Search results for: Learning Curve

Authors: Xiaocong Liu, Huazhen Wang, Ting He, Xiaozheng Li, Weihan Zhang, Jian Chen

Abstract:

The utilization of electronic medical record (EMR) data to establish the disease diagnosis model has become an important research content of biomedical informatics. Deep learning can automatically extract features from the massive data, which brings about breakthroughs in the study of EMR data. The challenge is that deep learning lacks semantic knowledge, which leads to impracticability in medical science. This research proposes a method of incorporating lexical-semantic knowledge from abundant entities into a convolutional neural network (CNN) framework for pediatric disease diagnosis. Firstly, medical terms are vectorized into Lexical Semantic Vectors (LSV), which are concatenated with the embedded word vectors of word2vec to enrich the feature representation. Secondly, the semantic distribution of medical terms serves as Semantic Decision Guide (SDG) for the optimization of deep learning models. The study evaluates the performance of LSV-SDG-CNN model on four kinds of Chinese EMR datasets. Additionally, CNN, LSV-CNN, and SDG-CNN are designed as baseline models for comparison. The experimental results show that LSV-SDG-CNN model outperforms baseline models on four kinds of Chinese EMR datasets. The best configuration of the model yielded an F1 score of 86.20%. The results clearly demonstrate that CNN has been effectively guided and optimized by lexical-semantic knowledge, and LSV-SDG-CNN model improves the disease classification accuracy with a clear margin.

Keywords: lexical semantics, feature representation, semantic decision, convolutional neural network, electronic medical record

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 585

Authors: K´evin Fernagut, Olivier Flauzac, Erick M. Gallegos R, Florent Nolot

Abstract:

The adoption of modern lightweight virtualization often comes with new threats and network vulnerabilities. This paper seeks to assess this with a different approach studying the behavior of a testbed built with tools such as Kernel-based Virtual Machine (KVM), LinuX Containers (LXC) and Docker, by performing stress tests within a platform where students experiment simultaneously with cyber-attacks, and thus observe the impact on the campus network and also find the best solution for cyber-security learning. Interesting outcomes can be found in the literature comparing these technologies. It is, however, difficult to find results of the effects on the global network where experiments are carried out. Our work shows that other physical hosts and the faculty network were impacted while performing these trials. The problems found are discussed, as well as security solutions and the adoption of new network policies.

Keywords: Containerization, containers, cyber-security, cyber-attacks, isolation, performance, security, virtualization, virtual machines.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 551

Authors: L. Mortari, A. Bevilacqua, R. Silva

Abstract:

The experience presented in this paper aims to understand how educational research can support the introduction and optimization of teaching innovations in legal education. In this increasingly complex context, a strong need to introduce paths aimed at acquiring not only professional knowledge and skills but also reflective, critical and problem-solving skills emerges. Through a peer observation intertwined with an analysis of discursive practices, researchers and the teacher worked together through a process of participatory and transformative accompaniment whose objective was to promote the active participation and engagement of students in learning processes, an element indispensable to work in the more specific direction of strengthening key competences. This reflective faculty development path led the teacher to activate metacognitive processes, becoming thus aware of the strengths and areas of improvement of his teaching innovation.

Keywords: Discursive analysis, faculty development, legal education, peer observation, teaching innovation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 344

Authors: Sedigheh Mirzaei S., Debasis Sengupta

Abstract:

Parametric models have been quite popular for studying human growth, particularly in relation to biological parameters such as peak size velocity and age at peak size velocity. Longitudinal data are generally considered to be vital for fittinga parametric model to individual-specific data, and for studying the distribution of these biological parameters in a human population. However, cross-sectional data are easier to obtain than longitudinal data. In this paper, we present a method of combining longitudinal and cross-sectional data for the purpose of estimating the distribution of the biological parameters. We demonstrate, through simulations in the special case ofthePreece Baines model, how estimates based on longitudinal data can be improved upon by harnessing the information contained in cross-sectional data.We study the extent of improvement for different mixes of the two types of data, and finally illustrate the use of the method through data collected by the Indian Statistical Institute.

Keywords: Preece-Baines growth model, MCMC method, Mixed effect model

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2135

Authors: Parisa Ahmadi Oliaei, Seyed Abolhassan Naeini

Abstract:

The knowledge of soil–reinforcement interaction parameters is particularly important in the design of reinforced soil structures. The pull-out test is one of the most widely used tests in this regard. The results of tensile tests may be very sensitive to boundary conditions, and more research is needed for a better understanding of the pull-out response of reinforcement, so numerical analysis using the finite element method can be a useful tool for the understanding of the pull-out response of soil-geogrid interaction. The main objective of the present study is to compare the numerical and experimental results of a pull-out test on geogrid-reinforced sandy soils interactions. Plaxis 2D finite element software is used for simulation. In the present study, the pull-out test modeling has been done on sandy soil. The effect of geogrid hardness was also investigated by considering two different types of geogrids. The numerical results curve had a good agreement with the pull-out laboratory results.

Keywords: Plaxis, pull-out test, sand, soil-geogrid interaction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 421

Authors: Hossein Mamusi, Abolfazl Masoumi, Ramezanali Mahdavinezhad

Abstract:

Tailor-welded Blanks (TWBs) are tailor made for different complex component designs by welding multiple metal sheets with different thicknesses, shapes, coatings or strengths prior to forming. In this study the Hemispherical Die Stretching (HDS) test (out-of-plane stretching) of TWBs were simulated via ABAQUS/Explicit to obtain the Forming Limit Diagrams (FLDs) of Stainless steel (AISI 304) laser welded blanks with different thicknesses. Two criteria were used to detect the start of necking to determine the FLD for TWBs and parent sheet metals. These two criteria are the second derivatives of the major and thickness strains that are given from the strain history of simulation. In the other word, in these criteria necking starts when the second derivative of thickness or major strain reaches its maximum. With having the time of onset necking, one can measure the major and minor strains at the critical area and determine the forming limit curve.

Keywords: TWB, Forming Limit Diagram, Necking criteria, ABAQUS/Explicit

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1640

Authors: Y. A. Adla, R. Soubra, M. Kasab, M. O. Diab, A. Chkeir

Abstract:

Over the past several years, researchers have shown a great interest in assessing the mobility of elderly people to measure their functional status. Usually, such an assessment is done by conducting tests that require the subject to walk a certain distance, turn around, and finally sit back down. Consequently, this study aims to provide an at home monitoring system to assess the patient’s status continuously. Thus, we proposed a technique to automatically detect when a subject sits down while walking at home. In this study, we utilized a Doppler radar system to capture the motion of the subjects. More than 20 features were extracted from the radar signals out of which 11 were chosen based on their Intraclass Correlation Coefficient (ICC > 0.75). Accordingly, the sequential floating forward selection wrapper was applied to further narrow down the final feature vector. Finally, five features were introduced to the Linear Discriminant Analysis classifier and an accuracy of 93.75% was achieved as well as a precision and recall of 95% and 90% respectively.

Keywords: Doppler radar system, stand-to-sit phase, TUG test, machine learning, classification

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 443

Authors: B. Bagheri Nakhjavanlo, T. J. Ellis, P. Raoofi, J. Dehmeshki

Abstract:

This paper presents a customized deformable model for the segmentation of abdominal and thoracic aortic aneurysms in CTA datasets. An important challenge in reliably detecting aortic aneurysm is the need to overcome problems associated with intensity inhomogeneities and image noise. Level sets are part of an important class of methods that utilize partial differential equations (PDEs) and have been extensively applied in image segmentation. A Gaussian kernel function in the level set formulation, which extracts the local intensity information, aids the suppression of noise in the extracted regions of interest and then guides the motion of the evolving contour for the detection of weak boundaries. The speed of curve evolution has been significantly improved with a resulting decrease in segmentation time compared with previous implementations of level sets. The results indicate the method is more effective than other approaches in coping with intensity inhomogeneities.

Keywords: Abdominal and thoracic aortic aneurysms, intensityinhomogeneity, level sets, local fitting binary.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1810

Authors: A. C. Sarmah

Abstract:

The Al-MoO3-P-CdTe-Al MOS sandwich structures were fabricated by vacuum deposition method on cleaned glass substrates. Capacitance versus voltage measurements were performed at different frequencies and sweep rates of applied voltages for oxide and semiconductor films of different thicknesses. In the negative voltage region of the C-V curve a high differential capacitance of the semiconductor was observed and at high frequencies (<10 kHz) the transition from accumulation to depletion and further to deep depletion was observed as the voltage was swept from negative to positive. A study have been undertaken to determine the value of acceptor density and some depletion layer parameters such as depletion layer capacitance, depletion width, impurity concentration, flat band voltage, Debye length, flat band capacitance, diffusion or built-in-potential, space charge per unit area etc. These were determined from C-V measurements for different oxide and semiconductor thicknesses.

Keywords: Debye length, Depletion width, flat band capacitance, impurity concentration.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1566

Authors: Mongkorn Klingajay, Tawatchai Jitson

Abstract:

The pipe inspection operation is the difficult detective performance. Almost applications are mainly relies on a manual recognition of defective areas that have carried out detection by an engineer. Therefore, an automation process task becomes a necessary in order to avoid the cost incurred in such a manual process. An automated monitoring method to obtain a complete picture of the sewer condition is proposed in this work. The focus of the research is the automated identification and classification of discontinuities in the internal surface of the pipe. The methodology consists of several processing stages including image segmentation into the potential defect regions and geometrical characteristic features. Automatic recognition and classification of pipe defects are carried out by means of using an artificial neural network technique (ANN) based on Radial Basic Function (RBF). Experiments in a realistic environment have been conducted and results are presented.

Keywords: Artificial neural network, Radial basic function, Curve fitting, CCTV, Image segmentation, Data acquisition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1808

Authors: Khalid Ghoulam, Belaid Bouikhalene, Zakaria Harmouch, Hicham Mouncif

Abstract:

The goal of engineering education is to prepare students to cope with problems of real devices and systems. Usually there are not enough devices or time for conducting experiments in a real lab. Other factors that prevent the use of lab devices directly by students are inaccessible or dangerous phenomena, or polluting chemical reactions. The technology brings additional strategies of learning and teaching, there are two types of online labs, virtual and remote labs RL. We present an example of a successful development and deployment of a remote lab in the field of engineering education, integrated in the Moodle platform, using very low-coast, high documented devices and free software. The remote lab is user friendly for both teachers and students. Our web 2.0 based user interface would attract and motivate students, as well as solving the problem of larger classes and expensive lab devices.

Keywords: Remote lab, online learning, Moodle, Arduino, SMSU, lab experimentation, engineering education, online engineering education.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1353

Authors: R. Sridevi, G. Jagajothi

Abstract:

Artificial Immune Systems (AIS), inspired by the human immune system, are algorithms and mechanisms which are self-adaptive and self-learning classifiers capable of recognizing and classifying by learning, long-term memory and association. Unlike other human system inspired techniques like genetic algorithms and neural networks, AIS includes a range of algorithms modeling on different immune mechanism of the body. In this paper, a mechanism of a human immune system based on apoptosis is adopted to build an Intrusion Detection System (IDS) to protect computer networks. Features are selected from network traffic using Fisher Score. Based on the selected features, the record/connection is classified as either an attack or normal traffic by the proposed methodology. Simulation results demonstrates that the proposed AIS based on apoptosis performs better than existing AIS for intrusion detection.

Keywords: Apoptosis, Artificial Immune System (AIS), Fisher Score, KDD dataset, Network intrusion detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2187

Authors: Revanth Sai Kosaraju, Michael Ramscar, Melody Dye

Abstract:

Accounts of language acquisition differ significantly in their treatment of the role of prediction in language learning. In particular, nativist accounts posit that probabilistic learning about words and word sequences has little to do with how children come to use language. The accuracy of this claim was examined by testing whether distributional probabilities and frequency contributed to how well 3-4 year olds repeat simple word chunks. Corresponding chunks were the same length, expressed similar content, and were all grammatically acceptable, yet the results of the study showed marked differences in performance when overall distributional frequency varied. It was found that a distributional model of language predicted the empirical findings better than a number of other models, replicating earlier findings and showing that children attend to distributional probabilities in an adult corpus. This suggested that language is more prediction-and-error based, rather than on abstract rules which nativist camps suggest.

Keywords: Abstractness, child psychology, language acquisition, prediction and error.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2086

Authors: U. Prasopchingchana

Abstract:

The temperature distribution and the heat transfer rates through a multi-layer door of a furnace were investigated. The inside of the door was in contact with hot air and the other side of the door was in contact with room air. Radiation heat transfer from the walls of the furnace to the door and the door to the surrounding area was included in the problem. This work is a two dimensional steady state problem. The Churchill and Chu correlation was used to find local convection heat transfer coefficients at the surfaces of the furnace door. The thermophysical properties of air were the functions of the temperatures. Polynomial curve fitting for the fluid properties were carried out. Finite difference method was used to discretize for conduction heat transfer within the furnace door. The Gauss-Seidel Iteration was employed to compute the temperature distribution in the door. The temperature distribution in the horizontal mid plane of the furnace door in a two dimensional problem agrees with the one dimensional problem. The local convection heat transfer coefficients at the inside and outside surfaces of the furnace door are exhibited.

Keywords: Conduction, heat transfer, multi-layer door, natural convection

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2090

Authors: Bharatendra Rai

Abstract:

Predictive data analysis and modeling involving machine learning techniques become challenging in presence of too many explanatory variables or features. Presence of too many features in machine learning is known to not only cause algorithms to slow down, but they can also lead to decrease in model prediction accuracy. This study involves housing dataset with 79 quantitative and qualitative features that describe various aspects people consider while buying a new house. Boruta algorithm that supports feature selection using a wrapper approach build around random forest is used in this study. This feature selection process leads to 49 confirmed features which are then used for developing predictive random forest models. The study also explores five different data partitioning ratios and their impact on model accuracy are captured using coefficient of determination (r-square) and root mean square error (rsme).

Keywords: Housing data, feature selection, random forest, Boruta algorithm, root mean square error.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1704

Authors: George Yungchih Wang

Abstract:

This paper argues that increased uncertainty, in certain situations, may actually encourage investment. Since earlier studies mostly base their arguments on the assumption of geometric Brownian motion, the study extends the assumption to alternative stochastic processes, such as mixed diffusion-jump, mean-reverting process, and jump amplitude process. A general approach of Monte Carlo simulation is developed to derive optimal investment trigger for the situation that the closed-form solution could not be readily obtained under the assumption of alternative process. The main finding is that the overall effect of uncertainty on investment is interpreted by the probability of investing, and the relationship appears to be an invested U-shaped curve between uncertainty and investment. The implication is that uncertainty does not always discourage investment even under several sources of uncertainty. Furthermore, high-risk projects are not always dominated by low-risk projects because the high-risk projects may have a positive realization effect on encouraging investment.

Keywords: real options, geometric Brownian motion, mixeddiffusion-jump process, mean- reverting process, jump amplitudeprocess

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1538

Authors: Firas Salih, Luban Hameed, Afaf Kamil, Armin Bolz

Abstract:

Diagnostic and detection of the arterial stiffness is very important; which gives indication of the associated increased risk of cardiovascular diseases. To make a cheap and easy method for general screening technique to avoid the future cardiovascular complexes , due to the rising of the arterial stiffness ; a proposed algorithm depending on photoplethysmogram to be used. The photoplethysmograph signals would be processed in MATLAB. The signal will be filtered, baseline wandering removed, peaks and valleys detected and normalization of the signals should be achieved .The area under the catacrotic phase of the photoplethysmogram pulse curve is calculated using trapezoidal algorithm ; then will used in cooperation with other parameters such as age, height, blood pressure in neural network for arterial stiffness detection. The Neural network were implemented with sensitivity of 80%, accuracy 85% and specificity of 90% were got from the patients data. It is concluded that neural network can detect the arterial STIFFNESS depending on risk factor parameters.

Keywords: Arterial stiffness, area under the catacrotic phase of the photoplethysmograph pulse, neural network

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1645

Authors: Rahib Hidayat Abiyev

Abstract:

This paper presents the development of recurrent neural network based fuzzy inference system for identification and control of dynamic nonlinear plant. The structure and algorithms of fuzzy system based on recurrent neural network are described. To train unknown parameters of the system the supervised learning algorithm is used. As a result of learning, the rules of neuro-fuzzy system are formed. The neuro-fuzzy system is used for the identification and control of nonlinear dynamic plant. The simulation results of identification and control systems based on recurrent neuro-fuzzy network are compared with the simulation results of other neural systems. It is found that the recurrent neuro-fuzzy based system has better performance than the others.

Keywords: Fuzzy logic, neural network, neuro-fuzzy system, control system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2368

Authors: Chonmapat Torasa

Abstract:

This paper presents the performance of electricity generation and consumption from solar generator installed at Rajabhat Suan Sunandha’s learning center in Samutsongkram. The result from the experiment showed that solar cell began to work and distribute the current into the system when the solar energy intensity was 340 w/m2, starting from 8:00 am to 4:00 pm (duration of 8 hours). The highest intensity read during the experiment was 1,051.64w/m2. The solar power was 38.74kWh/day. The electromotive force from solar cell averagely was 93.6V. However, when connecting solar cell with the battery charge controller system, the voltage was dropped to 69.07V. After evaluating the power distribution ability and electricity load of tested solar cell, the result showed that it could generate power to 11 units of 36-watt fluorescent lamp bulbs, which was altogether 396W. In the meantime, the AC to DC power converter generated 3.55A to the load, and gave 781VA.

Keywords: Solar Cell, Solar-cell power generating system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2056

Authors: Siyao Zhu, Yifang Xu

Abstract:

After natural and man-made disasters, robots can bypass the danger, expedite the search, and acquire unprecedented situational awareness to design rescue plans. Brain-computer interface is a promising option to overcome the limitations of tedious manual control and operation of robots in the urgent search-and-rescue tasks. This study aims to test the feasibility of using electroencephalography (EEG) signals to decode human intentions and detect the level of consensus on robot-provided information. EEG signals were classified using machine-learning and deep-learning methods to discriminate search intentions and agreement perceptions. The results show that the average classification accuracy for intention recognition and consensus assessment is 67% and 72%, respectively, proving the potential of incorporating recognizable users’ bioelectrical responses into advanced robot-assisted systems for emergency response.

Keywords: Consensus assessment, electroencephalogram, EEG, emergency response, human-robot collaboration, intention recognition, search and rescue.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 331

Authors: Fl. Loury, P. Sablonière, C. Lamoureux, G. Magnier, Th. Gutierrez

Abstract:

A method is proposed to extract faithful representative patterns from data set of observations when they are suffering from non-negligible fluctuations. Supposing time interval between measurements to be extremely small compared to observation time, it consists in defining first a subset of intermediate time intervals characterizing coherent behavior. Data projection on these intervals gives a set of curves out of which an ideally “perfect” one is constructed by taking the sup limit of them. Then comparison with average real curve in corresponding interval gives an efficiency parameter expressing the degradation consecutive to fluctuation effect. The method is applied to sunlight data collected in a specific place, where ideal sunlight is the one resulting from direct exposure at location latitude over the year, and efficiency is resulting from action of meteorological parameters, mainly cloudiness, at different periods of the year. The extracted information already gives interesting element of decision, before being used for analysis of plant control.

Keywords: Base Input Reconstruction, Data Mining, Efficiency Factor, Information Pattern Operator.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1522

Authors: Maziar Ramezani, Zaidi Mohd Ripin

Abstract:

In a metal forming process, the friction between the material and the tools influences the process by modifying the stress distribution of the workpiece. This frictional behaviour is often taken into account by using a constant coefficient of friction in the finite element simulations of sheet metal forming processes. However, friction coefficient varies in time and space with many parameters. The Stribeck friction model is investigated in this study to predict springback behaviour of AA6061-T4 sheets during V-bending process. The coefficient of friction in Stribeck curve depends on sliding velocity and contact pressure. The plane-strain bending process is simulated in ABAQUS/Standard. We compared the computed punch load-stroke curves and springback related to the constant coefficient of friction with the defined friction model. The results clearly showed that the new friction model provides better agreement between experiments and results of numerical simulations. The influence of friction models on stress distribution in the workpiece is also studied numerically

Keywords: Friction model, Stress distribution, V-bending.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2734

Authors: Quratulain N. Rajput, Sajjad Haider

Abstract:

This paper applies Bayesian Networks to support information extraction from unstructured, ungrammatical, and incoherent data sources for semantic annotation. A tool has been developed that combines ontologies, machine learning, and information extraction and probabilistic reasoning techniques to support the extraction process. Data acquisition is performed with the aid of knowledge specified in the form of ontology. Due to the variable size of information available on different data sources, it is often the case that the extracted data contains missing values for certain variables of interest. It is desirable in such situations to predict the missing values. The methodology, presented in this paper, first learns a Bayesian network from the training data and then uses it to predict missing data and to resolve conflicts. Experiments have been conducted to analyze the performance of the presented methodology. The results look promising as the methodology achieves high degree of precision and recall for information extraction and reasonably good accuracy for predicting missing values.

Keywords: Information Extraction, Bayesian Network, ontology, Machine Learning

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2223

Authors: Abdelbasit Gadour

Abstract:

A great number of children in mainstream schools across Libya is currently living with emotional, behavioural difficulties. This study aims to explore teachers’ perceptions of children’s emotional and behavioural difficulties (EBD) and their attributions of the causes of EBD. The relevance of this area of study to current educational practice is illustrated in the fact that primary school teachers in Libya find classroom behaviour problems one of the major difficulties they face. The information presented in this study was gathered from 182 teachers that responded back to the survey, of whom, 27 teachers were later interviewed. In general, teachers’ perceptions of EBD reflect personal experience, training, and attitudes. Teachers appear from this study to use words such as indifferent, frightened, withdrawn, aggressive, disobedient, hyperactive, less ambitious, lacking concentration, and academically weak to describe pupils with EBD. The implications of this study are envisaged as being extremely important to support teachers addressing children’s EBD and shed light on the contributing factors to EBD for a successful teaching-learning process in Libyan primary schools.

Keywords: Teachers, children, learning, emotional and behaviour difficulties.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 602

Authors: Akella Amarendra Babu, Rama Devi Yellasiri, Natukula Sainath

Abstract:

Where human beings can easily learn and adopt pronunciation variations, machines need training before put into use. Also humans keep minimum vocabulary and their pronunciation variations are stored in front-end of their memory for ready reference, while machines keep the entire pronunciation dictionary for ready reference. Supervised methods are used for preparation of pronunciation dictionaries which take large amounts of manual effort, cost, time and are not suitable for real time use. This paper presents an unsupervised adaptation model for building agile and dynamic pronunciation dictionaries online. These methods mimic human approach in learning the new pronunciations in real time. A new algorithm for measuring sound distances called Dynamic Phone Warping is presented and tested. Performance of the system is measured using an adaptation model and the precision metrics is found to be better than 86 percent.

Keywords: Pronunciation variations, dynamic programming, machine learning, natural language processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 792

Authors: Yun-Husan Huang

Abstract:

Many English for Specific Purposes (ESP) textbooks are distributed globally as the content development is often obliged to compromises between commercial and pedagogical demands. Therefore, the issue of regional application and usefulness of globally published ESP textbooks has received much debate. For ESP instructors, textbook selection is definitely a priority consideration for curriculum design. An appropriate ESP textbook can facilitate teaching and learning, while an inappropriate one may cause a disaster for both teachers and students. This study aims to investigate the regional application and usefulness of an ESP textbook for information technology (IT). Participants were 51 sophomores majoring in Applied Informatics and Multimedia at a university in Taiwan. As they were non-English majors, their English proficiency was mostly at elementary and elementary-to-intermediate levels. This course was offered for two semesters. The textbook selected was Oxford English for Information Technology. At class end, the students were required to complete a survey comprising five choices of Very Easy, Easy, Neutral, Difficult, and Very Difficult for each item. Based on the content design of the textbook, the survey investigated how the students viewed the difficulty of grammar, listening, speaking, reading, and writing materials of the textbook. In terms of difficulty, results reveal that only 22% of them found the grammar section difficult and very difficult. For listening, 71% responded difficult and very difficult. For general reading, 55% responded difficult and very difficult. For speaking, 56% responded difficult and very difficult. For writing, 78% responded difficult and very difficult. For advanced reading, 90% reported difficult and very difficult. These results indicate that, except the grammar section, more than half of the students found the textbook contents difficult in terms of listening, speaking, reading, and writing materials. Such contradictory results between the easy grammar section and the difficult four language skills sections imply that the textbook designers do not well understand the English learning background of regional ESP learners. For the participants, the learning contents of the grammar section were the general grammar level of junior high school, while the learning contents of the four language skills sections were more of the levels of college English majors. Implications from the findings are obtained for instructors and textbook designers. First of all, existing ESP textbooks for IT are few and thus textbook selections for instructors are insufficient. Second, existing globally published textbooks for IT cannot be applied to learners of all English proficiency levels, especially the low level. With limited textbook selections, third, instructors should modify the selected textbook contents or supplement extra ESP materials to meet the proficiency level of target learners. Fourth, local ESP publishers should collaborate with local ESP instructors who understand best the learning background of their students in order to develop appropriate ESP textbooks for local learners. Even though the instructor reduced learning contents and simplified tests in curriculum design, in conclusion, the students still found difficult. This implies that in addition to the instructor’s professional experience, there is a need to understand the usefulness of the textbook from learner perspectives.

Keywords: ESP textbooks, ESP materials, ESP textbook design, learner perspectives on ESP textbooks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1891

Authors: Ahmad Rouhani, Masoud Jabbari, Sima Honarmand

Abstract:

This paper introduces a method to optimal design of a hybrid Wind/Photovoltaic/Fuel cell generation system for a typical domestic load that is not located near the electricity grid. In this configuration the combination of a battery, an electrolyser, and a hydrogen storage tank are used as the energy storage system. The aim of this design is minimization of overall cost of generation scheme over 20 years of operation. The Matlab/Simulink is applied for choosing the appropriate structure and the optimization of system sizing. A teaching learning based optimization is used to optimize the cost function. An overall power management strategy is designed for the proposed system to manage power flows among the different energy sources and the storage unit in the system. The results have been analyzed in terms of technical and economic. The simulation results indicate that the proposed hybrid system would be a feasible solution for stand-alone applications at remote locations.

Keywords: Hybrid energy system, optimum sizing, power management, TLBO.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2555

Authors: Fazıl Gökgöz, Fahrettin Filiz

Abstract:

Load forecasting has become crucial in recent years and become popular in forecasting area. Many different power forecasting models have been tried out for this purpose. Electricity load forecasting is necessary for energy policies, healthy and reliable grid systems. Effective power forecasting of renewable energy load leads the decision makers to minimize the costs of electric utilities and power plants. Forecasting tools are required that can be used to predict how much renewable energy can be utilized. The purpose of this study is to explore the effectiveness of LSTM-based neural networks for estimating renewable energy loads. In this study, we present models for predicting renewable energy loads based on deep neural networks, especially the Long Term Memory (LSTM) algorithms. Deep learning allows multiple layers of models to learn representation of data. LSTM algorithms are able to store information for long periods of time. Deep learning models have recently been used to forecast the renewable energy sources such as predicting wind and solar energy power. Historical load and weather information represent the most important variables for the inputs within the power forecasting models. The dataset contained power consumption measurements are gathered between January 2016 and December 2017 with one-hour resolution. Models use publicly available data from the Turkish Renewable Energy Resources Support Mechanism. Forecasting studies have been carried out with these data via deep neural networks approach including LSTM technique for Turkish electricity markets. 432 different models are created by changing layers cell count and dropout. The adaptive moment estimation (ADAM) algorithm is used for training as a gradient-based optimizer instead of SGD (stochastic gradient). ADAM performed better than SGD in terms of faster convergence and lower error rates. Models performance is compared according to MAE (Mean Absolute Error) and MSE (Mean Squared Error). Best five MAE results out of 432 tested models are 0.66, 0.74, 0.85 and 1.09. The forecasting performance of the proposed LSTM models gives successful results compared to literature searches.

Keywords: Deep learning, long-short-term memory, energy, renewable energy load forecasting.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1585

Authors: R. Sharma, M. Gopal

Abstract:

This paper presents a new problem solving approach that is able to generate optimal policy solution for finite-state stochastic sequential decision-making problems with high data efficiency. The proposed algorithm iteratively builds and improves an approximate Markov Decision Process (MDP) model along with cost-to-go value approximates by generating finite length trajectories through the state-space. The approach creates a synergy between an approximate evolving model and approximate cost-to-go values to produce a sequence of improving policies finally converging to the optimal policy through an intelligent and structured search of the policy space. The approach modifies the policy update step of the policy iteration so as to result in a speedy and stable convergence to the optimal policy. We apply the algorithm to a non-holonomic mobile robot control problem and compare its performance with other Reinforcement Learning (RL) approaches, e.g., a) Q-learning, b) Watkins Q(λ), c) SARSA(λ).

Keywords: Markov Decision Process (MDP), Mobile robot, Policy iteration, Simulation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1441

Authors: Cristina-Maria Păcurar

Abstract:

The paper presents an original Python-based application that outlines the advantages of combining some elementary notions of mathematics with the study of a programming language. The application support refers to some of the first lessons of analytic geometry, meaning conics and quadrics and their reduction to a standard form, as well as some related notions. The chosen programming language is Python, not only for its closer to an everyday language syntax – and therefore, enhanced readability – but also for its highly reusable code, which is of utmost importance for a mathematician that is accustomed to exploit already known and used problems to solve new ones. The purpose of this paper is, on one hand, to support the idea that one of the most appropriate means to initiate one into programming is throughout mathematics, and reciprocal, one of the most facile and handy ways to assimilate some basic knowledge in the study of mathematics is to apply them in a personal project. On the other hand, besides being a mean of learning both programming and analytic geometry, the application subject to this paper is itself a useful tool for it can be seen as an independent original Python package for analytic geometry.

Keywords: Analytic geometry, conics, Python programming language, quadrics.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1570