Search results for: repetitive learning method
9104 Computational Intelligence Hybrid Learning Approach to Time Series Forecasting
Authors: Chunshien Li, Jhao-Wun Hu, Tai-Wei Chiang, Tsunghan Wu
Abstract:
Time series forecasting is an important and widely popular topic in the research of system modeling. This paper describes how to use the hybrid PSO-RLSE neuro-fuzzy learning approach to the problem of time series forecasting. The PSO algorithm is used to update the premise parameters of the proposed prediction system, and the RLSE is used to update the consequence parameters. Thanks to the hybrid learning (HL) approach for the neuro-fuzzy system, the prediction performance is excellent and the speed of learning convergence is much faster than other compared approaches. In the experiments, we use the well-known Mackey-Glass chaos time series. According to the experimental results, the prediction performance and accuracy in time series forecasting by the proposed approach is much better than other compared approaches, as shown in Table IV. Excellent prediction performance by the proposed approach has been observed.Keywords: forecasting, hybrid learning (HL), Neuro-FuzzySystem (NFS), particle swarm optimization (PSO), recursiveleast-squares estimator (RLSE), time series
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15499103 Determination of Skills Gap between School-Based Learning and Laboratory-Based Learning in Omar Al-Mukhtar University
Authors: Aisha Othman, Crinela Pislaru, Ahmed Impes
Abstract:
This paper provides an identification of the existing practical skills gap between school-based learning (SBL) and laboratory based learning (LBL) in the Computing Department within the Faculty of Science at Omar Al-Mukhtar University in Libya. A survey has been conducted and the first author has elicited the responses of two groups of stakeholders, namely the academic teachers and students.
The primary goal is to review the main strands of evidence available and argue that there is a gap between laboratory and school-based learning in terms of opportunities for experiment and application of skills. In addition, the nature of experimental work within the laboratory at Omar Al-Mukhtar University needs to be reconsidered. Another goal of our study was to identify the reasons for students’ poor performance in the laboratory and to determine how this poor performance can be eliminated by the modification of teaching methods. Bloom’s taxonomy of learning outcomes has been applied in order to classify questions and problems into categories, and the survey was formulated with reference to third year Computing Department students. Furthermore, to discover students’ opinions with respect to all the issues, an exercise was conducted. The survey provided questions related to what the students had learnt and how well they had learnt. We were also interested in feedback on how to improve the course and the final question provided an opportunity for such feedback.
Keywords: Bloom’s taxonomy, e-learning, Omar Al-Mukhtar University.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 24179102 Combining ILP with Semi-supervised Learning for Web Page Categorization
Authors: Nuanwan Soonthornphisaj, Boonserm Kijsirikul
Abstract:
This paper presents a semi-supervised learning algorithm called Iterative-Cross Training (ICT) to solve the Web pages classification problems. We apply Inductive logic programming (ILP) as a strong learner in ICT. The objective of this research is to evaluate the potential of the strong learner in order to boost the performance of the weak learner of ICT. We compare the result with the supervised Naive Bayes, which is the well-known algorithm for the text classification problem. The performance of our learning algorithm is also compare with other semi-supervised learning algorithms which are Co-Training and EM. The experimental results show that ICT algorithm outperforms those algorithms and the performance of the weak learner can be enhanced by ILP system.
Keywords: Inductive Logic Programming, Semi-supervisedLearning, Web Page Categorization
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16339101 Locating Critical Failure Surface in Rock Slope Stability with Hybrid Model Based on Artificial Immune System and Cellular Learning Automata (CLA-AIS)
Authors: Ramin Javadzadeh, Emad Javadzadeh
Abstract:
Locating the critical slip surface with the minimum factor of safety for a rock slope is a difficult problem. In recent years, some modern global optimization methods have been developed with success in treating various types of problems, but very few of such methods have been applied to rock mechanical problems. In this paper, use of hybrid model based on artificial immune system and cellular learning automata is proposed. The results show that the algorithm is an effective and efficient optimization method with a high level of confidence rate.
Keywords: CLA-AIS, failure surface, optimization methods, rock slope.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19999100 A Survey of Sentiment Analysis Based on Deep Learning
Authors: Pingping Lin, Xudong Luo, Yifan Fan
Abstract:
Sentiment analysis is a very active research topic. Every day, Facebook, Twitter, Weibo, and other social media, as well as significant e-commerce websites, generate a massive amount of comments, which can be used to analyse peoples opinions or emotions. The existing methods for sentiment analysis are based mainly on sentiment dictionaries, machine learning, and deep learning. The first two kinds of methods rely on heavily sentiment dictionaries or large amounts of labelled data. The third one overcomes these two problems. So, in this paper, we focus on the third one. Specifically, we survey various sentiment analysis methods based on convolutional neural network, recurrent neural network, long short-term memory, deep neural network, deep belief network, and memory network. We compare their futures, advantages, and disadvantages. Also, we point out the main problems of these methods, which may be worthy of careful studies in the future. Finally, we also examine the application of deep learning in multimodal sentiment analysis and aspect-level sentiment analysis.Keywords: Natural language processing, sentiment analysis, document analysis, multimodal sentiment analysis, deep learning.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 19789099 Authentic Learning for Computer Network with Mobile Device-Based Hands-On Labware
Authors: Kai Qian, Ming Yang, Minzhe Guo, Prabir Bhattacharya, Lixin Tao
Abstract:
Computer network courses are essential parts of college computer science curriculum and hands-on networking experience is well recognized as an effective approach to help students understand better about the network concepts, the layered architecture of network protocols, and the dynamics of the networks. However, existing networking labs are usually server-based and relatively cumbersome, which require a certain level of specialty and resource to set up and maintain the lab environment. Many universities/colleges lack the resources and build-ups in this field and have difficulty to provide students with hands-on practice labs. A new affordable and easily-adoptable approach to networking labs is desirable to enhance network teaching and learning. In addition, current network labs are short on providing hands-on practice for modern wireless and mobile network learning. With the prevalence of smart mobile devices, wireless and mobile network are permeating into various aspects of our information society. The emerging and modern mobile technology provides computer science students with more authentic learning experience opportunities especially in network learning. A mobile device based hands-on labware can provide an excellent ‘real world’ authentic learning environment for computer network especially for wireless network study. In this paper, we present our mobile device-based hands-on labware (series of lab module) for computer network learning which is guided by authentic learning principles to immerse students in a real world relevant learning environment. We have been using this labware in teaching computer network, mobile security, and wireless network classes. The student feedback shows that students can learn more when they have hands-on authentic learning experience.
Keywords: Mobile computing, android, network, labware.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 20639098 Learning Factory for Changeability
Authors: Dennis Gossmann, Habil Peter Nyhuis
Abstract:
Amongst the consistently fluctuating conditions prevailing today, changeability represents a strategic key factor for a manufacturing company to achieve success on the international markets. In order to cope with turbulences and the increasing level of incalculability, not only the flexible design of production systems but in particular the employee as enabler of change provide the focus here. It is important to enable employees from manufacturing companies to participate actively in change events and in change decisions. To this end, the learning factory has been created, which is intended to serve the development of change-promoting competences and the sensitization of employees for the necessity of changes.Keywords: Changeability, human resources, learning factory.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 17159097 Exploring Students’ Self-Evaluation on Their Learning Outcomes through an Integrated Cumulative Grade Point Average Reporting Mechanism
Authors: Suriyani Ariffin, Nor Aziah Alias, Khairil Iskandar Othman, Haslinda Yusoff
Abstract:
An Integrated Cumulative Grade Point Average (iCGPA) is a mechanism and strategy to ensure the curriculum of an academic programme is constructively aligned to the expected learning outcomes and student performance based on the attainment of those learning outcomes that is reported objectively in a spider web. Much effort and time has been spent to develop a viable mechanism and trains academics to utilize the platform for reporting. The question is: How well do learners conceive the idea of their achievement via iCGPA and whether quality learner attributes have been nurtured through the iCGPA mechanism? This paper presents the architecture of an integrated CGPA mechanism purported to address a holistic evaluation from the evaluation of courses learning outcomes to aligned programme learning outcomes attainment. The paper then discusses the students’ understanding of the mechanism and evaluation of their achievement from the generated spider web. A set of questionnaires were distributed to a group of students with iCGPA reporting and frequency analysis was used to compare the perspectives of students on their performance. In addition, the questionnaire also explored how they conceive the idea of an integrated, holistic reporting and how it generates their motivation to improve. The iCGPA group was found to be receptive to what they have achieved throughout their study period. They agreed that the achievement level generated from their spider web allows them to develop intervention and enhance the programme learning outcomes before they graduate.
Keywords: Learning outcomes attainment, iCGPA, programme learning outcomes, spider web, iCGPA reporting skills.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7599096 ORR Activity and Stability of Pt-Based Electrocatalysts in PEM Fuel Cell
Authors: S. Limpattayanate, M. Hunsom
Abstract:
A comparison of activity and stability of the as-formed Pt/C, Pt-Co and Pt-Pd/C electrocatalysts, prepared by a combined approach of impregnation and seeding, was performed. According to the activity test in a single Proton Exchange Membrane (PEM) fuel cell, the Oxygen Reduction Reaction (ORR) activity of the Pt-M/C electrocatalyst was slightly lower than that of Pt/C. The j0.9 V and E10 mA/cm2 of the as-prepared electrocatalysts increased in the order of Pt/C > Pt-Co/C > Pt-Pd/C. However, in the medium-to-high current density region, Pt-Pd/C exhibited the best performance. With regard to their stability in a 0.5 M H2SO4 electrolyte solution, the electrochemical surface area decreased as the number of rounds of repetitive potential cycling increased due to the dissolution of the metals within the catalyst structure. For long-term measurement, Pt- Pd/C was the most stable than the other three electrocatalysts.Keywords: ORR activity, Stability, Pt-based electrocatalysts, PEM fuel cell.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 24109095 Meditation Based Brain Painting Promoting Foreign Language Memory through Establishing a Brain-Computer Interface
Authors: Zhepeng Rui, Zhenyu Gu, Caitilin de Bérigny
Abstract:
In the current study, we designed an interactive meditation and brain painting application to cultivate users’ creativity, promote meditation, reduce stress, and improve cognition while attempting to learn a foreign language. User tests and data analyses were conducted on 42 male and 42 female participants to better understand sex-associated psychological and aesthetic differences. Our method utilized brain-computer interfaces to import meditation and attention data to create artwork in meditation-based applications. Female participants showed statistically significantly different language learning outcomes following three meditation paradigms. The art style of brain painting helped females with language memory. Our results suggest that the most ideal methods for promoting memory attention were meditation methods and brain painting exercises contributing to language learning, memory concentration promotion, and foreign word memorization. We conclude that a short period of meditation practice can help in learning a foreign language. These findings provide insights into meditation, creative language education, brain-computer interface, and human-computer interactions.
Keywords: Brain-computer interface, creative thinking, meditation, mental health.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5549094 Hacking the Spatial Limitations in Bridging Virtual and Traditional Teaching Methodologies in Sri Lanka
Authors: Manuela Nayantara Jeyaraj
Abstract:
Having moved into the 21st century, it is way past being arguable that innovative technology needs to be incorporated into conventional classroom teaching. Though the Western world has found presumable success in achieving this, it is still a concept under battle in developing countries such as Sri Lanka. Reaching the acme of implementing interactive virtual learning within classrooms is a struggling idealistic fascination within the island. In order to overcome this problem, this study is set to reveal facts that limit the implementation of virtual, interactive learning within the school classrooms and provide hacks that could prove the augmented use of the Virtual World to enhance teaching and learning experiences. As each classroom moves along with the usage of technology to fulfill its functionalities, a few intense hacks provided will build the administrative onuses on a virtual system. These hacks may divulge barriers based on social conventions, financial boundaries, digital literacy, intellectual capacity of the staff, and highlight the impediments in introducing students to an interactive virtual learning environment and thereby provide the necessary actions or changes to be made to succeed and march along in creating an intellectual society built on virtual learning and lifestyle. This digital learning environment will be composed of multimedia presentations, trivia and pop quizzes conducted on a GUI, assessments conducted via a virtual system, records maintained on a database, etc. The ultimate objective of this study could enhance every child's basic learning environment; hence, diminishing the digital divide that exists in certain communities.
Keywords: Digital divide, digital learning, digitization, Sri Lanka, teaching methodologies.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 11859093 Evolutionary Program Based Approach for Manipulator Grasping Color Objects
Authors: Y. Harold Robinson, M. Rajaram, Honey Raju
Abstract:
Image segmentation and color identification is an important process used in various emerging fields like intelligent robotics. A method is proposed for the manipulator to grasp and place the color object into correct location. The existing methods such as PSO, has problems like accelerating the convergence speed and converging to a local minimum leading to sub optimal performance. To improve the performance, we are using watershed algorithm and for color identification, we are using EPSO. EPSO method is used to reduce the probability of being stuck in the local minimum. The proposed method offers the particles a more powerful global exploration capability. EPSO methods can determine the particles stuck in the local minimum and can also enhance learning speed as the particle movement will be faster.Keywords: Color information, EPSO, hue, saturation, value (HSV), image segmentation, particle swarm optimization (PSO). Active Contour, GMM.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15669092 Online Topic Model for Broadcasting Contents Using Semantic Correlation Information
Authors: Chang-Uk Kwak, Sun-Joong Kim, Seong-Bae Park, Sang-Jo Lee
Abstract:
This paper proposes a method of learning topics for broadcasting contents. There are two kinds of texts related to broadcasting contents. One is a broadcasting script, which is a series of texts including directions and dialogues. The other is blogposts, which possesses relatively abstracted contents, stories, and diverse information of broadcasting contents. Although two texts range over similar broadcasting contents, words in blogposts and broadcasting script are different. When unseen words appear, it needs a method to reflect to existing topic. In this paper, we introduce a semantic vocabulary expansion method to reflect unseen words. We expand topics of the broadcasting script by incorporating the words in blogposts. Each word in blogposts is added to the most semantically correlated topics. We use word2vec to get the semantic correlation between words in blogposts and topics of scripts. The vocabularies of topics are updated and then posterior inference is performed to rearrange the topics. In experiments, we verified that the proposed method can discover more salient topics for broadcasting contents.
Keywords: Broadcasting script analysis, topic expansion, semantic correlation analysis, word2vec.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 17529091 The Students' Learning Effects on Dance Domain of Arts Education
Authors: Sheng-Min Cheng
Abstract:
The purpose of this study was to explore the learning effects on dance domain in Arts Curriculum at junior and senior high levels. A total of 1,366 students from 9th to 11th grade of different areas from Taiwan were administered a self-designed dance achievement test. Data were analyzed through descriptive analysis, independent sample t test, one-way ANOVA and Post hoc comparison analysis using Scheffé Test. The results showed (1) female studentsKeywords: arts education, dance learning effects, secondary level students, dance talented students
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 20939090 Machine Learning Techniques in Bank Credit Analysis
Authors: Fernanda M. Assef, Maria Teresinha A. Steiner
Abstract:
The aim of this paper is to compare and discuss better classifier algorithm options for credit risk assessment by applying different Machine Learning techniques. Using records from a Brazilian financial institution, this study uses a database of 5,432 companies that are clients of the bank, where 2,600 clients are classified as non-defaulters, 1,551 are classified as defaulters and 1,281 are temporarily defaulters, meaning that the clients are overdue on their payments for up 180 days. For each case, a total of 15 attributes was considered for a one-against-all assessment using four different techniques: Artificial Neural Networks Multilayer Perceptron (ANN-MLP), Artificial Neural Networks Radial Basis Functions (ANN-RBF), Logistic Regression (LR) and finally Support Vector Machines (SVM). For each method, different parameters were analyzed in order to obtain different results when the best of each technique was compared. Initially the data were coded in thermometer code (numerical attributes) or dummy coding (for nominal attributes). The methods were then evaluated for each parameter and the best result of each technique was compared in terms of accuracy, false positives, false negatives, true positives and true negatives. This comparison showed that the best method, in terms of accuracy, was ANN-RBF (79.20% for non-defaulter classification, 97.74% for defaulters and 75.37% for the temporarily defaulter classification). However, the best accuracy does not always represent the best technique. For instance, on the classification of temporarily defaulters, this technique, in terms of false positives, was surpassed by SVM, which had the lowest rate (0.07%) of false positive classifications. All these intrinsic details are discussed considering the results found, and an overview of what was presented is shown in the conclusion of this study.
Keywords: Artificial Neural Networks, ANNs, classifier algorithms, credit risk assessment, logistic regression, machine learning, support vector machines.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12589089 Learning to Recognize Faces by Local Feature Design and Selection
Authors: Yanwei Pang, Lei Zhang, Zhengkai Liu
Abstract:
Studies in neuroscience suggest that both global and local feature information are crucial for perception and recognition of faces. It is widely believed that local feature is less sensitive to variations caused by illumination, expression and illumination. In this paper, we target at designing and learning local features for face recognition. We designed three types of local features. They are semi-global feature, local patch feature and tangent shape feature. The designing of semi-global feature aims at taking advantage of global-like feature and meanwhile avoiding suppressing AdaBoost algorithm in boosting weak classifies established from small local patches. The designing of local patch feature targets at automatically selecting discriminative features, and is thus different with traditional ways, in which local patches are usually selected manually to cover the salient facial components. Also, shape feature is considered in this paper for frontal view face recognition. These features are selected and combined under the framework of boosting algorithm and cascade structure. The experimental results demonstrate that the proposed approach outperforms the standard eigenface method and Bayesian method. Moreover, the selected local features and observations in the experiments are enlightening to researches in local feature design in face recognition.Keywords: Face recognition, local feature, AdaBoost, subspace analysis.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15849088 A Primer to the Learning Readiness Assessment to Raise the Sharing of e-Health Knowledge amongst Libyan Nurses
Authors: Mohamed Elhadi M. Sharif, Mona Masood
Abstract:
The usage of e-health facilities is seen to be the first priority by the Libyan government. As such this paper focuses on how the key factors or elements of working size in terms of technological availability, structural environment, and other competence-related matters may affect nurses’ sharing of knowledge in e-health. Hence, this paper investigates learning readiness assessment to raise e-health for Libyan regional hospitals by using ehealth services in nursing education.
Keywords: Libyan nurses, e-Learning readiness, e-Health.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 21589087 Reducing SAGE Data Using Genetic Algorithms
Authors: Cheng-Hong Yang, Tsung-Mu Shih, Li-Yeh Chuang
Abstract:
Serial Analysis of Gene Expression is a powerful quantification technique for generating cell or tissue gene expression data. The profile of the gene expression of cell or tissue in several different states is difficult for biologists to analyze because of the large number of genes typically involved. However, feature selection in machine learning can successfully reduce this problem. The method allows reducing the features (genes) in specific SAGE data, and determines only relevant genes. In this study, we used a genetic algorithm to implement feature selection, and evaluate the classification accuracy of the selected features with the K-nearest neighbor method. In order to validate the proposed method, we used two SAGE data sets for testing. The results of this study conclusively prove that the number of features of the original SAGE data set can be significantly reduced and higher classification accuracy can be achieved.Keywords: Serial Analysis of Gene Expression, Feature selection, Genetic Algorithm, K-nearest neighbor method.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 15979086 The Attitude of Second Year Pharmacy Students towards Lectures, Exams and E-Learning
Authors: Ahmed T. Alahmar
Abstract:
There is an increasing trend toward student-centred interactive e-learning methods and students’ feedback is a valuable tool for improving learning methods. The aim of this study was to explore the attitude of second year pharmacy students at the University of Babylon, Iraq, towards lectures, exams and e-learning. Materials and methods: Ninety pharmacy students were surveyed by paper questionnaire about their preference for lecture format, use of e-files, theoretical lectures versus practical experiments, lecture and lab time. Students were also asked about their predilection for Moodle-based online exams, different types of exam questions, exam time and other extra academic activities. Results: Students prefer to read lectures on paper (73.3%), use of PowerPoint file (76.7%), short lectures of less than 10 pages (94.5%), practical experiments (66.7%), lectures and lab time of less than two hours (89.9% and 96.6 respectively) and intra-lecture discussions (68.9%). Students also like to have paper-based exam (73.3%), short essay (40%) or MCQ (34.4%) questions and also prefer to do extra activities like reports (22.2%), seminars (18.6%) and posters (10.8%). Conclusion: Second year pharmacy students have different attitudes toward traditional and electronic leaning and assessment methods. Using multimedia, e-learning and Moodle are increasingly preferred methods among some students.
Keywords: Pharmacy, students, lecture, exam, e-learning, Moodle.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14149085 A Prediction of Attractive Evaluation Objects Based On Complex Sequential Data
Authors: Shigeaki Sakurai, Makino Kyoko, Shigeru Matsumoto
Abstract:
This paper proposes a method that predicts attractive evaluation objects. In the learning phase, the method inductively acquires trend rules from complex sequential data. The data is composed of two types of data. One is numerical sequential data. Each evaluation object has respective numerical sequential data. The other is text sequential data. Each evaluation object is described in texts. The trend rules represent changes of numerical values related to evaluation objects. In the prediction phase, the method applies new text sequential data to the trend rules and evaluates which evaluation objects are attractive. This paper verifies the effect of the proposed method by using stock price sequences and news headline sequences. In these sequences, each stock brand corresponds to an evaluation object. This paper discusses validity of predicted attractive evaluation objects, the process time of each phase, and the possibility of application tasks.
Keywords: Trend rule, frequent pattern, numerical sequential data, text sequential data, evaluation object.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12269084 Iterative Learning Control of Two Coupled Nonlinear Spherical Tanks
Authors: A. R. Tavakolpour-Saleh, A. R. Setoodeh, E. Ansari
Abstract:
This paper presents modeling and control of a highly nonlinear system including, non-interacting two spherical tanks using iterative learning control (ILC). Consequently, the objective of the paper is to control the liquid levels in the nonlinear tanks. First, a proportional-integral-derivative (PID) controller is applied to the plant model as a suitable benchmark for comparison. Then, dynamic responses of the control system corresponding to different step inputs are investigated. It is found that the conventional PID control is not able to fulfill the design criteria such as desired time constant. Consequently, an iterative learning controller is proposed to accurately control the coupled nonlinear tanks system. The simulation results clearly demonstrate the superiority of the presented ILC approach over the conventional PID controller to cope with the nonlinearities presented in the dynamic system.Keywords: Iterative learning control, spherical tanks, nonlinear system.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 12379083 Active Islanding Detection Method Using Intelligent Controller
Authors: Kuang-Hsiung Tan, Chih-Chan Hu, Chien-Wu Lan, Shih-Sung Lin, Te-Jen Chang
Abstract:
An active islanding detection method using disturbance signal injection with intelligent controller is proposed in this study. First, a DC\AC power inverter is emulated in the distributed generator (DG) system to implement the tracking control of active power, reactive power outputs and the islanding detection. The proposed active islanding detection method is based on injecting a disturbance signal into the power inverter system through the d-axis current which leads to a frequency deviation at the terminal of the RLC load when the utility power is disconnected. Moreover, in order to improve the transient and steady-state responses of the active power and reactive power outputs of the power inverter, and to further improve the performance of the islanding detection method, two probabilistic fuzzy neural networks (PFNN) are adopted to replace the traditional proportional-integral (PI) controllers for the tracking control and the islanding detection. Furthermore, the network structure and the online learning algorithm of the PFNN are introduced in detail. Finally, the feasibility and effectiveness of the tracking control and the proposed active islanding detection method are verified with experimental results.
Keywords: Distributed generators, probabilistic fuzzy neural network, islanding detection, non-detection zone.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14119082 Virtual Science Hub: An Open Source Platform to Enrich Science Teaching
Authors: Enrique Barra, Aldo Gordillo, Juan Quemada
Abstract:
This paper presents the Virtual Science Hub platform. It is an open source platform that combines a social network, an e-learning authoring tool, a videoconference service and a learning object repository for science teaching enrichment. These four main functionalities fit very well together. The platform was released in April 2012 and since then it has not stopped growing. Finally we present the results of the surveys conducted and the statistics gathered to validate this approach.
Keywords: E-learning, platform, authoring tool, science teaching.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 35099081 The RK1GL2X3 Method for Initial Value Problems in Ordinary Differential Equations
Authors: J.S.C. Prentice
Abstract:
The RK1GL2X3 method is a numerical method for solving initial value problems in ordinary differential equations, and is based on the RK1GL2 method which, in turn, is a particular case of the general RKrGLm method. The RK1GL2X3 method is a fourth-order method, even though its underlying Runge-Kutta method RK1 is the first-order Euler method, and hence, RK1GL2X3 is considerably more efficient than RK1. This enhancement is achieved through an implementation involving triple-nested two-point Gauss- Legendre quadrature.
Keywords: RK1GL2X3, RK1GL2, RKrGLm, Runge-Kutta, Gauss-Legendre, initial value problem, local error, global error.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 13049080 Ontology-Navigated Tutoring System for Flipped-Mastery Model
Authors: Masao Okabe
Abstract:
Nowadays, in Japan, variety of students get into a university and one of the main roles of introductory courses for freshmen is to make such students well prepared for subsequent intermediate courses. For that purpose, the flipped-mastery model is not enough because videos usually used in a flipped classroom is not adaptive and does not fit all freshmen with different academic performances. This paper proposes an ontology-navigated tutoring system called EduGraph. Using EduGraph, students can prepare for and review a class, in a more flexibly personalizable way than by videos. Structuralizing learning materials by its ontology, EduGraph also helps students integrate what they learn as knowledge, and makes learning materials sharable. EduGraph was used for an introductory course for freshmen. This application suggests that EduGraph is effective.
Keywords: Adaptive e-learning, flipped classroom, mastery learning, ontology.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 9889079 An Expert System for Assessment of Learning Outcomes for ABET Accreditation
Authors: M. H. Imam, Imran A. Tasadduq, Abdul-Rahim Ahmad, Fahd M. Aldosari
Abstract:
Learning outcomes of a course (CLOs) and the abilities at the time of graduation referred to as Student Outcomes (SOs) are required to be assessed for ABET accreditation. A question in an assessment must target a CLO as well as an SO and must represent a required level of competence. This paper presents the idea of an Expert System (ES) to select a proper question to satisfy ABET accreditation requirements. For ES implementation, seven attributes of a question are considered including the learning outcomes and Bloom’s Taxonomy level. A database contains all the data about a course including course content topics, course learning outcomes and the CLO-SO relationship matrix. The knowledge base of the presented ES contains a pool of questions each with tags of the specified attributes. Questions and the attributes represent expert opinions. With implicit rule base the inference engine finds the best possible question satisfying the required attributes. It is shown that the novel idea of such an ES can be implemented and applied to a course with success. An application example is presented to demonstrate the working of the proposed ES.
Keywords: Expert system, student outcomes, course learning outcomes, question attributes.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 14949078 Off-Policy Q-learning Technique for Intrusion Response in Network Security
Authors: Zheni S. Stefanova, Kandethody M. Ramachandran
Abstract:
With the increasing dependency on our computer devices, we face the necessity of adequate, efficient and effective mechanisms, for protecting our network. There are two main problems that Intrusion Detection Systems (IDS) attempt to solve. 1) To detect the attack, by analyzing the incoming traffic and inspect the network (intrusion detection). 2) To produce a prompt response when the attack occurs (intrusion prevention). It is critical creating an Intrusion detection model that will detect a breach in the system on time and also challenging making it provide an automatic and with an acceptable delay response at every single stage of the monitoring process. We cannot afford to adopt security measures with a high exploiting computational power, and we are not able to accept a mechanism that will react with a delay. In this paper, we will propose an intrusion response mechanism that is based on artificial intelligence, and more precisely, reinforcement learning techniques (RLT). The RLT will help us to create a decision agent, who will control the process of interacting with the undetermined environment. The goal is to find an optimal policy, which will represent the intrusion response, therefore, to solve the Reinforcement learning problem, using a Q-learning approach. Our agent will produce an optimal immediate response, in the process of evaluating the network traffic.This Q-learning approach will establish the balance between exploration and exploitation and provide a unique, self-learning and strategic artificial intelligence response mechanism for IDS.Keywords: Intrusion prevention, network security, optimal policy, Q-learning.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 10099077 Image Ranking to Assist Object Labeling for Training Detection Models
Authors: Tonislav Ivanov, Oleksii Nedashkivskyi, Denis Babeshko, Vadim Pinskiy, Matthew Putman
Abstract:
Training a machine learning model for object detection that generalizes well is known to benefit from a training dataset with diverse examples. However, training datasets usually contain many repeats of common examples of a class and lack rarely seen examples. This is due to the process commonly used during human annotation where a person would proceed sequentially through a list of images labeling a sufficiently high total number of examples. Instead, the method presented involves an active process where, after the initial labeling of several images is completed, the next subset of images for labeling is selected by an algorithm. This process of algorithmic image selection and manual labeling continues in an iterative fashion. The algorithm used for the image selection is a deep learning algorithm, based on the U-shaped architecture, which quantifies the presence of unseen data in each image in order to find images that contain the most novel examples. Moreover, the location of the unseen data in each image is highlighted, aiding the labeler in spotting these examples. Experiments performed using semiconductor wafer data show that labeling a subset of the data, curated by this algorithm, resulted in a model with a better performance than a model produced from sequentially labeling the same amount of data. Also, similar performance is achieved compared to a model trained on exhaustive labeling of the whole dataset. Overall, the proposed approach results in a dataset that has a diverse set of examples per class as well as more balanced classes, which proves beneficial when training a deep learning model.Keywords: Computer vision, deep learning, object detection, semiconductor.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 8059076 Kernel’s Parameter Selection for Support Vector Domain Description
Authors: Mohamed EL Boujnouni, Mohamed Jedra, Noureddine Zahid
Abstract:
Support Vector Domain Description (SVDD) is one of the best-known one-class support vector learning methods, in which one tries the strategy of using balls defined on the feature space in order to distinguish a set of normal data from all other possible abnormal objects. As all kernel-based learning algorithms its performance depends heavily on the proper choice of the kernel parameter. This paper proposes a new approach to select kernel's parameter based on maximizing the distance between both gravity centers of normal and abnormal classes, and at the same time minimizing the variance within each class. The performance of the proposed algorithm is evaluated on several benchmarks. The experimental results demonstrate the feasibility and the effectiveness of the presented method.
Keywords: Gravity centers, Kernel’s parameter, Support Vector Domain Description, Variance.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 18199075 High Wire Act: the Perils, Pitfalls and Possibilities of Online Discussions
Authors: Karen Armstrong
Abstract:
Online discussions are an important component of both blended and online courses. This paper examines the varieties of online discussions and the perils, pitfalls and possibilities of this rather new technological tool for enhanced learning. The discussion begins with possible perils and pitfalls inherent in this educational tool and moves to a consideration of the advantages of the varieties of online discussions feasible for use in teacher education programs.Keywords: online discussions, computer-mediatedcommunication (CMC), computer-supported collaborative learning(CSCL), e-learning, teacher education
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2589