initiation and amplification of injury starting by activation of the rich-club. We performed an in-silico screen for putative interventions that have pleiotropic effects on rich-club components and we identified estrogen as a prominent candidate. Our findings show that complex network analysis of disease related interactomes may lead to a better understanding of pathogenic mechanisms and provide cost-effective and mechanism-based discovery of candidate therapeutics.

Visualization of Mechanics Problems based on Natural Language Processing

In The ability to visualize our thoughts has always fascinated us. Even more intriguing is the subject of making a computer able to visualize those thoughts, just by understanding the human language. In this paper, a text to scene generation system is proposed for the educational domain where a basic Newtonian physics problem is conveyed to the system in natural language and the scene depicting the problem is generated and displayed to the user. The paper describes the implementation of the system as well as the results obtained. It is based on the integration of advances in NLP and computer graphics technology to generate a virtual environment. It makes it easier for students to visualize the problems and also helps teachers in explaining better. General Terms Your general terms must be any term which can be used for general classification of the submitted material such as Pattern Recognition, Security, Algorithms et. al.

Mixed-complexity artificial grammar learning in humans and macaque monkeys: evaluating learning strategies.

Artificial grammars (AG) can be used to generate rule‐based sequences of stimuli. Some of these can be used to investigate sequence‐processing computations in non‐human animals that might be related to, but not unique to, human language. Previous AG learning studies in non‐human animals have used different AGs to separately test for specific sequence‐processing abilities. However, given that natural language and certain animal communication systems (in particular, song) have multiple levels of complexity, mixed‐complexity AGs are needed to simultaneously evaluate sensitivity to the different features of the AG. Here, we tested humans and Rhesus macaques using a mixed‐complexity auditory AG, containing both adjacent (local) and non‐adjacent (longer‐distance) relationships. Following exposure to exemplary sequences generated by the AG, humans and macaques were individually tested with sequences that were either consistent with the AG or violated specific adjacent or non‐adjacent relationships. We observed a considerable level of cross‐species correspondence in the sensitivity of both humans and macaques to the adjacent AG relationships and to the statistical properties of the sequences. We found no significant sensitivity to the non‐adjacent AG relationships in the macaques. A subset of humans was sensitive to this non‐adjacent relationship, revealing interesting between‐ and within‐species differences in AG learning strategies. The results suggest that humans and macaques are largely comparably sensitive to the adjacent AG relationships and their statistical properties. However, in the presence of multiple cues to grammaticality, the non‐adjacent relationships are less salient to the macaques and many of the humans.

Provably correct reactive control from natural language

This paper presents an integrated system for generating, troubleshooting, and executing correct-by-construction controllers for autonomous robots using natural language input, allowing non-expert users to command robots to perform high-level tasks. This system unites the power of formal methods with the accessibility of natural language, providing controllers for implementable high-level task specifications, easy-to-understand feedback on those that cannot be achieved, and natural language explanation of the reason for the robot's actions during execution. The natural language system uses domain-general components that can easily be adapted to cover the vocabulary of new applications. Generation of a linear temporal logic specification from the user's natural language input uses a novel data structure that allows for subsequent mapping of logical propositions back to natural language, enabling natural language feedback about problems with the specification that are only identifiable in the logical form. We demonstrate the robustness of the natural language understanding system through a user study where participants interacted with a simulated robot in a search and rescue scenario. Automated analysis and user feedback on unimplementable specifications is demonstrated using an example involving a robot assistant in a hospital.

An evaluation of retranslation methods in computing with words

To represent output fuzzy values of a computing with words (CW) system in natural language, a retranslation unit is required. In this work, retranslation methods applicable to a CW system are explored. Several methods that employ similarity measures of fuzzy sets, linguistic modifiers, or linguistic quantifiers have been applied to three real-world case studies. Performances of the applied methods have been evaluated through degree of validity, and comparison of characteristics of fuzzy sets such as fuzziness and specificity. Results show that invalid linguistic terms might be used in the retranslation process which also cause incomprehensible phrases in natural language.

Inference of Probabilistic Grammars in Different Rules Systems of Natural Languages

Abstract Grammar induction is an important and current area within computational linguistics. The results in induction of string transformation rules can be applied not only in linguistics but in many other pattern recognition tasks. The task of grammar induction belongs in general to NP -hard problems. The paper analyses the main rule types in natural languages and in string transformation systems. The paper focuses on the inference of inflectional rule systems which differs in many aspects from the traditional production rule system of Chomsky grammars. The work compares the main candidate methods on the learning of objective-case in the Hungarian language.

Natural Language Understanding and Prediction: from Formal Grammars to Large Scale Machine Learning

Scientists have long dreamed of creating machines humans could interact with by voice. Although one no longer believes Turing's prophecy that machines will be able to converse like humans in the near future, real progress has been made in the voice and text-based human-machine interaction. This paper is a light introduction and survey of some deployed natural language systems and technologies and their historical evolution. We review two fundamental problems involving natural language: the language prediction problem and the language understanding problem. While describing in detail all these technologies is beyond our scope, we do comment on some aspects less discussed in the literature such as language prediction using huge models and semantic labeling using Marcus contextual grammars.

Birth of an Abstraction: A Dynamical Systems Account of the Discovery of an Elsewhere Principle in a Category Learning Task

Human participants and recurrent ("connectionist") neural networks were both trained on a categorization system abstractly similar to natural language systems involving irregular ("strong") classes and a default class. Both the humans and the networks exhibited staged learning and a generalization pattern reminiscent of the Elsewhere Condition (Kiparsky, 1973). Previous connectionist accounts of related phenomena have often been vague about the nature of the networks' encoding systems. We analyzed our network using dynamical systems theory, revealing topological and geometric properties that can be directly compared with the mechanisms of non-connectionist, rule-based accounts. The results reveal that the networks "contain" structures related to mechanisms posited by rule-based models, partly vindicating the insights of these models. On the other hand, they support the one mechanism (OM), as opposed to the more than one mechanism (MOM), view of symbolic abstraction by showing how the appearance of MOM behavior can arise emergently from one underlying set of principles. The key new contribution of this study is to show that dynamical systems theory can allow us to explicitly characterize the relationship between the two perspectives in implemented models.

Towards a continuous population model for natural language vowel shift

The Great English Vowel Shift of 16th–19th centuries and the current Northern Cities Vowel Shift are two examples of collective language processes characterized by regular phonetic changes, that is, gradual changes in vowel pronunciation over time. Here we develop a structured population approach to modeling such regular changes in the vowel systems of natural languages, taking into account learning patterns and effects such as social trends. We treat vowel pronunciation as a continuous variable in vowel space and allow for a continuous dependence of vowel pronunciation in time and age of the speaker. The theory of mixtures with continuous diversity provides a framework for the model, which extends the McKendrick–von Foerster equation to populations with age and phonetic structures. We develop the general balance equations for such populations and propose explicit expressions for the factors that impact the evolution of the vowel pronunciation distribution. For illustration, we present two examples of numerical simulations. In the first one we study a stationary solution corresponding to a state of phonetic equilibrium, in which speakers of all ages share a similar phonetic profile. We characterize the variance of the phonetic distribution in terms of a parameter measuring a ratio of phonetic attraction to dispersion. In the second example we show how vowel shift occurs upon starting with an initial condition consisting of a majority pronunciation that is affected by an immigrant minority with a different vowel pronunciation distribution. The approach developed here for vowel systems may be applied also to other learning situations and other time-dependent processes of cognition in self-interacting populations, like opinions or perceptions.

SWSNL: Semantic Web Search Using Natural Language

As modern search engines are approaching the ability to deal with queries expressed in natural language, full support of natural language interfaces seems to be the next step in the development of future systems. The vision is that of users being able to tell a computer what they would like to find, using any number of sentences and as many details as requested. In this article we describe our effort to move towards this future using currently available technology. The Semantic Web framework was chosen as the best means to achieve this goal. We present our approach to building a complete Semantic Web Search Using Natural Language (SWSNL) system. We cover the complete process which includes preprocessing, semantic analysis, semantic interpretation, and executing a SPARQL query to retrieve the results. We perform an end-to-end evaluation on a domain dealing with accommodation options. The domain data come from an existing accommodation portal and we use a corpus of queries obtained by a Facebook campaign. In our paper we work with written texts in the Czech language. In addition to that, the Natural Language Understanding (NLU) module is evaluated on another domain (public transportation) and language (English). We expect that our findings will be valuable for the research community as they are strongly related to issues found in real-world scenarios. We struggled with inconsistencies in the actual Web data, with the performance of the Semantic Web engines on a decently sized knowledge base, and others.

2.2.2 A Decision Support Framework for the Software Systems Engineer

AbstractDecision Support Systems have been developed to aid the decision maker by providing ranked alternatives to consider. The ranking criteria may be resource‐based, schedule‐based or performance‐based. When considering decisions that affect the overall architecture of a software system and the integration of its subsystems/components, the software systems engineer may have some unique challenges when legacy software is part of the system. For the software systems engineer, reengineering legacy systems or integrating legacy systems with new development and/or off‐the‐shelf software are areas where effective software system analysis is critical. When analyzing a software system, a range of system artifacts may be used. These artifacts include a description of the system in natural language, program design language descriptions, flow charts, object diagrams, as well as the system code. All of these artifact types can contain equivalent information at their specific level of abstraction. However, deriving information that is needed from each of these types of artifacts for reengineering or integrating a legacy system may present varying levels of difficulty. For example, analyzing a poorly structured system with little or no design documentation is a particularly challenging problem. Software modelling using an abstract architecture representation as a common description vehicle can “level the playing field” by creating representations of disparate systems at the same level of abstraction. This is of importance to reengineering efforts; it shows promise for supporting and facilitating integration efforts between legacy and other systems. It also allows the definition and development of a framework for a decision support system to aid the software system engineer. This paper is a follow on to previous work by the authors and presents a proposed decision support system framework that makes the use of the technical analyses and techniques presented.

Analogical Dialogue Acts: Supporting Learning by Reading Analogies in Instructional Texts

Analogy is heavily used in instructional texts. We introduce the concept of analogical dialogue acts (ADAs), which represent the roles utterances play in instructional analogies. We describe a catalog of such acts, based on ideas from structure-mapping theory. We focus on the operations that these acts lead to while understanding instructional texts, using the Structure-Mapping Engine (SME) and dynamic case construction in a computational model. We test this model on a small corpus of instructional analogies expressed in simplified English, which were understood via a semi-automatic natural language system using analogical dialogue acts. The model enabled a system to answer questions after understanding the analogies that it was not able to answer without them.