Search results for: allosteric learning model

Authors: Ying Su, Morgan C. Wang

Abstract:

Long-term time series forecasting is an important research area for automated machine learning (AutoML). Currently, forecasting based on either machine learning or statistical learning is usually built by experts, and it requires significant manual effort, from model construction, feature engineering, and hyper-parameter tuning to the construction of the time series model. Automation is not possible since there are too many human interventions. To overcome these limitations, this article proposed to use recurrent neural networks (RNN) through the memory state of RNN to perform long-term time series prediction. We have shown that this proposed approach is better than the traditional Autoregressive Integrated Moving Average (ARIMA). In addition, we also found it is better than other network systems, including Fully Connected Neural Networks (FNN), Convolutional Neural Networks (CNN), and Nonpooling Convolutional Neural Networks (NPCNN).

Keywords: automated machines learning, autoregressive integrated moving average, neural networks, time series analysis

Procedia PDF Downloads 104

Authors: Diego Tancara, Raul Coto, Ariel Norambuena, Hoseein T. Dinani, Felipe Fanchini

Abstract:

Quantum machine learning is a growing research field that aims to perform machine learning tasks assisted by a quantum computer. Kernel-based quantum machine learning models are paradigmatic examples where the kernel involves quantum states, and the Gram matrix is calculated from the overlapping between these states. With the kernel at hand, a regular machine learning model is used for the learning process. In this paper we investigate the quantum support vector machine and quantum kernel ridge models to predict the degree of non-Markovianity of a quantum system. We perform digital quantum simulation of amplitude damping and phase damping channels to create our quantum dataset. We elaborate on different kernel functions to map the data and kernel circuits to compute the overlapping between quantum states. We observe a good performance of the models.

Keywords: quantum, machine learning, kernel, non-markovianity

Procedia PDF Downloads 177

Authors: Nighat Bibi, Jane Courtney, Kathleen M. Curran

Abstract:

Magnetic Resonance Imaging (MRI) is essential for the differential diagnosis of brain diseases, with deep learning methods showing promise for enhancing diagnostic accuracy. This study develops an ensemble learning model incorporating DenseNet121, EfficientNetB1, and ResNet50 architectures for the accurate classification of diverse brain conditions, including glioma, meningioma, pituitary tumors, and multiple sclerosis (MS). The model is trained on publicly available MRI datasets, utilizing Gradient-weighted Class Activation Mapping (Grad-CAM) to increase interpretability by highlighting crucial image regions, thereby enhancing transparency in AI-assisted diagnostics. The ensemble model achieved a notable classification accuracy of 99.84%, demonstrating its reliability in distinguishing multiple brain conditions. Grad-CAM visualizations further support the model’s decision-making, fostering trust in clinical applications. This approach offers a valuable tool for MRI-based diagnosis, emphasizing both accuracy and interpretability in neuroimaging. Future research will expand to larger, diverse datasets to ensure robustness across varied clinical settings.

Keywords: brain tumor, ensemble learning, explainability, grad-cam, glioma, interpretability, meningioma, multiple sclerosis, pituitary, XAI

Procedia PDF Downloads 6

Authors: Giselle Maggie-Fer Castañeda Lozano

Abstract:

The main objective of this research was to explore the role of neural network architectures in simulating behavioral phenomena as a potential explanation for selective associations, specifically related to biological constraints on learning. Biological constraints on learning refer to the limitations observed in conditioning procedures, where learning is expected to occur. The study involved simulations of five different experiments exploring various phenomena and sources of biological constraints in learning. These simulations included the interaction between response and reinforcer, stimulus and reinforcer, specificity of stimulus-reinforcer associations, species differences, neuroanatomical constraints, and learning in uncontrolled conditions. The overall results demonstrated that by manipulating neural network architectures, conditions can be created to model and explain diverse biological constraints frequently reported in comparative psychology literature as learning typicities. Additionally, the simulations offer predictive content worthy of experimental testing in the pursuit of new discoveries regarding the specificity of learning. The implications and limitations of these findings are discussed. Finally, it is suggested that this research could inaugurate a line of inquiry involving the use of neural networks to study biological factors in behavior, fostering the development of more ethical and precise research practices.

Keywords: comparative psychology, connectionism, conditioning, experimental analysis of behavior, neural networks

Procedia PDF Downloads 70

Authors: Zabeehullah, Fahim Arif, Yawar Abbas

Abstract:

Software Defined Network (SDN) is a next genera-tion networking model which simplifies the traditional network complexities and improve the utilization of constrained resources. Currently, most of the SDN based Internet of Things(IoT) environments use traditional network routing strategies which work on the basis of max or min metric value. However, IoT network heterogeneity, dynamic traffic flow and complexity demands intelligent and self-adaptive routing algorithms because traditional routing algorithms lack the self-adaptions, intelligence and efficient utilization of resources. To some extent, SDN, due its flexibility, and centralized control has managed the IoT complexity and heterogeneity but still Software Defined IoT (SDIoT) lacks intelligence. To address this challenge, we proposed a model called Deep Routing Strategy (DRS) which uses Deep Learning algorithm to perform routing in SDIoT intelligently and efficiently. Our model uses real-time traffic for training and learning. Results demonstrate that proposed model has achieved high accuracy and low packet loss rate during path selection. Proposed model has also outperformed benchmark routing algorithm (OSPF). Moreover, proposed model provided encouraging results during high dynamic traffic flow.

Keywords: SDN, IoT, DL, ML, DRS

Procedia PDF Downloads 110

Authors: Mirna Diab

Abstract:

Mobile learning, or M-learning, drives the development of new teaching, learning, and assessment strategies in schools and colleges. With initiatives across states, districts, and institutions, the United States leads mobile learning, significantly impacting education. Since 2010, over 2,3 million American pupils have received their education via mobile devices, demonstrating its rapid expansion. Nonetheless, mobile learning lacks a consistent and explicit definition that helps educators, students, and stakeholders grasp its essence and implement it effectively. This article addresses the need for a revised definition by introducing readers to various mobile learning concepts and understandings. It seeks to raise awareness, clarify, and encourage making well-informed decisions regarding its incorporation as a potent learning tool.

Keywords: mobile learning, mobile pedagogy, mobile technological devices, learner mobility

Procedia PDF Downloads 65

Authors: Bitewulign Mekonnen

Abstract:

Context: This scientific paper focuses on the use of near-infrared (NIR) spectroscopy to determine glucose concentration in aqueous solutions accurately and rapidly. The study compares six different machine learning methods for predicting glucose concentration and also explores the development of a deep learning model for classifying NIR spectra. The objective is to optimize the detection model and improve the accuracy of glucose prediction. This research is important because it provides a comprehensive analysis of various machine-learning techniques for estimating aqueous glucose concentrations. Research Aim: The aim of this study is to compare and evaluate different machine-learning methods for predicting glucose concentration from NIR spectra. Additionally, the study aims to develop and assess a deep-learning model for classifying NIR spectra. Methodology: The research methodology involves the use of machine learning and deep learning techniques. Six machine learning regression models, including support vector machine regression, partial least squares regression, extra tree regression, random forest regression, extreme gradient boosting, and principal component analysis-neural network, are employed to predict glucose concentration. The NIR spectra data is randomly divided into train and test sets, and the process is repeated ten times to increase generalization ability. In addition, a convolutional neural network is developed for classifying NIR spectra. Findings: The study reveals that the SVMR, ETR, and PCA-NN models exhibit excellent performance in predicting glucose concentration, with correlation coefficients (R) > 0.99 and determination coefficients (R²)> 0.985. The deep learning model achieves high macro-averaging scores for precision, recall, and F1-measure. These findings demonstrate the effectiveness of machine learning and deep learning methods in optimizing the detection model and improving glucose prediction accuracy. Theoretical Importance: This research contributes to the field by providing a comprehensive analysis of various machine-learning techniques for estimating glucose concentrations from NIR spectra. It also explores the use of deep learning for the classification of indistinguishable NIR spectra. The findings highlight the potential of machine learning and deep learning in enhancing the prediction accuracy of glucose-relevant features. Data Collection and Analysis Procedures: The NIR spectra and corresponding references for glucose concentration are measured in increments of 20 mg/dl. The data is randomly divided into train and test sets, and the models are evaluated using regression analysis and classification metrics. The performance of each model is assessed based on correlation coefficients, determination coefficients, precision, recall, and F1-measure. Question Addressed: The study addresses the question of whether machine learning and deep learning methods can optimize the detection model and improve the accuracy of glucose prediction from NIR spectra. Conclusion: The research demonstrates that machine learning and deep learning methods can effectively predict glucose concentration from NIR spectra. The SVMR, ETR, and PCA-NN models exhibit superior performance, while the deep learning model achieves high classification scores. These findings suggest that machine learning and deep learning techniques can be used to improve the prediction accuracy of glucose-relevant features. Further research is needed to explore their clinical utility in analyzing complex matrices, such as blood glucose levels.

Keywords: machine learning, signal processing, near-infrared spectroscopy, support vector machine, neural network

Procedia PDF Downloads 92

Authors: S. Art-in

Abstract:

This research was a survey research. The objective of this study was to study current situation and need in learning management for developing the analytical thinking of teachers in basic education of Thailand. The target group consisted of 400 teachers teaching in basic education level. They were selected by multi-stage random sampling. The instrument used in this study was the questionnaire asking current situation and need in learning management for developing the analytical thinking, 5 level rating scale. Data were analyzed by calculating the frequency, mean, standard deviation, percentage and content analysis. The research found that: 1) For current situation, the teachers provided learning management for developing analytical thinking, in overall, in “high” level. The issue with lowest level of practice: the teachers had competency in designing and establishing the learning management plan for developing the students’ analytical thinking. Considering each aspect it was found that: 1.1) the teacher aspect; the issue with lowest level of practice was: the teachers had competency in designing and establishing the learning management plan for developing the students’ analytical thinking, and 1.2) the learning management aspect for developing the students’ analytical thinking, the issue with lowest level of practice was: the learning activities provided opportunity for students to evaluate their analytical thinking process in each learning session. 2) The teachers showed their need in learning management for developing the analytical thinking, in overall, in “the highest” level. The issue with highest level of the need was: to obtain knowledge and competency in model, technique, and method for learning management or steps of learning management for developing the students’ analytical thinking. Considering each aspect it was found that: 2.1) teacher aspect; the issue with highest level of the need was: to obtain knowledge and comprehension in model, technique, and method for learning management or steps of learning management for developing the students’ analytical thinking, and 2.2) learning management aspect for developing the analytical thinking, the issue with highest level of need consisted of the determination of learning activities as problem situation, and the opportunity for students to comprehend the problem situation as well as practice their analytical thinking in order to find the answer.

Keywords: current situation and need, learning management, analytical thinking, teachers in basic education level, Thailand

Procedia PDF Downloads 351

Authors: Kim Bower

Abstract:

Evaluating and researching motivation in language learning is a complex and multi-faceted activity. Various models for investigating learner motivation have been proposed in the literature, but no one model supplies a complex and coherent model for investigating a range of motivational characteristics. Here, such a methodological framework, which includes exemplification of sources of evidence and potential methods of investigation, is proposed. The process model for the investigation of motivation within language learning settings proposed is based on a complex dynamic systems perspective that takes account of cognition and affects. It focuses on three overarching aspects of motivation: the learning environment, learner engagement and learner identities. Within these categories subsets are defined: the learning environment incorporates teacher, course and group specific aspects of motivation; learner engagement addresses the principal characteristics of learners' perceived value of activities, their attitudes towards language learning, their perceptions of their learning and engagement in learning tasks; and within learner identities, principal characteristics of self-concept and mastery of the language are explored. Exemplifications of potential sources of evidence in the model reflect the multiple influences within and between learner and environmental factors and the possible changes in both that may emerge over time. The model was initially developed as a framework for investigating different models of Content and Language Integrated Learning (CLIL) in contrasting contexts in secondary schools in England. The study, from which examples are drawn to exemplify the model, aimed to address the following three research questions: (1) in what ways does CLIL impact on learner motivation? (2) what are the main elements of CLIL that enhance motivation? and (3) to what extent might these be transferable to other contexts? This new model has been tried and tested in three locations in England and reported as case studies. Following an initial visit to each institution to discuss the qualitative research, instruments were developed according to the proposed model. A questionnaire was drawn up and completed by one group prior to a 3-day data collection visit to each institution, during which interviews were held with academic leaders, the head of the department, the CLIL teacher(s), and two learner focus groups of six-eight learners. Interviews were recorded and transcribed verbatim. 2-4 naturalistic observations of lessons were undertaken in each setting, as appropriate to the context, to provide colour and thereby a richer picture. Findings were subjected to an interpretive analysis by the themes derived from the process model and are reported elsewhere. The model proved to be an effective and coherent framework for planning the research, instrument design, data collection and interpretive analysis of data in these three contrasting settings, in which different models of language learning were in place. It is hoped that the proposed model, reported here together with exemplification and commentary, will enable teachers and researchers in a wide range of language learning contexts to investigate learner motivation in a systematic and in-depth manner.

Keywords: investigate, language-learning, learner motivation model, dynamic systems perspective

Procedia PDF Downloads 267

Authors: Pajaree Naramitpanich, Natchanok Jansawang, Panwilai Chomchid

Abstract:

The aims of this are compared between the students’ logical thinking abilities of their learning for designing the 5-lesson instructional plans of the 2-instructional methods, namely; the STEM Education and the Creative Learning Process (CLP) for developing students’ logical thinking abilities that a sample consisted of 90 students from two chemistry classes of different learning outcomes in Wapi Phathum School with the cluster random sampling technique was used at the 11th grade level. To administer of their learning environments with the 45-experimenl student group by the STEM Education method and the 45-controlling student group by the Creative Learning Process. These learning different groups were obtained using the 5 instruments; the 5-lesson instructional plans of the STEM Education and the Creative Learning Process to enhance the logical thinking tests on Mineral issue were used. The efficiency of the Creative Learning Processes (CLP) Model and the STEM Education’s innovations of these each five instructional lesson plans based on criteria are higher than of 80/80 standard level with the IOC index from the expert educators. The averages mean scores of students’ learning achievement motives were assessed with the Pre and Post Techniques and Logical Thinking Ability Test (LTAT) and dependent t-test analysis were differentiated between the CLP and the STEM, significantly. Students’ perceptions of their chemistry classroom environment inventories with the MCI with the CLP and the STEM methods also were found, differently. Associations between students’ perceptions of their chemistry classroom learning environment inventories on the CLP Model and the STEM Education learning designs toward their logical thinking abilities toward chemistry, the predictive efficiency of R2 values indicate that 68% and 76% of the variances in students’ logical thinking abilities toward chemistry to their controlling and experimental chemistry classroom learning environmental groups with the MCI were correlated at .05 levels, significantly. Implementations of this result are showed the students’ learning by the CLP of the potential thinking life-changing roles in most their logical thinking abilities that it is revealed that the students perceive their abilities to be highly learning achievement in chemistry group are differentiated with the STEM education of students’ outcomes.

Keywords: design, the lesson instructional plans, the stem education, the creative learning process, logical thinking ability, different, learning outcome, student, chemistry class

Procedia PDF Downloads 319

Authors: Abobaker Mohammed Qasem Farhan, Shangming Yang, Mohammed Al-Nehari

Abstract:

Pneumonia is a kind of lung disease that creates congestion in the chest. Such pneumonic conditions lead to loss of life of the severity of high congestion. Pneumonic lung disease is caused by viral pneumonia, bacterial pneumonia, or Covidi-19 induced pneumonia. The early prediction and classification of such lung diseases help to reduce the mortality rate. We propose the automatic Computer-Aided Diagnosis (CAD) system in this paper using the deep learning approach. The proposed CAD system takes input from raw computerized tomography (CT) scans of the patient's chest and automatically predicts disease classification. We designed the Hybrid Deep Learning Algorithm (HDLA) to improve accuracy and reduce processing requirements. The raw CT scans have pre-processed first to enhance their quality for further analysis. We then applied a hybrid model that consists of automatic feature extraction and classification. We propose the robust 2D Convolutional Neural Network (CNN) model to extract the automatic features from the pre-processed CT image. This CNN model assures feature learning with extremely effective 1D feature extraction for each input CT image. The outcome of the 2D CNN model is then normalized using the Min-Max technique. The second step of the proposed hybrid model is related to training and classification using different classifiers. The simulation outcomes using the publically available dataset prove the robustness and efficiency of the proposed model compared to state-of-art algorithms.

Keywords: CT scan, Covid-19, deep learning, image processing, lung disease classification

Procedia PDF Downloads 153

Authors: Noppadon Phumeechanya, Panita Wannapiroon

Abstract:

The purpose of this research is to design the ubiquitous scaffold learning environment using problem-based learning activities that enhance problem-solving skills and context awareness, and to evaluate the suitability of the ubiquitous scaffold learning environment using problem-based learning activities. We divide the research procedures into two phases. The first phase is to design the ubiquitous scaffold learning environment using problem-based learning activities, and the second is to evaluate the ubiquitous scaffold learning environment using problem-based learning activities. The sample group in this study consists of five experts selected using the purposive sampling method. We analyse data by arithmetic mean and standard deviation. The research findings are as follows; the ubiquitous scaffold learning environment using problem-based learning activities consists of three major steps, the first is preparation before learning. This prepares learners to acknowledge details and learn through u-LMS. The second is the learning process, where learning activities happen in the ubiquitous learning environment and learners learn online with scaffold systems for each step of problem solving. The third step is measurement and evaluation. The experts agree that the ubiquitous scaffold learning environment using problem-based learning activities is highly appropriate.

Keywords: ubiquitous learning environment scaffolding, learning activities, problem-based learning, problem-solving skills, context awareness

Procedia PDF Downloads 497

Authors: Budhvin T. Withana, Sulochana Rupasinghe

Abstract:

The educational system faces a significant concern with regards to Dyslexia and Dysgraphia, which are learning disabilities impacting reading and writing abilities. This is particularly challenging for children who speak the Sinhala language due to its complexity and uniqueness. Commonly used methods to detect the risk of Dyslexia and Dysgraphia rely on subjective assessments, leading to limited coverage and time-consuming processes. Consequently, delays in diagnoses and missed opportunities for early intervention can occur. To address this issue, the project developed a hybrid model that incorporates various deep learning techniques to detect the risk of Dyslexia and Dysgraphia. Specifically, Resnet50, VGG16, and YOLOv8 models were integrated to identify handwriting issues. The outputs of these models were then combined with other input data and fed into an MLP model. Hyperparameters of the MLP model were fine-tuned using Grid Search CV, enabling the identification of optimal values for the model. This approach proved to be highly effective in accurately predicting the risk of Dyslexia and Dysgraphia, providing a valuable tool for early detection and intervention. The Resnet50 model exhibited a training accuracy of 0.9804 and a validation accuracy of 0.9653. The VGG16 model achieved a training accuracy of 0.9991 and a validation accuracy of 0.9891. The MLP model demonstrated impressive results with a training accuracy of 0.99918, a testing accuracy of 0.99223, and a loss of 0.01371. These outcomes showcase the high accuracy achieved by the proposed hybrid model in predicting the risk of Dyslexia and Dysgraphia.

Keywords: neural networks, risk detection system, dyslexia, dysgraphia, deep learning, learning disabilities, data science

Procedia PDF Downloads 63

Authors: Sai Maniveer Adapa, Sai Guptha Perla, Adithya Reddy P.

Abstract:

Thyroid nodules can often be diagnosed with ultrasound imaging, although differentiating between benign and malignant nodules can be challenging for medical professionals. This work suggests a novel approach to increase the precision of thyroid nodule identification by combining machine learning and deep learning. The new approach first extracts information from the ultrasound pictures using a deep learning method known as a convolutional autoencoder. A support vector machine, a type of machine learning model, is then trained using these features. With an accuracy of 92.52%, the support vector machine can differentiate between benign and malignant nodules. This innovative technique may decrease the need for pointless biopsies and increase the accuracy of thyroid nodule detection.

Keywords: thyroid tumor diagnosis, ultrasound images, deep learning, machine learning, convolutional auto-encoder, support vector machine

Procedia PDF Downloads 56

Authors: M. E. Marwan, A. R. Madar, N. Fuad

Abstract:

Mobile learning (m-learning) is a new method in teaching and learning process which combines technology of mobile device with learning materials. It can enhance student's engagement in learning activities and facilitate them to access the learning materials at anytime and anywhere. In Kolej Poly-Tech Mara (KPTM), this method is seen as an important effort in teaching practice and to improve student learning performance. The aim of this paper is to discuss the development of m-learning application called Mobile EEF Learning System (MEEFLS) to be implemented for Electric and Electronic Fundamentals course using Flash, XML (Extensible Markup Language) and J2ME (Java 2 micro edition). System Development Life Cycle (SDLC) was used as an application development approach. It has three modules in this application such as notes or course material, exercises and video. MEELFS development is seen as a tool or a pilot test for m-learning in KPTM.

Keywords: flash, mobile device, mobile learning, teaching and learning, SDLC, XML

Procedia PDF Downloads 522

Authors: Sabahat Nadeem, Farman Ali Khan

Abstract:

Emotions are considered as a vital factor affecting the process of information handling, level of attention, memory capacity and decision making. Latest e-Learning systems are therefore taking into consideration the effective state of learners to make the learning process more effective and enjoyable. One such use of user’s affective information is in the systems that tend to regulate users’ emotions to a state optimally desirable for learning. So for, this objective has been tried to be achieved with the help of teaching strategies, background music, guided imagery, video clips and odors. Nevertheless, we know that colors can affect human emotions. Relationship between color and emotions has a strong influence on how we perceive our environment. Similarly, the colors of the interface can also affect the user positively as well as negatively. This affective behavior of color and its use as emotion regulation agent is not yet exploited. Therefore, this research proposes a Color-based Emotion Regulation Model (CERM), a new framework that can automatically adapt its colors according to user’s emotional state and her personality type and can help in producing a desirable emotional effect, aiming at providing an unobtrusive emotional support to the users of e-learning environment. The evaluation of CERM is carried out by comparing it with classical non-adaptive, static colored learning management system. Results indicate that colors of the interface, when carefully selected has significant positive impact on learner’s emotions.

Keywords: effective learning, e-learning, emotion regulation, emotional design

Procedia PDF Downloads 304

Authors: Budhvin T. Withana, Sulochana Rupasinghe

Abstract:

The problem of Dyslexia and Dysgraphia, two learning disabilities that affect reading and writing abilities, respectively, is a major concern for the educational system. Due to the complexity and uniqueness of the Sinhala language, these conditions are especially difficult for children who speak it. The traditional risk detection methods for Dyslexia and Dysgraphia frequently rely on subjective assessments, making it difficult to cover a wide range of risk detection and time-consuming. As a result, diagnoses may be delayed and opportunities for early intervention may be lost. The project was approached by developing a hybrid model that utilized various deep learning techniques for detecting risk of Dyslexia and Dysgraphia. Specifically, Resnet50, VGG16 and YOLOv8 were integrated to detect the handwriting issues, and their outputs were fed into an MLP model along with several other input data. The hyperparameters of the MLP model were fine-tuned using Grid Search CV, which allowed for the optimal values to be identified for the model. This approach proved to be effective in accurately predicting the risk of Dyslexia and Dysgraphia, providing a valuable tool for early detection and intervention of these conditions. The Resnet50 model achieved an accuracy of 0.9804 on the training data and 0.9653 on the validation data. The VGG16 model achieved an accuracy of 0.9991 on the training data and 0.9891 on the validation data. The MLP model achieved an impressive training accuracy of 0.99918 and a testing accuracy of 0.99223, with a loss of 0.01371. These results demonstrate that the proposed hybrid model achieved a high level of accuracy in predicting the risk of Dyslexia and Dysgraphia.

Keywords: neural networks, risk detection system, Dyslexia, Dysgraphia, deep learning, learning disabilities, data science

Procedia PDF Downloads 112

Authors: Maryam Zarabian, Hector Guzman, Pedro Pereira-Almao, Abraham Fapojuwo

Abstract:

Dry reforming of methane (DRM) has sparked significant industrial and scientific interest not only as a viable alternative for addressing the environmental concerns of two main contributors of the greenhouse effect, i.e., carbon dioxide (CO₂) and methane (CH₄), but also produces syngas, i.e., a mixture of hydrogen (H₂) and carbon monoxide (CO) utilized by a wide range of downstream processes as a feedstock for other chemical productions. In this study, we develop an AI-enable syngas production model to tackle the problem of achieving an equivalent H₂/CO ratio [1:1] with respect to the most efficient conversion. Firstly, the unsupervised density-based spatial clustering of applications with noise (DBSAN) algorithm removes outlier data points from the original experimental dataset. Then, random forest (RF) and deep neural network (DNN) models employ the error-free dataset to predict the DRM results. DNN models inherently would not be able to obtain accurate predictions without a huge dataset. To cope with this limitation, we employ reusing pre-trained layers’ approaches such as transfer learning and greedy layer-wise pretraining. Compared to the other deep models (i.e., pure deep model and transferred deep model), the greedy layer-wise pre-trained deep model provides the most accurate prediction as well as similar accuracy to the RF model with R² values 1.00, 0.999, 0.999, 0.999, 0.999, and 0.999 for the total outlet flow, H₂/CO ratio, H₂ yield, CO yield, CH₄ conversion, and CO₂ conversion outputs, respectively.

Keywords: artificial intelligence, dry reforming of methane, artificial neural network, deep learning, machine learning, transfer learning, greedy layer-wise pretraining

Procedia PDF Downloads 84

Authors: Sameh Baccari, Mahmoud Neji

Abstract:

In recent years, the rapid developments on mobile devices and wireless technologies enable new dimension capabilities for the learning domain. This dimension facilitates people daily activities and shortens the distances between individuals. When these technologies have been used in learning, a new paradigm has been emerged giving birth to mobile learning. Because of the mobility feature, m-learning courses have to be adapted dynamically to the learner’s context. The main challenge in context-aware mobile learning is to develop an approach building the best learning resources according to dynamic learning situations. In this paper, we propose a context-aware mobile learning system called Collaborative and Context-aware Mobile Learning System (CCMLS). It takes into account the requirements of Mobility, Collaboration and Context-Awareness. This system is based on the semantic modeling of the learning context and the learning content. The adaptation part of this approach is made up of adaptation rules to propose and select relevant resources, learning partners and learning activities based not only on the user’s needs, but also on its current context.

Keywords: mobile learning, mobile technologies, context-awareness, collaboration, semantic web, adaptation engine, adaptation strategy, learning object, learning context

Procedia PDF Downloads 306

Authors: Lei Zhang, Li Ji

Abstract:

Formal learning and informal learning represent two distinct learning styles: one is systematic and organized, another is causal and unstructured. Although there are many factors influencing online learning satisfaction, including self-regulation, self-efficacy, and interaction, factors influencing online formal learning and informal learning satisfaction may differ from each other. This paper investigated and compared influential factors of online formal and informal learning. Two questionnaires were created based on previous studies to explore factors influencing online formal learning and online informal learning satisfaction, respectively. A sample of 105 college students from different departments in a university located in the eastern part of China was selected to participate in this study. They all had an online learning experience and agreed to fill out questionnaires. Correlation analysis, variance analysis, and regression analysis were employed in this study. In addition, five participants were chosen for interviews. The study found that student-content, interaction, self-regulation, and self-efficacy related positively to both online formal learning and informal learning satisfaction. In addition, compared to online formal learning, student-content interaction in informal learning was the most influential factor for online learning satisfaction, perhaps that online informal learning was more goal-oriented and learners paid attention to the quality of content. In addition, results also revealed that interactions among students or teachers had little impact on online informal learning satisfaction. This study compared influential factors in online formal and informal learning satisfaction helped to add discussions to online learning satisfaction and contributed to further practices of online learning.

Keywords: learning satisfaction, formal learning, informal learning, online learning

Procedia PDF Downloads 162

Authors: Seyed-Ali Sadegh-Zadeh, Kaveh Kavianpour, Hamed Atashbar, Elham Heidari, Saeed Shiry Ghidary, Amir M. Hajiyavand

Abstract:

Evaluation metrics play a critical role in assessing the performance of machine learning models. In this review paper, we provide a comprehensive overview of performance measurement approaches for machine learning models. For each category, we discuss the most widely used metrics, including their mathematical formulations and interpretation. Additionally, we provide a comparative analysis of performance measurement approaches for metric combinations. Our review paper aims to provide researchers and practitioners with a better understanding of performance measurement approaches and to aid in the selection of appropriate evaluation metrics for their specific applications.

Keywords: evaluation metrics, performance measurement, supervised learning, unsupervised learning, reinforcement learning, model robustness and stability, comparative analysis

Procedia PDF Downloads 70

Authors: Jinan Fiaidhi, Sabah Mohammed

Abstract:

Diagnosing cataract severity is an important factor in deciding to undertake surgery. It is usually conducted by an ophthalmologist or through taking a variety of fundus photography that needs to be examined by the ophthalmologist. This paper carries out an investigation using a Siamese neural net that can be trained with small anchor samples to score cataract severity. The model used in this paper is based on a triplet loss function that takes the ophthalmologist best experience in rating positive and negative anchors to a specific cataract scaling system. This approach that takes the heuristics of the ophthalmologist is generally called the thick data approach, which is a kind of machine learning approach that learn from a few shots. Clinical Relevance: The lens of the eye is mostly made up of water and proteins. A cataract occurs when these proteins at the eye lens start to clump together and block lights causing impair vision. This research aims at employing thick data machine learning techniques to rate the severity of the cataract using Siamese neural network.

Keywords: thick data analytics, siamese neural network, triplet-loss model, few shot learning

Procedia PDF Downloads 109

Authors: Donald C. Lantu, Henndy Ginting, M. Yorga Permana, Dany M. A. Ramdlany

Abstract:

We constructed a training program for top-talents of a Bank with Scharmer Theory-U as the model. In this training program, we implemented the action learning perspective, as it is claimed to be the most effective one currently available. In the process, participants were encouraged to be more involved, especially compared to traditional lecturing. The goal of this study is to assess the effectiveness of this particular training. The program consists of six days non-residential workshop within two months. Between each workshop, the participants were involved in the works of action learning group. They were challenged by dealing with the real problem related to their tasks at work. The participants of the program were 30 best talents who were chosen according to their yearly performance. Using paired difference statistical test in the behavioral assessment, we found that the training was not effective to increase participants’ leadership competencies. For the future development program, we suggested to modify the goals of the program toward the next stage of development.

Keywords: action learning, behavior, leadership development, Theory-U

Procedia PDF Downloads 193

Authors: Todd Zhou, Mikhail Yurochkin

Abstract:

Out-of-distribution (OOD) detection is receiving increasing amounts of attention in the machine learning research community, boosted by recent technologies, such as autonomous driving and image processing. This newly-burgeoning field has called for the need for more effective and efficient methods for out-of-distribution generalization methods. Without accessing the label information, deploying machine learning models to out-of-distribution domains becomes extremely challenging since it is impossible to evaluate model performance on unseen domains. To tackle this out-of-distribution detection difficulty, we designed a model selection pipeline algorithm and developed a model selection framework with different impurity and randomness measurements to evaluate and choose the best-performing models for out-of-distribution data. By exploring different randomness scores based on predicted probabilities, we adopted the out-of-distribution entropy and developed a custom-designed score, ”CombinedScore,” as the evaluation criterion. This proposed score was created by adding labeled source information into the judging space of the uncertainty entropy score using harmonic mean. Furthermore, the prediction bias was explored through the equality of opportunity violation measurement. We also improved machine learning model performance through model calibration. The effectiveness of the framework with the proposed evaluation criteria was validated on the Folktables American Community Survey (ACS) datasets.

Keywords: model selection, domain generalization, model fairness, randomness measurements, bias index

Procedia PDF Downloads 123

Authors: Luis F. Recalde, Daniela A. Bastidas, Dayana E. Gallegos, Patricia N. Constante, Victor H. Andaluz

Abstract:

This article presents a non-immersive virtual lab-oratory to emulate the behavior of the Mitsubishi Melfa RV 2SDB robotic arm, allowing students and users to acquire skills and experience related to real robots, augmenting the access and learning of robotics in Universidad de las Fuerzas Armadas (ESPE). It was developed using the mathematical model of the robotic arm, thus defining the parameters for virtual recreation. The environment, interaction, and behavior of the robotic arm were developed in a graphic engine (Unity3D) to emulate learning tasks such as in a robotics laboratory. In the virtual system, four inputs were developed for the movement of the robot arm; further, to program the robot, a user interface was created where the user selects the trajectory such as point to point, line, arc, or circle. Finally, the hypothesis of the industrial robotic learning process is validated through the level of knowledge acquired after using the system.

Keywords: virtual learning, robot arm, non-immersive reality, mathematical model

Procedia PDF Downloads 96

Authors: Gizem Eser Erdek

Abstract:

This study is research on predicting the remaining life of industrial cutting tools used in the industrial production process with deep learning methods. When the life of cutting tools decreases, they cause destruction to the raw material they are processing. This study it is aimed to predict the remaining life of the cutting tool based on the damage caused by the cutting tools to the raw material. For this, hole photos were collected from the hole-drilling machine for 8 months. Photos were labeled in 5 classes according to hole quality. In this way, the problem was transformed into a classification problem. Using the prepared data set, a model was created with convolutional neural networks, which is a deep learning method. In addition, VGGNet and ResNet architectures, which have been successful in the literature, have been tested on the data set. A hybrid model using convolutional neural networks and support vector machines is also used for comparison. When all models are compared, it has been determined that the model in which convolutional neural networks are used gives successful results of a %74 accuracy rate. In the preliminary studies, the data set was arranged to include only the best and worst classes, and the study gave ~93% accuracy when the binary classification model was applied. The results of this study showed that the remaining life of the cutting tools could be predicted by deep learning methods based on the damage to the raw material. Experiments have proven that deep learning methods can be used as an alternative for cutting tool life estimation.

Keywords: classification, convolutional neural network, deep learning, remaining life of industrial cutting tools, ResNet, support vector machine, VggNet

Procedia PDF Downloads 75

Authors: Yang Yang, Dan Liu

Abstract:

Anomaly detection models for network flow based on machine learning have poor detection performance under extremely unbalanced training data conditions and also have slow detection speed and large resource consumption when deploying on network edge devices. Embedding multi-teacher knowledge distillation (mKD) in anomaly detection can transfer knowledge from multiple teacher models to a single model. Inspired by this, we proposed a state-of-the-art model, mKDNAD, to improve detection performance. mKDNAD mine and integrate the knowledge of one-dimensional sequence and two-dimensional image implicit in network flow to improve the detection accuracy of small sample classes. The multi-teacher knowledge distillation method guides the train of the student model, thus speeding up the model's detection speed and reducing the number of model parameters. Experiments in the CICIDS2017 dataset verify the improvements of our method in the detection speed and the detection accuracy in dealing with the small sample classes.

Keywords: network flow anomaly detection (NAD), multi-teacher knowledge distillation, machine learning, deep learning

Procedia PDF Downloads 119

Authors: Yashi Ye

Abstract:

Principles play vital roles in shaping the school culture and promoting teachers' professional learning by exerting their leadership. In the changing time of the 21st century, the creative leadership of school leaders is increasingly important in cultivating the professional learning communities of teachers for eventually improving student performance in every continent. This study examines under what conditions and how principal creative leadership contributes to teachers’ professional learning and school culture. Data collected from 632 teachers in 30 primary and middle schools in the cities of Chengdu and Chongqing in mainland China are analyzed using structural equation modeling and bootstrapping tests. A moderated mediation model of principle creative leadership effects is used to analyze professional teacher learning and school culture in which the mediator will be school culture and the moderator will be power distance orientation. The results indicate that principal creative leadership has significant direct and indirect effects on teachers' professional learning. A positive correlation between principal creative leadership, professional teacher learning, and school culture is observed. Further model testing found that teacher power distance orientation moderated the significant effect of principal creative leadership on school culture. When teachers perceived higher power distance in teacher-principal relations, the effects of principal creative leadership were stronger than for those who perceived low power distance. The results indicate the “culture change” in the young generation of teachers in China, and further implications to understanding the cultural context in the field of educational leadership are discussed.

Keywords: power distance orientation, principal creative leadership, school culture, teacher professional learning

Procedia PDF Downloads 140

Authors: Prasita Sooksamran, Wareerat Kaewurai

Abstract:

STEM is an integrated teaching approach promoted by the Ministry of Education in Thailand. STEM Education is an integrated approach to teaching Science, Technology, Engineering, and Mathematics. It has been questioned by Thai teachers on the grounds of how to integrate STEM into the classroom. Therefore, the main objective of this study is to develop a science instructional model based on the STEM approach to enhance scientific mind and problem-solving skills for primary students. This study is participatory action research, and follows the following steps: 1) develop a model 2) seek the advice of experts regarding the teaching model. Developing the instructional model began with the collection and synthesis of information from relevant documents, related research and other sources in order to create prototype instructional model. 2) The examination of the validity and relevance of instructional model by a panel of nine experts. The findings were as follows: 1. The developed instructional model comprised of principles, objective, content, operational procedures and learning evaluation. There were 4 principles: 1) Learning based on the natural curiosity of primary school level children leading to knowledge inquiry, understanding and knowledge construction, 2) Learning based on the interrelation between people and environment, 3) Learning that is based on concrete learning experiences, exploration and the seeking of knowledge, 4) Learning based on the self-construction of knowledge, creativity, innovation and 5) relating their findings to real life and the solving of real-life problems. The objective of this construction model is to enhance scientific mind and problem-solving skills. Children will be evaluated according to their achievements. Lesson content is based on science as a core subject which is integrated with technology and mathematics at grade 6 level according to The Basic Education Core Curriculum 2008 guidelines. The operational procedures consisted of 6 steps: 1) Curiosity 2) Collection of data 3) Collaborative planning 4) Creativity and Innovation 5) Criticism and 6) Communication and Service. The learning evaluation is an authentic assessment based on continuous evaluation of all the material taught. 2. The experts agreed that the Science Instructional Model based on the STEM Education Approach had an excellent level of validity and relevance (4.67 S.D. 0.50).

Keywords: instructional model, STEM education, scientific mind, problem solving

Procedia PDF Downloads 189

Authors: Sara ElElimy, Samir Moustafa

Abstract:

Mobile network operators start to face many challenges in the digital era, especially with high demands from customers. Since mobile network operators are considered a source of big data, traditional techniques are not effective with new era of big data, Internet of things (IoT) and 5G; as a result, handling effectively different big datasets becomes a vital task for operators with the continuous growth of data and moving from long term evolution (LTE) to 5G. So, there is an urgent need for effective Big data analytics to predict future demands, traffic, and network performance to full fill the requirements of the fifth generation of mobile network technology. In this paper, we introduce data science techniques using machine learning and deep learning algorithms: the autoregressive integrated moving average (ARIMA), Bayesian-based curve fitting, and recurrent neural network (RNN) are employed for a data-driven application to mobile network operators. The main framework included in models are identification parameters of each model, estimation, prediction, and final data-driven application of this prediction from business and network performance applications. These models are applied to Telecom Italia Big Data challenge call detail records (CDRs) datasets. The performance of these models is found out using a specific well-known evaluation criteria shows that ARIMA (machine learning-based model) is more accurate as a predictive model in such a dataset than the RNN (deep learning model).

Keywords: big data analytics, machine learning, CDRs, 5G

Procedia PDF Downloads 138