Vocabulary Keywords Research Articles

Enabling Spatial Data Interoperability through the Use of a Semantic Meta-Model—The Peatland Example from the JRC SEPLA Project

Numerous geographic data on peatland exist but definitions vary, and the correspondent classes are often neither harmonized nor interoperable. This hinders the efforts to employ the available national datasets on peatlands and wetlands for policy monitoring and reporting. The existing meta-languages, such as ISO-Land Cover Meta Language (LCML) and EAGLE, offer the possibility to “deconstruct” the relevant nomenclatures in an object-oriented manner, allowing the comparability and interoperable use of related information. The complex nature of peatlands calls for a dedicated and structured vocabulary of keywords and terms, comprising the biotic substrate and the soil. In the SEPLA project, a semantic meta-model has been developed, combining the hierarchical ontology of the LCML with the matrix structure of the EAGLE model. The necessary elements were provided to describe peatland bio-physical characteristics, while representing the definitions in a concise and user-friendly manner (semantic passports). The proposed semantic meta-model is innovative as it enables the documentation of the spatial distribution of peatland characteristics, considering also their temporal dimension, their intrinsic relation with land use, and the soil. It has been successfully implemented for the translation of the national peatland nomenclature into common land categories relevant for reporting under LULUCF regulation, as part of the EU Climate Law.

Improved Small-Footprint ASR-Based Solution for Open Vocabulary Keyword Spotting

Improved Small-Footprint ASR-Based Solution for Open Vocabulary Keyword Spotting

Max Liboiron, Pollution is Colonialism

oiron Pollution is Colonialism Durham, North Carolina: Duke University Press, 2021 ISBN: 978-1-4780-1413-3 (PB) $24.95. 214 pp. Amidst growing discussions about how to mitigate the current climate and pollution crisis, Max Liboiron presents a stunning critique of colonial practices in Western scientific research methodologies. Throughout the work Pollution is Colonialism, Max Liboiron, founder of the Civic Laboratory for Environmental Action Research (CLEAR) Memorial University in Newfoundland, argues three interrelated ideas. First, settler science supports violent structures of colonialism through assuming access to Indigenous land without permission and perceiving the environment as a manageable waste sink. Scientific research, even if well-intentioned, can reproduce colonial structures by working to manage, not eliminate, industrial toxicants. Second, anticolonial science is possible through specific, contextual and place-based methods which attend to scientists' obligations to their relations. Anticolonial science is community-oriented and respects Indigenous traditions, claims to the land and the local right to refuse a scientific study; the local community should own the results of research. Third, 'methodologies - whether scientific, writerly, readerly or otherwise - are always already part of Land relations and thus are a key site in which to enact good relations (sometimes called ethics)' (pp 6-7). Liboiron's work challenges management-based practices towards the environment and pollution, the colonial assumption that researchers own research, and the settler practice of conducting research in places they never were never granted permission to enter. Liboiron's demonstration of anticolonial praxis begins in the acknowledgements. These acknowledge that the text was written on the ancestral homeland of the Beothuk, that the island of Newfoundland is the ancestral homeland of the Mi'kmaq and Beothuk, and recognise 'the Inuit of Nunatsiavut and NunatuKavut and the Innu of Nitassinan, and their ancestors, as the original people of Labrador' (p. vii). Liboiron then details personal guiding relations of family and genealogy, the ethic of gratitude and reciprocity enacted through footnotes, and those whose presence and advice made the book possible. The acknowledgements are a clear demonstration of the book's guiding ethic: knowledge is not a thing that is owned, but a relationship that is shared. What must be acknowledged is not only sources of funding and a researcher's support structure, but on whose Land the research was conducted. The first two chapters, 'Land, Nature, Resource, Property' and 'Scale, Harm, Violence, Land', generate a keyword vocabulary to describe colonial science's relationship to Indigenous Land. This review uses Liboiron's capitalisation to denote such

End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations

End-to-End Open Vocabulary Keyword Search With Multilingual Neural Representations

The Role of Metadata and Vocabulary Standards in Enabling Scientific Data Interoperability: A Study of Earth System Science Data Facilities

Objective: Journal publishers within many sciences are increasingly expecting data to be deposited into repositories that support the FAIR principles. Data repositories are thus needing to determine what implications the FAIR principles have on their existing services and systems. Metadata standards and controlled vocabularies are specifically called out as core components of the FAIR principles related to interoperability.Methods: This paper looks specifically at the ways that metadata standards and controlled vocabularies are used by Earth system science data repositories. Data sets from 55 data facilities were examined to determine which metadata standards and controlled subject / keyword vocabularies were used. Results: The findings indicate that only the ISO 19115:2003 and DataCite metadata standards are used by more than 40% of the data facilities, and the NASA Global Change Master Directory (GCMD) keywords are the only keyword vocabulary of broad use within this community. Conclusions: These findings raise questions about the extent to which metadata standards and keyword vocabularies can facilitate interoperability beyond narrow sub-sections of the data facility communities. This study also points to systematic challenges related to migration to new standards.

Keyword-augmented and semi-automatic generation of FESS reports: a proof-of-concept study

IntroductionSurgical reports are usually written after a procedure and must often be reproduced from memory. Thus, this is an error-prone, and time-consuming task which increases the workload of physicians. In this proof-of-concept study, we developed and evaluated a software tool using Artificial Intelligence (AI) for semi-automatic intraoperative generation of surgical reports for functional endoscopic sinus surgery (FESS).Materials and methodsA vocabulary of keywords for developing a neural language model was created. With an encoder-decoder-architecture, artificially coherent sentence structures, as they would be expected in general operation reports, were generated. A first set of 48 conventional operation reports were used for model training. After training, the reports were generated again and compared to those before training. Established metrics were used to measure optimization of the model objectively. A cohort of 16 physicians corrected and evaluated three randomly selected, generated reports in four categories: “quality of the generated operation reports,” “time-saving,” “clinical benefits” and “comparison with the conventional reports.” The corrections of the generated reports were counted and categorized.ResultsObjective parameters showed improvement in performance after training the language model (p < 0.001). 27.78% estimated a timesaving of 1–15 and 61.11% of 16–30 min per day. 66.66% claimed to see a clinical benefit and 61.11% a relevant workload reduction. Similarity in content between generated and conventional reports was seen by 33.33%, similarity in form by 27.78%. 66.67% would use this tool in the future. An average of 23.25 ± 12.5 corrections was needed for a subjectively appropriate surgery report.ConclusionThe results indicate existing limitations of applying deep learning to text generation of operation reports and show a high acceptance by the physicians. By taking over this time-consuming task, the tool could reduce workload, optimize clinical workflows and improve the quality of patient care. Further training of the language model is needed.

Infertility and subsequent recurrent miscarriage: Current state of the literature and future considerations for practice and research

Background: Recurrent miscarriage (RM) and infertility are independently associated with adverse pregnancy outcomes, in addition to psychological sequelae. Experiencing pregnancy loss alongside infertility is particularly difficult. International guidance regarding RM is conflicting, and applicability to women with infertility is undetermined. The aim of this study was to: (i) establish if women/couples with a history of infertility are recognised in the literature on the investigation and management of RM, and (ii) determine if the specific needs of women/couples experiencing RM and infertility are ascertained and incorporated into clinical management strategies. Methods: We examined the wide-ranging literature to ascertain what gaps existed. Studies were retrieved through searches of PubMed and Google Scholar up to 21 January 2021 using appropriate controlled vocabulary and combinations of key words. No language or study design restrictions were applied. Results: While women/couples experiencing RM after infertility appear in studies evaluating investigations and proposed treatments, high-quality studies are lacking. Furthermore, they are largely excluded from international clinical guidance and qualitative research. Conclusions: The experiences of women/couples with RM and infertility and their specific care needs within maternity and fertility services are underexplored. It is unclear from current RM guidelines how best to manage and support this complex cohort. Women/couples with infertility and RM are underserved in the literature and in clinical guidance. Further robust studies are warranted to examine pregnancy outcomes, investigations and treatments currently used. Qualitative research is also required to identify their medical and psychological needs to better support this vulnerable group.

Spatial position constraint for unsupervised learning of speech representations.

The success of supervised learning techniques for automatic speech processing does not always extend to problems with limited annotated speech. Unsupervised representation learning aims at utilizing unlabelled data to learn a transformation that makes speech easily distinguishable for classification tasks, whereby deep auto-encoder variants have been most successful in finding such representations. This paper proposes a novel mechanism to incorporate geometric position of speech samples within the global structure of an unlabelled feature set. Regression to the geometric position is also added as an additional constraint for the representation learning auto-encoder. The representation learnt by the proposed model has been evaluated over a supervised classification task for limited vocabulary keyword spotting, with the proposed representation outperforming the commonly used cepstral features by about 9% in terms of classification accuracy, despite using a limited amount of labels during supervision. Furthermore, a small keyword dataset has been collected for Kadazan, an indigenous, low-resourced Southeast Asian language. Analysis for the Kadazan dataset also confirms the superiority of the proposed representation for limited annotation. The results are significant as they confirm that the proposed method can learn unsupervised speech representations effectively for classification tasks with scarce labelled data.

Advancements in urban logistics toward smart, sustainable reforms in developing enabling technologies and markets

PurposeThe research article's primary purpose is to understand the advancements in urban logistics and allied fields over time along with a consideration of its enabling technologies.Design/methodology/ApproachAn initial review is used to build a keyword vocabulary, combinations of which were then applied to the Scopus, ScienceDirect, Emerald Insights, the Web of Science (WOS), Elsevier, Taylor and Francis, Wiley, Inderscience, Springer, Google Scholar and IEEE Xplore for extracting academic publication collection. The first part includes bibliometric analysis; network analysis is done based on the finally selected 645 papers (only those articles include either of the keywords mentioned above in title, abstract, and keywords). The second part conducts a review of the existing literature review studies (only 21 literature review studies out of 645 articles). The last one discusses the advancement in the topics based on the selected research articles.FindingsThis research discussed the advancement of the urban logistics and allied field, key academic forums and key researchers. It is evident from the analysis that the research related to key emerging themes like implementing innovative concepts and sustainability; application of green technologies; data collection, visualization, monitoring and sharing; and automatic logistic systems are still in the nascent stage. However, these research areas gained momentum in the recent past.Research limitationsUrban logistics are essential and play a crucial role for such rapidly growing cities to function. Despite playing a vital role, urban ecosystem logistics is often neglected in formal urban planning. Hence, as a response to customer and business demand, private entities regularly invest in new technologies and solutions. Since such investments are toward profits, various environmental, social and economic challenges arise.Originality/valueThis research investigates the advancements in urban logistics toward smart, sustainable reforms in developing enabling technologies and markets. The obtained research articles are subjected to bibliometric, descriptive, network and content analysis to present a rundown of advancements, relationships and trends in emerging research gaps.

Universidades en Google: hacia un modelo de análisis multinivel del posicionamiento web académico

Se propone un modelo de análisis del posicionamiento web de universidades basado en un vocabulario de palabras clave categorizadas según las distintas misiones universitarias, que se aplica a una universidad (Universitat Politècnica de València) para comprobar su idoneidad. A partir de un vocabulario de 164 palabras clave se construyeron 290 consultas web que fueron ejecutadas en Google, recopilando los 20 primeros resultados obtenidos para cada consulta. Los resultados confirman que las universidades obtienen un posicionamiento web variable en función de la dimensión vinculada a la consulta web y que las páginas web vinculadas a la docencia (especialmente Grados) son las que mejor posicionan, incluso para consultas web orientadas a investigación. Con todo, se observa un posicionamiento bajo no sólo para la UPV sino para las universidades públicas presenciales españolas (sólo el 27% del total de resultados en el Top 20 corresponde a alguna de estas universidades). Se concluye que el análisis multinivel es necesario para estudiar el posicionamiento web de las universidades y que el modelo propuesto es viable y escalable. No obstante, se han identificado ciertas limitaciones (dependencia del vocabulario utilizado y alta variabilidad de datos) que deben tenerse en cuenta en el diseño de este tipo de modelos de análisis.

How to Identify Hot Topics in Psychology Using Topic Modeling

Abstract. Latent topics and trends in psychological publications were examined to identify hotspots in psychology. Topic modeling was contrasted with a classification-based scientometric approach in order to demonstrate the benefits of the former. Specifically, the psychological publication output in the German-speaking countries containing German- and English-language publications from 1980 to 2016 documented in the PSYNDEX database was analyzed. Topic modeling based on latent Dirichlet allocation (LDA) was applied to a corpus of 314,573 publications. Input for topic modeling was the controlled terms of the publications, that is, a standardized vocabulary of keywords in psychology. Based on these controlled terms, 500 topics were determined and trending topics were identified. Hot topics, indicated by the highest increasing trends in this data, were facets of neuropsychology, online therapy, cross-cultural aspects, traumatization, and visual attention. In conclusion, the findings indicate that topics can reveal more detailed insights into research trends than standardized classifications. Possible applications of this method, limitations, and implications for research synthesis are discussed.

Emphasizing Essential Words for Sentiment Classification Based on Recurrent Neural Networks

With the explosion of online communication and publication, texts become obtainable via forums, chat messages, blogs, book reviews and movie reviews. Usually, these texts are much short and noisy without sufficient statistical signals and enough information for a good semantic analysis. Traditional natural language processing methods such as Bow-of-Word (BOW) based probabilistic latent semantic models fail to achieve high performance due to the short text environment. Recent researches have focused on the correlations between words, i.e., term dependencies, which could be helpful for mining latent semantics hidden in short texts and help people to understand them. Long short-term memory (LSTM) network can capture term dependencies and is able to remember the information for long periods of time. LSTM has been widely used and has obtained promising results in variants of problems of understanding latent semantics of texts. At the same time, by analyzing the texts, we find that a number of keywords contribute greatly to the semantics of the texts. In this paper, we establish a keyword vocabulary and propose an LSTM-based model that is sensitive to the words in the vocabulary; hence, the keywords leverage the semantics of the full document. The proposed model is evaluated in a short-text sentiment analysis task on two datasets: IMDB and SemEval-2016, respectively. Experimental results demonstrate that our model outperforms the baseline LSTM by 1%~2% in terms of accuracy and is effective with significant performance enhancement over several non-recurrent neural network latent semantic models (especially in dealing with short texts). We also incorporate the idea into a variant of LSTM named the gated recurrent unit (GRU) model and achieve good performance, which proves that our method is general enough to improve different deep learning models.

Projection of gene-protein networks to the functional space of the proteome and its application to analysis of organism complexity

We consider the problem of biological complexity via a projection of protein-coding genes of complex organisms onto the functional space of the proteome. The latter can be defined as a set of all functions committed by proteins of an organism. Alternative splicing (AS) allows an organism to generate diverse mature RNA transcripts from a single mRNA strand and thus it could be one of the key mechanisms of increasing of functional complexity of the organism's proteome and a driving force of biological evolution. Thus, the projection of transcription units (TU) and alternative splice-variant (SV) forms onto proteome functional space could generate new types of relational networks (e.g. SV-protein function networks, SFN) and lead to discoveries of novel evolutionarily conservative functional modules. Such types of networks might provide new reliable characteristics of organism complexity and a better understanding of the evolutionary integration and plasticity of interconnection of genome-transcriptome-proteome functions.ResultsWe use the InterPro and UniProt databases to attribute descriptive features (keywords) to protein sequences. UniProt database includes a controlled and curated vocabulary of specific descriptors or keywords. The keywords have been assigned to a protein sequence via conserved domains or via similarity with annotated sequences. Then we consider the unique combinations of keywords as the protein functional labels (FL), which characterize the biological functions of the given protein and construct the contingency tables and graphs providing the projections of transcription units (TU) and alternative splice-variants (SV) onto all FL of the proteome of a given organism. We constructed SFNs for organisms with different evolutionary history and levels of complexity, and performed detailed statistical parameterization of the networks.ConclusionsThe application of the algorithm to organisms with different evolutionary history and level of biological complexity (nematode, fruit fly, vertebrata) reveals that the parameters describing SFN correlate with the complexity of a given organism. Using statistical analysis of the links of the functional networks, we propose new features of evolution of protein function acquisition. We reveal a group of genes and corresponding functions, which could be attributed to an early conservative part of the cellular machinery essential for cell viability and survival. We identify and provide characteristics of functional switches in the polyform group of TUs in different organisms. Based on comparison of mouse and human SFNs, a role of alternative splicing as a necessary source of evolution towards more complex organisms is demonstrated.The entire set of FL across many organisms could be used as a draft of the catalogue of the functional space of the proteome world.

Identifying hypothetical genetic influences on complex disease phenotypes

BackgroundStatistical interactions between disease-associated loci of complex genetic diseases suggest that genes from these regions are involved in a common mechanism impacting, or impacted by, the disease. The computational problem we address is to discover relationships among genes from these interacting regions that may explain the observed statistical interaction and the role of these genes in the disease phenotype.ResultsWe describe a heuristic algorithm for generating hypothetical gene relationships from loci associated with a complex disease phenotype. This approach, called Prioritizing Disease Genes by Analysis of Common Elements (PDG-ACE), mines biomedical keywords from text descriptions of genes and uses them to relate genes close to disease-associated loci. A keyword common to, and significantly over-represented in, a pair of gene descriptions may represent a preliminary hypothesis about the biological relationship between the genes, and suggest the role the genes play in the disease phenotype.ConclusionOur experimentation shows that the approach finds previously published relationships, while failing to find relationships that don't exist. The results also indicate that the approach is robust to differences in keyword vocabulary. We outline a brief case study in which results from a recently published Type 2 Diabetes association study are used to identify potential hypotheses.

Integrated patch model: A generative model for image categorization based on feature selection

Image categorization could be treated as an effective solution to enable keyword-based image retrieval. In this paper, we propose a novel image categorization approach by learning semantic concepts of image categories. In order to choose representative features and meanwhile reduce noisy features, a three-step feature selection strategy is proposed. First, salient patches are detected. Then all the detected salient patches are clustered and the visual keyword vocabulary is constructed. Finally, the region of dominance and the salient entropy measure are calculated to reduce the similar and non-common noises of salient patches. Based on the selected visual keywords, the Integrated Patch (IP) model is proposed to describe and categorize images. As a generative model, the IP model represents the appearance of the combination of the visual keywords, considering the diversity of the object or the scene. The parameters are estimated by the EM algorithm. The experimental results on the Corel image dataset demonstrate that the proposed feature selection and the image description model are effective in image categorization.

Computer-aided Document Indexing System

An enormous number of documents is being produced that have to be stored, searched and accessed. Document indexing represents an efficient way to tackle this problem. Contributing to the document indexing process, we developed the Computer-Aided Document Indexing System (CADIS) that applies controlled vocabulary keywords from the EUROVOC thesaurus. The main contribution of this paper is the introduction of the special CADIS internal data structure that copes with the morphological complexity of the Croatian language. CADIS internal data structure ensures efficient statistical analysis of input documents and quick visual feedback generation that helps indexing documents more quickly, accurately and uniformly than by manual indexing.

Broadcast speech recognition system for keyword monitoring

Broadcast audio (10) is automatically monitored for information of interest. A computer processor (12) has a memory associated therewith for storing a vocabulary (24) of keywords of interest. An audio receiver receives an audio broadcast. A voice recognition system (12, 22) associated with the audio receiver and computer processor detects (26) when one of the keywords appears in a received audio segment. A report generator (14) associated with the computer processor and responsive to the detection of a keyword generates a report detailing the detected keyword and its context. The system is particularly well suited to the verification of commercial messages and to assist in public relations efforts.

Subsyllable-based discriminative segmental Bayesian network for Mandarin speech keyword spotting

A continuous Mandarin speech keyword spotting system based on context-dependent subsyllables is presented. In this vocabulary-independent system, users can define their own keywords and most frequently occurring non-keywords without retraining the system. A set of 176 monosyllables and 483 balanced words or sentences are used to extract the context-dependent subsyllables (i.e. initials or finals in Mandarin speech), for training. Each subsyllable is represented by a proposed discriminative segmental Bayesian network (DSBN). In the training process, the generalised probabilistic descent (GPD) algorithm is used for discriminative training. The most frequently occurring non-keywords are divided into keyword predecessors and successors. Non-keyword garbage models for keyword predecessors, keyword successors and extraneous speech are separately constructed. In the recognition process, a final part preprocessor is used to screen out unreasonable hypotheses in order to reduce the recognition time. Using a test set of 750 conversational speech utterances from 20 speakers (ten males and ten females), word spotting rates of 92.0% when the vocabulary word was embedded in unconstrained extraneous speech, were obtained for a user-defined 20 keyword vocabulary.

A fast algorithm for large vocabulary keyword spotting application

Presents a fast algorithm for spotting a large number of keywords in unconstrained, continuous speech using an HMM-based continuous speech recognizer. This fast algorithm is based on a two stage scheme. In the first stage, the forward backward search is performed for detecting N most likely common subwords. In the second stage, the tree-trellis search is carried out to determine the optimum keyword by traversing the tree-structural vocabulary in an effective way. Compared with the conventional whole-word based keyword spotting algorithm, the proposed fast algorithm can drastically reduce the computational cost. >

A Residency-based Information System

The medical literature has grown to an unmanageable size, and we need to develop information systems that provide better accessibility. An internal medicine residency program is an ideal setting for developing a system in which articles are screened for usefulness in clinical decision making and organized to be rapidly accessible and relatively inexpensive to use. In our residency training program, we use clinical problem-solving, journal club article review, and teaching conference preparation to generate appropriate articles from PAPERCHASE (MEDLINE) database searches. With the chief resident supervising the process, articles are then selected and organized in a computerized article file. The software program, Reference Manager (Research Information Systems, Inc., Encinitas, California), stores article citations and supports searches using standard, medical subject heading vocabulary key words. In an 18-month period, residents have collected more than 1800 references appropriate to clinical decision making and have established an article file that is now used daily as a medical information source.