122 research outputs found
Turkish sentiment analysis: a comparative study on different sentiment dictionaries with generated features and presenting a new sentiment dictionary
No full textSentiment analysis is a research area that aims to find out people\"s opinions by matching data to topics, notions, etc. There are several approaches for sentiment analysis (e.g., machine learning-based, dictionary-based, hybrid-based, etc.). In this study, we presented a new tripolar Turkish sentiment dictionary, SentiMenTR, which consists of bigrams and unigrams. To compare the performances of SentiMenTR and other Turkish sentiment dictionaries (SWNetTR++ and SentiTurkNet), we conducted experiments on two Turkish datasets containing documents labeled as negative or positive. For experiments, firstly, we vectorized the documents by features extracted using polarity scores belonging to dictionaries. Afterward, we fitted machine learning models with these features. According to the experiment results, SentiMenTR performed better than other dictionaries. We aim to extend our dictionary, develop a negation handler module, and conduct more comprehensive experiments with deep learning methods in the future
A multicenter survey of childhood asthma in Turkey - II: Utilization of asthma drugs, control levels and their determinants
No full textPubMedID: 18823358Soyer OU, Beyhun NE, Demir E, Yildirim S, Bingöl Boz A, Altinel N, Cevit O, Karakaş T, Anlar Y, Söüt A, Altintaş D, Canitez Y, Büyükdereli Z and Sekerel BE. A multicenter survey of childhood asthma in Turkey - II: Utilization of asthma drugs, control levels and their determinants. Pediatr Allergy Immunol 2009: 20: 172-179. © 2008 The Authors Journal compilation © 2008 Blackwell Munksgaard Many surveys worldwide have consistently demonstrated a low level of asthma control and under-utilization of preventive asthma drugs. However, these studies have been frequently criticized for using population-based samples, which include many patients with no or irregular follow-ups. Our aim, in this study, was to define the extent of asthma drug utilization, control levels, and their determinants among children with asthma attending to pediatric asthma centers in Turkey. Asthmatic children (age range: 6-18 yr) with at least 1-yr follow-up seen at 12 asthma outpatient clinics during a 1-month period with scheduled or unscheduled visits were included and were surveyed with a questionnaire-guided interview. Files from the previous year were evaluated retrospectively to document control levels and their determinants. From 618 children allocated, most were mild asthmatics (85.6%). Almost 30% and 15% of children reported current use of emergency service and hospitalization, respectively; and 51.4% and 53.1% of children with persistent and intermittent disease, respectively, were on daily preventive therapy, including inhaled corticosteroids. Disease severity [odds ratio: 12.6 (95% confidence intervals: 5.3-29.8)], hospitalization within the last year [3.4 (1.4-8.2)], no use of inhaled steroids [2.9 (1.1- 7.3)], and female gender [2.3 (1.1-5.4)] were major predictors of poor asthma control as defined by their physicians. In this national pediatric asthma study, we found a low level of disease control and discrepancies between preventive drug usage and disease severity, which shows that the expectations of guidelines have not been met even in facilitated centers, thus indicating the need to revise the severity-based approach of asthma guidelines. Efforts to implement the control-based approach of new guidelines (Global Initiative for Asthma 2006) would be worthwhile. © 2008 Blackwell Munksgaard
Conocimiento, creencias y prácticas sobre cáncer testicular y su autoexamen en estudiantes universitarios de Nuevo León.
No full textEl Cáncer Testicular (CT) representa el principal tipo de tumor que afecta al sexo masculino en la juventud (15 a 40 años aproximadamente). El Autoexamen Testicular (AT) es una técnica recomendada para hombres en este rango de edad que puede llevar a la detección oportuna del CT, sin embargo según la literatura científica alrededor del mundo, el conocimiento sobre esta técnica y su aplicación son considerablemente bajos.
En este estudio se exploró el nivel de conocimiento y actitudes sobre el CT y AT en estudiantes universitarios, esto validando y después utilizando la escala del modelo de creencias en salud para cáncer testicular desarrollada por Altinel y Avci, la cantidad de participantes fue de 310 estudiantes universitarios (Media de edad= 19.8, DE=2.8) quienes fueron seleccionados a conveniencia para esta investigación. Se encontraron diferencias estadísticamente significativas entre los estudiantes que sabían realizar el AT (19.7%) y los que nó (80.3%) en cuanto a sus respuestas en las dimensiones de Autoeficacia, Barreras y Severidad percibida (p=<.05), los estudiantes que sí conocían el AT, reportaron más autoeficacia, menos barreras y menos severidad, mientras que quienes realizaban el AT mostraron además mayores beneficios percibidos que los que no lo hacían. Por ultimo se hizo énfasis en la validación del instrumento ya mencionado, el cual reportó una buena consistencia interna en cada una de sus dimensiones(alfas de cronbach superiores a .790), además tras un análisis factorial exploratorio se encontraron 6 factores relacionados y se decidió suprimir ciertos items con baja carga factorial para mejorar la consistencia de la escala. En suma el conocimiento sobre el AT encontrado en los sujetos es bajo, pero se encuentra dentro del rango de los países en desarrollo, las diferencias entre sujetos que sabían o no efectuar el AT son consistentes con otros estudios y la validación de la Escala de Creencias de Salud para el Autoexamen de Detección de Cáncer Testicular representa una mayor facilidad para futuros estudios similares en México
A novel semantic smoothing kernel for text classification with class-based weighting
No full textAltınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author) -- #articleinpress#Altınel, Berna (Dogus Author), Diri, Banu (Dogus Author), Ganiz, Murat Can (Dogus Author)In this study, we propose a novel methodology to build a semantic smoothing kernel to use with Support Vector Machines (SVM) for text classification. The suggested approach is based on two key concepts; class-based term weighting and changing the orthogonality of vector space. A class-based term weighting methodology is used for transformation of documents from the original space to the feature space. This class-based weighting basically groups terms based on their importance for each class and consequently smooths the representation of documents. This is accomplished by changing the orthogonality of the Vector Space Model (VSM) with introducing class-based dependencies between terms. As a result, on the extreme case, two documents can be seen as similar even if they do not share any terms but their terms are similarly weighted for a particular class. The resulting semantic kernel can directly make use of class information in extracting semantic information between terms, therefore it can be considered as a supervised kernel. For our experimental evaluation, we analyze the performance of the suggested kernel with a large number of experiments on benchmark textual datasets and present results with respect to varying experimental conditions. To the best of our knowledge, this is the first study to use class-based term weighting in order to build a supervised semantic kernel for SVM. We compare our results with kernels that are commonly used in SVM such as linear kernel, polynomial kernel, Radial Basis Function (RBF) kernel and with several corpus-based semantic kernels. According to our experimental results the proposed method favorably improves classification accuracy over linear kernel and several corpus-based semantic kernels in terms of both accuracy and speed
Artificial intelligence-based prediction of transfusion in the intensive care unit in patients with gastrointestinal bleeding
No full textObjective Gastrointestinal (GI) bleeding commonly requires intensive care unit (ICU) in cases of potentialhaemodynamiccompromise or likely urgent intervention. However, manypatientsadmitted to the ICU stop bleeding and do not require further intervention, including blood transfusion. The present work proposes an artificial intelligence (AI) solution for the prediction of rebleeding in patients with GI bleeding admitted to ICU. Methods A machine learning algorithm was trained and tested using two publicly available ICU databases, the Medical Information Mart for Intensive Care V.1.4 database and eICU Collaborative Research Database using freedom from transfusion as a proxy for patients who potentially did not require ICU-level care. Multiple initial observation time frames were explored using readily available data including labs, demographics and clinical parameters for a total of 20 covariates. Results The optimal model used a 5-hour observation period to achieve an area under the curve of the receiving operating curve (ROC-AUC) of greater than 0.80. The model was robust when tested against both ICU databases with a similar ROC-AUC for all. Conclusions The potential disruptive impact of AI in healthcare innovation is acknowledge, but awareness of AI-related risk on healthcare applications and current limitations should be considered before implementation and deployment. The proposed algorithm is not meant to replace but to inform clinical decision making. Prospective clinical trial validation as a triage tool is warranted
SMOQE: A System for Providing Secure Access to XML
No full textXML views have been widely used to enforce access control, support dataintegration, and speed up query answering. In many applications, e.g., XMLsecurity enforcement, it is prohibitively expensive to materialize and maintaina large number of views. Therefore, views are necessarily virtual. An immediate question then is how to answer queries on XML virtual views. A common approach is to rewrite a query on the view to an equivalent one on the underlying document, and evaluate the rewritten query. This is the approach used in the Secure MOdular Query Engine (SMOQE). The demo presents SMOQE, the first system to provide efficient support for answering queries over virtual and possibly recursively defined XML views. We demonstrate a set of novel techniques for the specification of views, therewriting, evaluation and optimization of XML queries. Moreover, we provideinsights into the internals of the engine by a set of visual tools
Efficient Distribution-Based Event Filtering
No full textEvent notification services are used in various applications, for example, stock tickers, environmental monitoring, and facility management. Several filtering algorithms for such services have been proposed. The best performance results are achieved by tree-based algorithms. However, to our knowledge existing algorithms do not consider the influence of event and profile distribution on the filter performance. In this paper we propose a distribution-dependent improvement of the tree-algorithm. We present the test results of our prototypical implementation that show the influence of various distribution-based measures on the performance
A novel extremophilic xylanase produced on wheat bran from Aureobasidium pullulans NRRL Y-2311-1: Effects on dough rheology and bread quality
No full textAn extremophilic xylanase from Aureobasidium pullulans NRRL Y-2311-1 was produced on wheat bran and its performance in bread making was investigated for the first time. Two different world-wide-used commercial xylanase preparations were also applied in bread making as comparison. Effects of different enzyme dosages on various farinograph and extensograph properties of the dough and bread quality were evaluated in detail. The novel xylanase provided increase in water absorption, development time and stability of the dough and decrease in dough softening degree and mixing tolerance index at a dosage of 100 U/100 g flour. None of the enzymes provided reasonable increase in dough extensibility. There was no direct correlation between the extensograph properties (mainly, resistance and extension) of the bread dough and the bread specific volume. A. pullulans xylanase (125 U/100 g flour) provided remarkable improvement (30%) in bread specific volume as compared to the commercial counterparts. The moisture content values of all the bread samples were within the ideal limits (35-40%). A. pullulans xylanase was the most effective enzyme in decreasing the crumb firmness. Slight improvements in cohesiveness and remarkable decline in springiness and gumminess were observed for all the enzymes tested. The results of this study provide an opportunity for A. pullulans xylanase to be used in bread making at industrial scale. (C) 2018 Elsevier Ltd. All rights reserved.Scientific and Technological Research Council of Turkey-TUBITAKTurkiye Bilimsel ve Teknolojik Arastirma Kurumu (TUBITAK) [TOVAG 112O521]This work was financially supported by The Scientific and Technological Research Council of Turkey-TUBITAK (Project No: TOVAG 112O521). This work is also included in Cost Action FP1306 entitled as "Valorisation of Lignocellulosic Biomass Side Streams for Sustainable Production of Chemicals, Materials & Fuels Using Low Environmental Impact Technologies"
Querying XML data streams from wireless sensor networks: an evaluation of query engines
No full textAs the deployment of wireless sensor networks increase and their application domain widens, the opportunity for effective use of XML filtering and streaming query engines is ever more present. XML filtering engines aim to provide efficient real-time querying of streaming XML encoded data. This paper provides a detailed analysis of several such engines, focusing on the technology involved, their capabilities, their support for XPath and their performance. Our experimental evaluation identifies which filtering engine is best suited to process a given query based on its properties. Such metrics are important in establishing the best approach to filtering XML streams on-the-fly
Distribution of first and last author sex.
No full textArtificial intelligence (AI) and machine learning are central components of today’s medical environment. The fairness of AI, i.e. the ability of AI to be free from bias, has repeatedly come into question. This study investigates the diversity of members of academia whose scholarship poses questions about the fairness of AI. The articles that combine the topics of fairness, artificial intelligence, and medicine were selected from Pubmed, Google Scholar, and Embase using keywords. Eligibility and data extraction from the articles were done manually and cross-checked by another author for accuracy. Articles were selected for further analysis, cleaned, and organized in Microsoft Excel; spatial diagrams were generated using Public Tableau. Additional graphs were generated using Matplotlib and Seaborn. Linear and logistic regressions were conducted using Python to measure the relationship between funding status, number of citations, and the gender demographics of the authorship team. We identified 375 eligible publications, including research and review articles concerning AI and fairness in healthcare. Analysis of the bibliographic data revealed that there is an overrepresentation of authors that are white, male, and are from high-income countries, especially in the roles of first and last author. Additionally, analysis showed that papers whose authors are based in higher-income countries were more likely to be cited more often and published in higher impact journals. These findings highlight the lack of diversity among the authors in the AI fairness community whose work gains the largest readership, potentially compromising the very impartiality that the AI fairness community is working towards.</div
- …
