Natural Language Phrases Research Articles

Effectiveness of lifestyle interventions for glycaemic control among adults with type 2 diabetes in West Africa: a systematic review and meta-analysis protocol

BackgroundLifestyle interventions are key to the control of diabetes and the prevention of complications, especially when used with pharmacological interventions. This protocol aims to review the effectiveness of lifestyle interventions in relation to nutrition and physical activity within the West African region. This systematic review and meta-analysis seeks to understand which interventions for lifestyle modification are implemented for the control of diabetes in West Africa at the individual and community level, what evidence is available on their effectiveness in improving glycaemic control and why these interventions were effective.MethodsWe will review randomised control trials and quasi-experimental designs on interventions relating to physical activity and nutrition in West Africa. Language will be restricted to English and French as these are the most widely spoken languages in the region. No other filters will be applied. Searching will involve four electronic databases — PubMed, Scopus, Africa Journals Online and Cairn.info using natural-language phrases plus reference/citation checking.Two reviewers will independently screen results according to titles and abstracts against the inclusion and exclusion criteria to identify eligible studies. Upon full-text review, all selected studies will be assessed using Cochrane’s Collaboration tool for assessing the risk of bias of a study and the ROBINS-I tool before data extraction. Evidence will be synthesised narratively and statistically where appropriate. We will conduct a meta-analysis when the interventions and contexts are similar enough for pooling and compare the treatment effects of the interventions in rural to urban settings and short term to long term wherever possible.DiscussionWe anticipate finding a number of studies missed by previous reviews and providing evidence of the effectiveness of different nutrition and physical activity interventions within the context of West Africa. This knowledge will support practitioners and policymakers in the design of interventions that are fit for context and purpose within the West African region.Systematic review registrationThis systematic review has been registered in the International Prospective Register for Systematic Reviews — PROSPERO, with registration number CRD42023435116. All amendments to this protocol during the process of the review will be explained accordingly.

Human Intention Space - Natural Language Phrase Driven Approach to Place Social Computing Interaction in a Designed Space

Human Intention Space - Natural Language Phrase Driven Approach to Place Social Computing Interaction in a Designed Space

A comprehensive review on zero-shot-learning techniques

Advancements in computational capabilities have enabled the implementation of advanced deep learning models across various domains of knowledge, yet the increasing complexity and scarcity of data in specialized areas pose significant challenges. Zero-shot learning (ZSL), a subset of transfer learning, has emerged as an innovative solution to these challenges, focusing on classifying unseen categories present in the test set but absent during training. Unlike traditional methods, ZSL utilizes semantic descriptions, like attribute lists or natural language phrases, to map intermediate features from the training data to unseen categories effectively, enhancing the model’s applicability across diverse and complex domains. This review provides a concise synthesis of the advancements, methodologies, and applications in the field of zero-shot learning, highlighting the milestones achieved and possible future directions. We aim to offer insights into the contemporary developments in ZSL, serving as a comprehensive reference for researchers exploring the potentials and challenges of implementing ZSL-based methodologies in real-world scenarios.

A comprehensive review on zero-shot-learning techniques

Advancements in computational capabilities have enabled the implementation of advanced deep learning models across various domains of knowledge, yet the increasing complexity and scarcity of data in specialized areas pose significant challenges. Zero-shot learning (ZSL), a subset of transfer learning, has emerged as an innovative solution to these challenges, focusing on classifying unseen categories present in the test set but absent during training. Unlike traditional methods, ZSL utilizes semantic descriptions, like attribute lists or natural language phrases, to map intermediate features from the training data to unseen categories effectively, enhancing the model’s applicability across diverse and complex domains. This review provides a concise synthesis of the advancements, methodologies, and applications in the field of zero-shot learning, highlighting the milestones achieved and possible future directions. We aim to offer insights into the contemporary developments in ZSL, serving as a comprehensive reference for researchers exploring the potentials and challenges of implementing ZSL-based methodologies in real-world scenarios.

Causality and signalling of garden-path sentences.

Sheaves are mathematical objects that describe the globally compatible data associated with open sets of a topological space. Original examples of sheaves were continuous functions; later they also became powerful tools in algebraic geometry, as well as logic and set theory. More recently, sheaves have been applied to the theory of contextuality in quantum mechanics. Whenever the local data are not necessarily compatible, sheaves are replaced by the simpler setting of presheaves. In previous work, we used presheaves to model lexically ambiguous phrases in natural language and identified the order of their disambiguation. In the work presented here, we model syntactic ambiguities and study a phenomenon in human parsing called garden-pathing. It has been shown that the information-theoretic quantity known as 'surprisal' correlates with human reading times in natural language but fails to do so in garden-path sentences. We compute the degree of signalling in our presheaves using probabilities from the large language model BERT and evaluate predictions on two psycholinguistic datasets. Our degree of signalling outperforms surprisal in two ways: (i) it distinguishes between hard and easy garden-path sentences (with a [Formula: see text]-value [Formula: see text]), whereas existing work could not, (ii) its garden-path effect is larger in one of the datasets (32 ms versus 8.75 ms per word), leading to better prediction accuracies. This article is part of the theme issue 'Quantum contextuality, causality and freedom of choice'.

Actional Atomic-Concept Learning for Demystifying Vision-Language Navigation

Vision-Language Navigation (VLN) is a challenging task which requires an agent to align complex visual observations to language instructions to reach the goal position. Most existing VLN agents directly learn to align the raw directional features and visual features trained using one-hot labels to linguistic instruction features. However, the big semantic gap among these multi-modal inputs makes the alignment difficult and therefore limits the navigation performance. In this paper, we propose Actional Atomic-Concept Learning (AACL), which maps visual observations to actional atomic concepts for facilitating the alignment. Specifically, an actional atomic concept is a natural language phrase containing an atomic action and an object, e.g., ``go up stairs''. These actional atomic concepts, which serve as the bridge between observations and instructions, can effectively mitigate the semantic gap and simplify the alignment. AACL contains three core components: 1) a concept mapping module to map the observations to the actional atomic concept representations through the VLN environment and the recently proposed Contrastive Language-Image Pretraining (CLIP) model, 2) a concept refining adapter to encourage more instruction-oriented object concept extraction by re-ranking the predicted object concepts by CLIP, and 3) an observation co-embedding module which utilizes concept representations to regularize the observation representations. Our AACL establishes new state-of-the-art results on both fine-grained (R2R) and high-level (REVERIE and R2R-Last) VLN benchmarks. Moreover, the visualization shows that AACL significantly improves the interpretability in action decision. Code will be available at https://gitee.com/mindspore/models/tree/master/research/cv/VLN-AACL.

Representing and Predicting Everyday Behavior

The prediction of everyday human behavior is a central goal in the behavioral sciences. However, efforts in this direction have been limited, as (1) the behaviors studied in most surveys and experiments represent only a small fraction of all possible behaviors, and (2) it has been difficult to generalize data from existing studies to predict arbitrary behaviors, owing to the difficulty in adequately representing such behaviors. Our paper attempts to address each of these problems. First, by sampling frequent verb phrases in natural language and refining these through human coding, we compile a dataset of nearly 4000 common human behaviors. Second, we use distributed semantic models to obtain vector representations for our behaviors, and combine these with demographic and psychographic data, to build supervised, deep neural network models of behavioral propensities for a representative sample of the US population. Our best models achieve reasonable accuracy rates when predicting propensities for novel (out-of-sample) participants as well as novel behaviors, and offer new insights for modeling psychographic and demographic differences in behavior. This work is a first step towards building predictive theories of everyday behavior, and thus improving the generality and naturalism of research in the behavioral sciences.

Устойчивое развитие природы и общества Сибири и Дальнего Востока: анализ документопотока электронной коллекции ГПНТБ СО РАН

“Living substance is biosphere's driving force whom main activity type is biogeochemical work in biosphere”. Biosphere development problems came up in twentieth century. Conception of sustainable development as translational headway which provides with society rational needs and conservation of environment state for future generations was accepted in 1922 at UN international conference, Rio de Janeiro. In Russia in 1992 were issued presidential decree № 236 and government decree № 217-r. In SPSTL SB RAS creation of bibliographical database of society and nature sustainable development started in 1997 due to theme actuality and large amount of scientific researches in Russia and in the world entirely. Since 2011 it is introduced in “Scientific Sibirica” database as a topic section. Deposit copies and public access scientific publications are the main database information sources. To cover all theme aspects literature selection and systematization methods and documents analytical processing structure were set up. State Classifier of Scientific and Technical information, not normative key words and thematic classifier are used as language support. Thematic classifier reflects main civilization problems represented in SB RAS research establishments priority areas of activity: general issues (historical, philosophical, law, informational), social and ecological issues, economical issues of sustainable development. Analysis of database information since 1998 to 2017 showed up next points: • reflection of social processes in data store by focus on particular sustainable development directions. • topic actuality. Annual large stock of publications (8700 documents averagely) which are highly dissipated (1236 periodicals (39%) and 5899 scientific events materials (35%) are represented by only one publication). • prevalence of scientific publications for: 13623 symposiums, seminars, conferences (81%); 605 productive Russian journals included in scientometrical systems; 448 (43%) productive Russian journals included in State Commission for Academic Degrees and Titles list. • various access levels of articles full texts: on electronic scientific library (E-library.ru) 531 (68%) productive Russian journals are at free access, the others may be viewed only directly from library or by subscription; on E-library and in Open Access systems materials of symposiums, conferences and seminars are partly accessible and temporarily published on the Internet. In the interest of specialists' and scientists' time economy they receive information by selective dissemination of information system since 1997. Since 1998 information is available by WEB-oriented information retrieval system of SPSTL SB RAS which includes optimized set of retrieval parameters; retrieval by all resources or by some certain resources; various retrieval modes; retrieval result improvement by retrieval vocabulary or hyperlinks in publications description; retrieval result saving in common formats; retrieval request saving and edition; subscription on new arrivals by e-mail; access to information by retrieval result (publications full texts browsing, print media order). From 2019 new WEB-oriented IRS will be supplemented by: full-text retrieval, retrieval by natural language phrase; faceted navigation; various access rights and full texts references registration and other options. Bibliographical resources in-formational opportunities will extend as 22655 documents (17%) from “Sustainable development of society and nature” section of “Scientific Siberica” database already have links on their full text (publications having DOI, publications on CD-ROM; publications in the public domain; published scientific events materials which arrive at libraries in printed format). For Internet-oriented customers on the web-site there are: topic sector description allowing switch to WEB IRS; virtual literature collections are represented (in xml-format) allowing full texts browsing and switch to WEB IRS. There is another advanced research direction -semantic WEB projects based on authority files of authors, personalities, organizations, subject and geographical headings. Update of “Scientific Siberica” bibliographical database and specially “Sustainable development of society and nature” section is directed at next points: documents analytic processing extension; functional files of authors, personalities, organizations, subject and geographical headings authority setting up; links on opened fulltext Internet resources addition; user interface development in WEB IRS and on the SPSTL SB RAS web-site.

Towards combinational relation linking over knowledge graphs

Given a knowledge graph and a natural language phrase, relation linking aims to find relations (predicates or properties) from the underlying knowledge graph to match the phrase. It is very useful in many applications, such as natural language question answering, personalized recommendation and text summarization. However, the previous relation linking algorithms usually produce a single relation for the input phrase and pay little attention to the more general and challenging problem, i.e., combinational relation linking that extracts a subgraph pattern to match the compound phrase (e.g. father-in-law). In this paper, we focus on the task of combinational relation linking over knowledge graphs. To resolve the problem, we define several elementary meta patterns which can be used to build any combinational relation. Then we design a systematic method based on the data-driven relation assembly technique, which is performed under the guidance of meta patterns. To enhance the system’s understanding ability, we introduce external knowledge during the linking process. Finally, extensive experiments over the real knowledge graph confirm the effectiveness of the proposed method.

TREE-BASED SEMANTIC ANALYSIS METHOD FOR NATURAL LANGUAGE PHRASE TO FORMAL QUERY CONVERSION

Context. This work is devoted to the problem of natural language interface construction for ontological graph databases. The focus here is on the methods for the conversion of natural language phrases into formal queries in SPARQL and CYPHER query languages. Objective. The goals of the work are the creation of a semantic analysis method for the input natural language phrases semantic type determination and obtaining meaningful entities from them for query template variables initialization, construction of flexible query templates for the types, development of program implementation of the proposed technique. Method. A tree-based method was developed for semantic determination of a user’s phrase type and obtaining a set of terms from it to put them into certain places of the most suiting formal query template. The proposed technique solves the tasks of the phrase type determination (and this is the criterion of the formal query template selection) and obtaining meaningful terms, which are to initialize variables of the chosen template. In the current work only interrogative and incentive user’s phrases are considered i.e. ones that clearly propose the system to answer or to do something. It is assumed that the considered dialog or reference system uses a graph ontological database, which directly impacts the formal query patterns – the resulting queries are destined to be in SPARQL or Cypher query languages. The semantic analysis examples considered in this work are aimed primarily at inflective languages, especially, Ukrainian and Russian, but the basic principles could be suitable to most of the other languages. Results. The developed method of natural language phrase to a formal query in SPARQL and CYPHER conversion has been implemented in software for Ukrainian and Norwegian languages using narrow subjected ontologies and tested against formal performance criteria. Conclusions. The proposed method allows the dialog system fast and with minimum number of steps to select the most suitable query template and extract informative entities from a natural language phrase given the huge phrase variability in inflective languages. Carried out experiments have shown high precision and reliability of the constructed system and its potential for practical usage and further development.

Development of Natural Language Dialogue Software Systems

This paper discusses problems of natural language dialogue systems development. Here are considered different types of such program systems and their peculiarities based on our development experience. The systems discussed here are aimed at using graph ontological databases. Five basic types of natural language systems are highlighted: reference system, reference system with clarifying and technical questions, questionnaire system, active initiative system, control system. The first four of them are considered in detail. The reference system type, which is the basic type, is shown and discussed. Problems considered here are: natural language phrase to a formal query transformation, semantic analysis and answer formation. The ways are shown for these processes program implementation. For all the types the problem of the current dialogue subject keeping is considered, especially for an active initiative system where it is the most crucial. The different approaches are shown and compared for the principles of graph database (ontology) construction, which could be used in the dialogue systems.

Revisiting Image-Language Networks for Open-Ended Phrase Detection.

Most existing work that grounds natural language phrases in images starts with the assumption that the phrase in question is relevant to the image. In this paper we address a more realistic version of the natural language grounding task where we must both identify whether the phrase is relevant to an image and localize the phrase. This can also be viewed as a generalization of object detection to an open-ended vocabulary, introducing elements of few- and zero-shot detection. We propose an approach for this task that extends Faster R-CNN to relate image regions and phrases. By carefully initializing the classification layers of our network using canonical correlation analysis (CCA), we encourage a solution that is more discerning when reasoning between similar phrases, resulting in over double the performance compared to a naive adaptation on three popular phrase grounding datasets, Flickr30K Entities, ReferIt Game, and Visual Genome, with test-time phrase vocabulary sizes of 5K, 32K, and 159K, respectively.

Information Retrieval with Root- and Rule-Based Terms

Root- and rule-based terms are structured representations of natural language phrases that can be automatically generated using a combination of statistical and symbolic methods. These terms are able to represent and normalize syntactic information about natural language phrases, making them richer than basic n-grams while greatly reducing the vocabulary size. In this paper, we discuss the use of root- and rule-based terms for information retrieval. We represent documents and queries as collections of root- and rule-based terms and show that this improves conventional information retrieval methods such as Latent Semantic Indexing and Latent Direchlet Allocation. Root- and rule-based terms improve on state of the art evaluation scores for the TREC 2016 clinical decision support track.

Simulation Framework for Evaluation of Indoor Navigation Systems

In this paper, we introduce the first simulation framework that provides cost-effective means to evaluate indoor navigation systems for different user groups (e.g., users with visual impairment), various positioning techniques, and navigation instructions algorithms. The simulation engine uses the Unity game engine, which tracks the virtual user interaction and motion in a virtual environment that represents the physical environment in which the user navigates. The framework includes the following modules that will be defined by the indoor navigation system developers: 1) Positioning module which simulates indoor localization techniques and their observed accuracy; 2) Indoor navigation algorithm module that generates the navigation instructions using natural language phrases, and 3) Virtual user model (VUM) that includes human perception and information processing that can understand navigation instructions, perceive the surrounding environment, and act based on this information. The framework also includes a performance analysis module that evaluates the indoor navigation system performance in terms of navigation success rate and route similarity.

An algorithm to generate short sentences in natural language from linked open data based on linguistic templates

The generation of natural language phrases from Linked Open Data can benefit from a significant amount of information available on the internet, as well as from the existence of properties within them, which appears, mostly, in the RDF format. These properties can represent semantic relationships between concepts that might help in creating sentences in natural language. Nevertheless, research in this field tends not to use the information in RDF. We support that this is a factor that might foster the generation of more natural phrases. In this scenario, this research explores these RDF properties for the generation of natural language phrases. The short sentences generated by the algorithm implementation were evaluated regarding their fluency by linguists and native English speakers. The results show that the sentences generated are promising regarding sentence fluency.

An algorithm to generate short sentences in natural language from linked open data based on linguistic templates

The generation of natural language phrases from Linked Open Data can benefit from a significant amount of information available on the internet, as well as from the existence of properties within them, which appears, mostly, in the RDF format. These properties can represent semantic relationships between concepts that might help in creating sentences in natural language. Nevertheless, research in this field tends not to use the information in RDF. We support that this is a factor that might foster the generation of more natural phrases. In this scenario, this research explores these RDF properties for the generation of natural language phrases. The short sentences generated by the algorithm implementation were evaluated regarding their fluency by linguists and native English speakers. The results show that the sentences generated are promising regarding sentence fluency.

Analysis of Linguistic Methods of Informational Impact on Human Consciousness

Modern mass media give the opportunity to use a variety of technologies of consciousness manipulation, but it is not important in what way or method this manipulation takes place, it is important what information needs to be conveyed to the ideologists of the organization. It is fundamentally important that almost every technology of mind manipulation creates its own image of the enemy, which provokes all the cruelty and aggression of terrorist organization members. The authors analyze various linguistic methods for detecting information influence on human consciousness. The paper shows that the classical approach to determining the types of speech actions is very convenient. However, this approach almost completely eludes the real semantics of natural language phrases that carry out speech actions, which can be compensated by the method of using content analysis and the use of databases for the information system.

Constructing Uyghur Commonsense Knowledge Base by Knowledge Projection

Although considerable effort has been devoted to building commonsense knowledge bases (CKB), it is still not available for many low-resource languages such as Uyghur because of expensive construction cost. Focusing on this issue, we proposed a cross-lingual knowledge-projection method to construct an Uyghur CKB by projecting ConceptNet’s Chinese facts into Uyghur. We used a Chinese–Uyghur bilingual dictionary to get high-quality entity translation in facts and employed a back-translation method to eliminate the entity-translation ambiguity. Moreover, to tackle the inner relation ambiguity in translated facts, we made a hand-crafted rule to convert the structured facts into natural-language phrases and built the Chinese–Uyghur lingual phrases based on the similarity of phrases that corresponded to the bilingual semantic similarity scoring model. Experimental results show that the accuracy of our semantic similarity scoring model reached 94.75% for our task, and they successfully project 55,872 Chinese facts into Uyghur as well as obtain 67,375 Uyghur facts within a very short period.

Use of Health Technologies for Low Back Pain in the Military Health System

Low back pain (LBP) is a widespread condition that poses specific occupational risks for active-duty military personnel. Barriers to treatment are also greater for military personnel and veterans. The authors conducted a review of the literature to identify technology solutions to help offset barriers to treatment for military personnel and veterans. Researchers conducted a PubMed search using subject headings and natural language terms and phrases. Three research psychologists screened abstracts and full text articles. After application of exclusion criteria, 28 key articles on treatment of LBP leveraging technology were included. There is evidence in medical literature for the efficacy of health technology in the treatment of LBP. Telehealth can be effective in reducing pain levels, and more convenient and cost-effective than treatment as usual (TAU). Smartphone and mobile apps are effective for managing LBP by providing self-management tools and tracking methods. Internet-based programs are highly accessible, cost-effective, and less stigmatized than TAU. Wearable technologies, such as biofeedback and electronic sensors, are effective for reducing pain and improve physical rehabilitation assessment and treatment for chronic LBP. Findings for emerging technologies are reviewed and discussed. Considerable evidence exists for incorporating technology into treatment programs for LBP. The most promising technologies were as follows: telehealth, internet-based programs, and mobile apps. Benefits of using technology in treatment programs included not only reduction in pain symptoms, but also enabling self-management, allowing anonymous participation, portability of treatment, and cost effectiveness.

Understanding uncertain information in vocal description for creating virtual spatial maps

Abstract Assistive robots are developed for supporting daily activities of elderly people to uplift the living standards. The assistive robots should be friendly, reliable, active, and comprehensible in order to satisfy the needs of elderly population. Human activities are frequently related to navigational tasks and human tend to use descriptions which include natural language phrases and uncertain terms such as “near”, “little”, “far”, “small”, “large”, “close”to describe about spatial information. Therefore assistive robots should be capable of analysing and understanding descriptions which contain natural language phrases with uncertain terms and creating a conceptual map for effective navigation. This paper proposes a method to understand spatial information in a description with uncertain terms and creates a conceptual map in a robot memory which can be linked with spatial map for purposeful, effective and human friendly navigation task. Human studies have been carried out to study different types of descriptions related to navigation tasks. The Virtual Spatial Data Identifier (VSDI) and Uncertain Term Identifier (UTI) modules have been introduced in order to evaluate the spatial information in description to create a virtual map. Results of the system have been compared with the results of a human study in order to evaluate performance of the proposed system.