Search results for: learning performance

Authors: Fidelia A. Orji, Julita Vassileva

Abstract:

This research aims to develop machine learning models for students' academic performance and study strategy prediction, which could be generalized to all courses in higher education. Key learning attributes (intrinsic, extrinsic, autonomy, relatedness, competence, and self-esteem) used in building the models are chosen based on prior studies, which revealed that the attributes are essential in students’ learning process. Previous studies revealed the individual effects of each of these attributes on students’ learning progress. However, few studies have investigated the combined effect of the attributes in predicting student study strategy and academic performance to reduce the dropout rate. To bridge this gap, we used Scikit-learn in python to build five machine learning models (Decision Tree, K-Nearest Neighbour, Random Forest, Linear/Logistic Regression, and Support Vector Machine) for both regression and classification tasks to perform our analysis. The models were trained, evaluated, and tested for accuracy using 924 university dentistry students' data collected by Chilean authors through quantitative research design. A comparative analysis of the models revealed that the tree-based models such as the random forest (with prediction accuracy of 94.9%) and decision tree show the best results compared to the linear, support vector, and k-nearest neighbours. The models built in this research can be used in predicting student performance and study strategy so that appropriate interventions could be implemented to improve student learning progress. Thus, incorporating strategies that could improve diverse student learning attributes in the design of online educational systems may increase the likelihood of students continuing with their learning tasks as required. Moreover, the results show that the attributes could be modelled together and used to adapt/personalize the learning process.

Keywords: classification models, learning strategy, predictive modeling, regression models, student academic performance, student motivation, supervised machine learning

Procedia PDF Downloads 94

Authors: Chi H. Lin, Chi C. Lin, Chih L. Hsieh

Abstract:

Motor Imagery (MI) alone or combined with action observation (AO) has been shown to enhance motor performance and skill learning. The most effective way to combine these techniques has received limited scientific scrutiny. In the present study, we examined the effects of simultaneous (i.e., observing an action whilst imagining carrying out the action concurrently), alternate (i.e., observing an action and then doing imagery related to that action consecutively) and synthesis (alternately perform action observation and imagery action and then perform observation and imagery action simultaneously) AOMI combinations on improving golf putting performance and learning. Participants, 45 university students who had no formal experience of using imagery for the study, were randomly allocated to one of four training groups: simultaneous action observation and motor imagery (S-AOMI), alternate action observation and motor imagery (A-AOMI), synthesis action observation and motor imagery (A-S-AOMI), and a control group. And it was applied 'Different Experimental Groups with Pre and Post Measured' designs. Participants underwent eighteen times of different interventions, which were happened three times a week and lasting for six weeks. We analyzed the information we received based on two-factor (group × times) mixed between and within analysis of variance to discuss the real effects on participants' golf putting performance and learning about different intervention methods of different types of combined action observation and motor imagery. After the intervention, we then used imagery questionnaire and journey to understand the condition and suggestion about different motor imagery and action observation intervention from the participants. The results revealed that the three experimental groups both are effective in putting performance and learning but not for the control group, and the A-S-AOMI group is significantly better effect than S-AOMI group on golf putting performance and learning. The results confirmed the effect of motor imagery combined with action observation on the performance and learning of golf putting. In particular, in the groups of synthesis, motor imagery, or action observation were alternately performed first and then performed motor imagery, and action observation simultaneously would have the best effectiveness.

Keywords: motor skill learning, motor imagery, action observation, simulation

Procedia PDF Downloads 108

Authors: Mahasa Zahirnia

Abstract:

The purpose of this paper is to explore new and effective ways of optimizing satellite images using artificial intelligence, and the process of implementing reinforcement learning to enhance the quality of data captured within the image. In our implementation of Bellman's Reinforcement Learning equations, associated state diagrams, and multi-stage image processing, we were able to enhance image quality, detect and define objects. Reinforcement learning is the differentiator in the area of artificial intelligence, and Q-Learning relies on trial and error to achieve its goals. The reward system that is embedded in Q-Learning allows the agent to self-evaluate its performance and decide on the best possible course of action based on the current and future environment. Results show that within a simulated environment, built on the images that are commercially available, the rate of detection was 40-90%. Reinforcement learning through Q-Learning algorithm is not just desired but required design criteria for image optimization and enhancements. The proposed methods presented are a cost effective method of resolving uncertainty of the data because reinforcement learning finds ideal policies to manage the process using a smaller sample of images.

Keywords: Q-learning, image optimization, reinforcement learning, Markov decision process

Procedia PDF Downloads 185

Authors: Narimen Hamdini

Abstract:

The mastery of a foreign language often implies adequate speaking competency and communication. However, it has been marked that the Algerian students’ oral performance is affected by the lack of language practice opportunities. The present study aims at investigating the impact of cooperative learning strategies on the learners’ oral performance through integrating some learning strategies in oral expression classes. Thus, a quasi-experimental study with one group pretest-posttest design was conducted. A convenience sample of 27 second-year students from the University of Jijel, Algeria, was taught during three consecutive weeks through cooperative learning activities in conjunction with regular language instruction in oral expression classes. Regarding data collection, the study makes use of students’ questionnaire, a semi-structured interview with the teachers of oral expression, and orally scored pre-posttest. While the students’ questionnaire aims at exploring the learners ‘speaking difficulties and attitudes towards the implementation of the strategy, the semi-structured interview aims at revealing the teachers’ instructional practices and attitudes toward the integration of CL activities. Finally, the oral tests were conducted before and after the intervention to measure the effect of the strategy on the learners’ oral production. The findings showed that the experimental group scored higher in the posttest. Cooperative learning promotes not only the learner’s oral performances, but also motivation and social skills. Consequently, its implementation in the oral expression classes is validated and recommended.

Keywords: cooperative learning, learning, oral performance, teaching

Procedia PDF Downloads 105

Authors: Maria Janet, Anita David, P. Vijayasamundeeswarimaria

Abstract:

Introduction: The main purpose of this study is to compare the effectiveness of web-based and blended learning on Paediatric Basic Life Support on competency among undergraduate nursing students in selected nursing colleges in Chennai. Materials and methods: A descriptive pre-test and post-test study design were used for this study. Samples of 100 Fourth year B.Sc., nursing students at Sri Ramachandra Faculty of Nursing SRIHER, Chennai, 100 Fourth year B.Sc., nursing students at Apollo College of Nursing, Chennai, were selected by purposive sampling technique. The instrument used for data collection was Knowledge Questionnaire on Paediatric Basic Life Support (PBLS). It consists of 29 questions on the general expansion of Basic Life Support and Cardiopulmonary Resuscitation, Prerequisites of Basic Life Support, and Knowledge on Paediatric Basic Life Support in which each question has four multiple choices answers, each right answer carrying one mark and no negative scoring. This questionnaire was formed with reference to AHA 2020 (American Heart Association) revised guidelines. Results: After the post-test, in the web-based learning group, 58.8% of the students had an inadequate level of objective performance score, while 41.1% of them had an adequate level of objective performance score. In the blended learning group, 26.5% of the students had an inadequate level of an objective performance score, and 73.4% of the students had an adequate level of an objective performance score. There was an association between the post-test level of knowledge and the demographic variables of undergraduate nursing students undergoing blended learning. The age was significant at a p-value of 0.01, and the performance of BLS before was significant at a p-value of 0.05. The results show that there was a significant positive correlation between knowledge and objective performance score of undergraduate nursing students undergoing web-based learning on paediatric basic life support.

Keywords: basic life support, paediatric basic life support, web-based learning, blended learning

Procedia PDF Downloads 40

Authors: Konstantinos Sofianos, Michael Stefanidakis

Abstract:

Nowadays, advances in information and communication technologies offer a range of opportunities for new approaches, methods, and tools in the field of education and training. Teacher-centered learning has changed to student-centered learning. E-learning has now matured and enables the design and construction of intelligent learning systems. A smart learning system fully adapts to a student's needs and provides them with an education based on their preferences, learning styles, and learning backgrounds. It is a wise friend and available at any time, in any place, and with any digital device. In this paper, we propose an intelligent learning system, which includes an ontology with all elements of the learning process (learning objects, learning activities) and a massive open online course (MOOC) system. This intelligent learning system can be used in music education.

Keywords: intelligent learning systems, e-learning, music education, ontology, semantic web

Procedia PDF Downloads 280

Authors: Sam Khozama, Ali M. Mayya

Abstract:

Breast cancer is one of the most common types in women. Early prediction of breast cancer helps physicians detect cancer in its early stages. Big cancer data needs a very powerful tool to analyze and extract predictions. Machine learning and deep learning are two of the most efficient tools for predicting cancer based on textual data. In this study, we developed a fusion model of two machine learning and deep learning models. To obtain the final prediction, Long-Short Term Memory (LSTM) and ensemble learning with hyper parameters optimization are used, and score-level fusion is used. Experiments are done on the Breast Cancer Surveillance Consortium (BCSC) dataset after balancing and grouping the class categories. Five different training scenarios are used, and the tests show that the designed fusion model improved the performance by 3.3% compared to the individual models.

Keywords: machine learning, deep learning, cancer prediction, breast cancer, LSTM, fusion

Procedia PDF Downloads 131

Authors: Mahesh Balan U, Rohith Srinivaas Mohanakrishnan, Venkat Subramanian

Abstract:

Positive and unlabeled learning (PU) has gained more attention in both academic and industry research literature recently because of its relevance to existing business problems today. Yet, there still seems to be some existing challenges in terms of validating the performance of PU learning, as the actual truth of unlabeled data points is still unknown in contrast to a binary classification where we know the truth. In this study, we propose a novel PU learning technique based on the Pseudo-F measure, where we address this research gap. In this approach, we train the PU model to discriminate the probability distribution of the positive and unlabeled in the validation and spy data. The predicted probabilities of the PU model have a two-fold validation – (a) the predicted probabilities of reliable positives and predicted positives should be from the same distribution; (b) the predicted probabilities of predicted positives and predicted unlabeled should be from a different distribution. We experimented with this approach on a credit marketing case study in one of the world’s biggest fintech platforms and found evidence for benchmarking performance and backtested using historical data. This study contributes to the existing literature on semi-supervised learning.

Keywords: PU learning, semi-supervised learning, pseudo f measure, classification

Procedia PDF Downloads 201

Authors: N. Bolong, J. Makinda, I. Saad

Abstract:

Laboratory activities have produced benefits in student learning. With current drives of new technology resources and evolving era of education methods, renewal status of learning and teaching in laboratory methods are in progress, for both learners and the educators. To enhance learning outcomes in laboratory works particularly in engineering practices and testing, learning via hands-on by instruction may not sufficient. This paper describes and compares techniques and implementation of traditional (expository) with open-ended laboratory (problem-based) for two consecutive cohorts studying environmental laboratory course in civil engineering program. The transition of traditional to problem-based findings and effect were investigated in terms of course assessment student feedback survey, course outcome learning measurement and student performance grades. It was proved that students have demonstrated better performance in their grades and 12% increase in the course outcome (CO) in problem-based open-ended laboratory style than traditional method; although in perception, students has responded less favorable in their feedback.

Keywords: engineering education, open-ended laboratory, environmental engineering lab

Procedia PDF Downloads 283

Authors: Yonatan Sverdlov, Shimon Ullman

Abstract:

Artificial neural networks encounter a notable challenge known as continual learning, which involves acquiring knowledge of multiple tasks over an extended period. This challenge arises due to the tendency of previously learned weights to be adjusted to suit the objectives of new tasks, resulting in a phenomenon called catastrophic forgetting. Most approaches to this problem seek a balance between maximizing performance on the new tasks and minimizing the forgetting of previous tasks. In contrast, our approach attempts to maximize the performance of the new task, while ensuring zero forgetting. This is accomplished through the introduction of task-specific modulation parameters for each task, and only these parameters are learned for the new task, after a set of initial tasks have been learned. Through comprehensive experimental evaluations, our model demonstrates superior performance in acquiring and retaining novel tasks that pose difficulties for other multi-task models. This emphasizes the efficacy of our approach in preventing catastrophic forgetting while accommodating the acquisition of new tasks.

Keywords: continual learning, life-long learning, neural analogies, adaptive modulation

Procedia PDF Downloads 43

Authors: Amani Alsaqqaf, Frederick W. B. Li

Abstract:

Field trips are essential methods of learning in different subjects, and in recent times, there has been a reduction in the number of field trips (FTs) across all learning levels around the world. Virtual field trips (VFTs) in game environments provide FT experience based on the experiential learning theory (ELT). A conceptual framework for designing virtual field trip games (VFTGs) is developed with an aim to support game designers and educators to produce an effective FT experience where technology would enhance education. The conceptual framework quantifies ELT as an internal economy to link learning elements to game mechanics such as feedback loops which leads to facilitating VFTGs design and implementation. This study assesses the conceptual framework for designing VFTGs by investigating the possibility of applying immersive VFTGs in a secondary classroom and compare them with traditional learning that uses video clips and PowerPoint slides from the viewpoint of students’ perceived motivation, presence, and learning. The assessment is achieved by evaluating the learning performance and learner experience of a prototype VFT game, Island of Volcanoes. A quasi-experiment was conducted with 60 secondary school students. The findings of this study are that the VFTG enhanced learning performance to a better level than did the traditional way of learning, and in addition, it provided motivation and a general feeling of presence in the VFTG environment.

Keywords: conceptual framework, game-based learning, game design, virtual field trip game

Procedia PDF Downloads 194

Authors: Gladys Gakenia Njoroge

Abstract:

Mathematics is a compulsory subject in both primary and secondary schools in Kenya. However, learners’ poor performance in the subject in Kenya national examinations year in year out remains a serious concern for teachers of Mathematics, parents, curriculum developers, and the general public. This is particularly worrying because of the importance attached to the subject in national development hence the need to find out what could be affecting learning of Mathematics in Kenyan schools. The research on which this paper is based sought to examine the factors that influence performance in Mathematics in Kenyan schools; identify the characteristics of Mathematics learning disabilities; determine how the learners with such learning disabilities can be assessed and identified and interventions for these difficulties implemented. A case study was undertaken on class six learners in a primary school in Nairobi County. The tools used for the research were: classroom observations and an Individualized Education Program (IEP) developed by the teachers with the help of the researcher. This paper therefore highlights the findings from the research, discusses the implications of the findings and suggests the way forward as far as teaching, learning and assessment of Mathematics in Kenyan schools is concerned. Perhaps with the application of the right interventions, poor performance in Mathematics in the national examinations in Kenya will be a thing of the past.

Keywords: demystifying mathematics, individualized education program, learning difficulties, assessment

Procedia PDF Downloads 47

Authors: Amira Khelil, Habib Affes

Abstract:

The XXIst century is characterized by promoting teamwork as one of the main drivers of firms` performance. Collective competence is becoming crucial in developing and maintaining a firm’s competitive advantage, as well as its contributions to organizational innovation. In other words, the improvement of collective competence for a firm is no longer a choice, but rather an obligation. Learning capabilities of a firm in the context of knowledge management are assumed to be the main drivers of collective competence. Although there are some efforts to consider these concepts together; they are mostly discussed separately in the management theory. Thus, this paper aims to offer a holistic approach for development collective competence on the basis of Knowledge Management and Organizational Learning Capabilities. A theoretical model that defines a relationship between knowledge management, organizational learning and collective competence is presented at the end of this paper.

Keywords: collective competence, exploitation learning, exploration learning, knowledge management, organizational learning capabilities

Procedia PDF Downloads 469

Authors: Mesfer Al Duhayyim, Paul Newbury

Abstract:

This paper proposes an effective adaptive e-learning system that uses a coloured concept map to show the learner's knowledge level for each concept in the chosen subject area. A Fuzzy logic system is used to evaluate the learner's knowledge level for each concept in the domain, and produce a ranked concept list of learning materials to address weaknesses in the learner’s understanding. This system obtains information on the learner's understanding of concepts by an initial pre-test before the system is used for learning and a post-test after using the learning system. A Fuzzy logic system is used to produce a weighted concept map during the learning process. The aim of this research is to prove that such a proposed novel adapted e-learning system will enhance learner's performance and understanding. In addition, this research aims to increase participants' overall understanding of their learning level by providing a coloured concept map of understanding followed by a ranked concepts list of learning materials.

Keywords: adaptive e-learning system, coloured concept map, fuzzy logic, ranked concept list

Procedia PDF Downloads 254

Authors: Mary Luna Mancao Ninal

Abstract:

This study aimed to determine the performance of the kindergarten teachers and its relation to pupils’ achievement in different learning areas in the Division of Kabankalan City. Using the standardized assessment and evaluation of the Department of Education secondary data, 100 kinder teachers and 2901 kinder pupils were investigated to determine the performance of the kindergarten teachers based on their Competency–Based Performance Appraisal System for Teachers and the periodic assessment of kinder pupils collected as secondary data. Weighted mean, Pearson–r, chi-square, Analysis of Variance were used in the study. Findings revealed that the kindergarten teacher respondents were 26-31 years old and most of them were female and married; they spent teaching for two years and less and passed the Licensure Examination for Teachers. They were very satisfactory as to instructional competences, school, and home and community involvement, personal, social, and professional characteristics. It also revealed that performance of the kindergarten pupils on their period of assessment shows that they were slightly advanced in their development. It also shows that domain as to performance of the kindergarten pupils were average overall development. Based on the results, it is recommended that Kindergarten teacher must augment their educational qualification and pursue their graduate studies and must develop the total personality of the children for them to achieve high advanced development to become productive individual.

Keywords: performance, kindergarten teacher, learning areas, professional, pupil

Procedia PDF Downloads 333

Authors: Edward W. Staley, Corban G. Rivera, Ashley J. Llorens

Abstract:

Advances in reinforcement learning (RL) have resulted in recent breakthroughs in the application of artificial intelligence (AI) across many different domains. An emerging landscape of development environments is making powerful RL techniques more accessible for a growing community of researchers. However, most existing frameworks do not directly address the problem of learning in complex operating environments, such as dense urban settings or defense-related scenarios, that incorporate distributed, heterogeneous teams of agents. To help enable AI research for this important class of applications, we introduce the AI Arena: a scalable framework with flexible abstractions for distributed multi-agent reinforcement learning. The AI Arena extends the OpenAI Gym interface to allow greater flexibility in learning control policies across multiple agents with heterogeneous learning strategies and localized views of the environment. To illustrate the utility of our framework, we present experimental results that demonstrate performance gains due to a distributed multi-agent learning approach over commonly-used RL techniques in several different learning environments.

Keywords: reinforcement learning, multi-agent, deep learning, artificial intelligence

Procedia PDF Downloads 132

Authors: Ijaz Ahmed, Anwar Zada, Muhammad Waqas, M. U. Ashraf

Abstract:

We demonstrate the performance of a very efficient tagger applies on hadronically decaying top quark pairs as signal based on deep neural network algorithms and compares with the QCD multi-jet background events. A significant enhancement of performance in boosted top quark events is observed with our limited computing resources. We also compare modern machine learning approaches and perform a multivariate analysis of boosted top-pair as well as single top quark production through weak interaction at √s = 14 TeV proton-proton Collider. The most relevant known background processes are incorporated. Through the techniques of Boosted Decision Tree (BDT), likelihood and Multlayer Perceptron (MLP) the analysis is trained to observe the performance in comparison with the conventional cut based and count approach

Keywords: top tagger, multivariate, deep learning, LHC, single top

Procedia PDF Downloads 77

Authors: I Putu Indra Kusuma

Abstract:

The rapid development of Information and Communication Technology (ICT) tools has improved the implementation of flipped classrooms in English Language Teaching (ELT), especially in speaking course. Flipped classrooms have therefore evolved from the oldest version, which uses recorded videos to the newest one (3.0 version), which combines various materials and enables out-of-class interaction and learning engagement. However, how the latest version of flipped classrooms affects students’ speaking performance and influences students’ learning engagement remains unclear. This study therefore sought (1) to examine the effect of flipped classrooms 3.0 towards students’ speaking performance and (2) to explore the students’ learning engagement during the implementation of flipped classrooms in the speaking course. This study then employed explanatory sequential mixed-method design. This study conducted a quasi-experimental study by recruiting 164 twelfth grade students of a public senior high school in Indonesia as the sample. They were distributed into experimental (80 students) and control (84 students) groups. The experimental group was treated by implementing flipped classrooms with various use of ICT tools such as Schoology, Youtube, websites, and Flipgrid for eight weeks. Meanwhile, the control group implemented a conventional method. Furthermore, there were two variables examined in this study, such as the implementation of flipped classrooms 3.0 as the independent variable and students’ speaking performance as the dependent variable. The data of these two variables were then collected through administering a speaking test to both groups. The data from this experimental study were analyzed by using independent t-test analysis. Also, five students were invited to participate in semi-structured interviews to explore their learning engagement during the implementation of flipped classrooms. The findings revealed that there was a significant difference in students’ speaking performance between experimental where t (df = 162) = 5.810, p < 0.001, d = 0.91 in which experimental group performed better in speaking than the control group. Also, the results of interviews showed that the students had positive learning engagement during the implementation of flipped classrooms 3.0, especially on out-of-class interactions and face-to-face meetings. Some relevant implications to ELT, especially in speaking courses, are also drawn from the data findings. From the findings, it can be concluded that flipped classrooms 3.0 has a significant effect on students’ speaking performance and it promotes students’ learning engagement. Therefore, flipped classrooms 3.0 should be embraced as the newest version of flipped classrooms that promotes interaction outside the classrooms and learning engagement.

Keywords: Flipped Classrooms 3.0, learning engagement, teaching speaking with technology, technology-enhanced language learning

Procedia PDF Downloads 95

Authors: Haohao Wang

Abstract:

Technology, multimedia in Open Educational Resources, can contribute positively to student performance in an online instructional environment. Student performance data of past four years were obtained from an online course entitled Applied Calculus (MA139). This paper examined the data to determine whether multimedia (independent variable) had any impact on student performance (dependent variable) in online math learning, and how students felt about the value of the technology. Two groups of student data were analyzed, group 1 (control) from the online applied calculus course that did not use multimedia instructional materials, and group 2 (treatment) of the same online applied calculus course that used multimedia instructional materials. For the MA139 class, results indicate a statistically significant difference (p = .001) between the two groups, where group 1 had a final score mean of 56.36 (out of 100), group 2 of 70.68. Additionally, student testimonials were discussed in which students shared their experience in learning applied calculus online with multimedia instructional materials.

Keywords: online learning, open educational resources, multimedia, technology

Procedia PDF Downloads 348

Authors: Jinwooung Kim, Jae-Hwan Jung, Seong-Jun Kim, Sung-Ah Kim

Abstract:

Because the building envelope has a significant impact on the operation and maintenance stage of the building, designing the facade considering the performance can improve the performance of the building and lower the maintenance cost of the building. In general, however, optimizing two or more performance factors confronts the limits of time and computational tools. The optimization phase typically repeats infinitely until a series of processes that generate alternatives and analyze the generated alternatives achieve the desired performance. In particular, as complex geometry or precision increases, computational resources and time are prohibitive to find the required performance, so an optimization methodology is needed to deal with this. Instead of directly analyzing all the alternatives in the optimization process, applying experimental techniques (heuristic method) learned through experimentation and experience can reduce resource waste. This study proposes and verifies a method to optimize the double envelope of a building composed of a perforated panel using machine learning to the design geometry and quantitative performance. The proposed method is to achieve the required performance with fewer resources by supplementing the existing method which cannot calculate the complex shape of the perforated panel.

Keywords: building envelope, machine learning, perforated metal, multi-factor optimization, façade

Procedia PDF Downloads 193

Authors: Lihong Wu, Raymond Young

Abstract:

The ability to learn is one of the most critical skills in the information age. However, many students do not have a clear understanding of what learning is, what they are learning, and why they are learning. Many students study simply to pass rather than to learn something useful for their career and their life. They have a misconception about learning and a wrong attitude towards learning. This research explores student attitudes to study in management education and explores how to intercede to lead students from shallow to deeper modes of learning.

Keywords: knowledge, surface learning, deep learning, education

Procedia PDF Downloads 467

Authors: Shafaq Rubab

Abstract:

The inclusion of special needs students in a classroom is prevailing gradually in developing countries. Digital game-based learning is one the most effective instructional methodology for special needs students. Digital game-based learning facilitates special needs students who actually face challenges and obstacles in their learning processes. This study aimed to evaluate the effectiveness of digital game-based learning on the educational progress of special needs students in developing countries. The quasi-experimental research was conducted by using purposively selected sample size of eight special needs students. Results of both experimental and control group showed that performance of the experimental group students was better than the control group students and there was a significant difference between both groups’ results. This research strongly recommended that digital game-based learning can help special needs students in an inclusive classroom. It also revealed that special needs students can learn efficiently by using pedagogically sound learning games and game-based learning helps a lot for the self-paced fast-track learning system.

Keywords: inclusive education, special needs, digital game-based learning, fast-track learning

Procedia PDF Downloads 261

Authors: Jimeng Shi, Mahek Jain, Giri Narasimhan

Abstract:

Time Series Forecasting (TSF) is used to predict the target variables at a future time point based on the learning from previous time points. To keep the problem tractable, learning methods use data from a fixed-length window in the past as an explicit input. In this paper, we study how the performance of predictive models changes as a function of different look-back window sizes and different amounts of time to predict the future. We also consider the performance of the recent attention-based Transformer models, which have had good success in the image processing and natural language processing domains. In all, we compare four different deep learning methods (RNN, LSTM, GRU, and Transformer) along with a baseline method. The dataset (hourly) we used is the Beijing Air Quality Dataset from the UCI website, which includes a multivariate time series of many factors measured on an hourly basis for a period of 5 years (2010-14). For each model, we also report on the relationship between the performance and the look-back window sizes and the number of predicted time points into the future. Our experiments suggest that Transformer models have the best performance with the lowest Mean Average Errors (MAE = 14.599, 23.273) and Root Mean Square Errors (RSME = 23.573, 38.131) for most of our single-step and multi-steps predictions. The best size for the look-back window to predict 1 hour into the future appears to be one day, while 2 or 4 days perform the best to predict 3 hours into the future.

Keywords: air quality prediction, deep learning algorithms, time series forecasting, look-back window

Procedia PDF Downloads 126

Authors: Atiwit Sinyoo, Sekh Thanprasertsuk, Sithiporn Agthong, Pasakorn Watanatada, Shaun Peter Qureshi, Saknan Bongsebandhu-Phubhakdi

Abstract:

Background: Since the SARS-CoV-2 virus became extensively disseminated throughout the world, online learning has become one of the most hotly debated topics in educational reform. While some studies have already shown the advantage of online learning, there are still questions concerning how online learning affects students’ learning behavior and academic achievement when each student learns in a different way. Hence, we aimed to develop a guide for preclinical medical students to avoid drawbacks and get benefits from online learning that possibly a double-edged sword. Methods: We used a multiple-choice questionnaire to evaluate the learning behavior of second-year Thai medical students in the neuroscience course. All traditional face-to-face lecture classes were video-recorded and promptly posted to the online learning platform throughout this course. Students could pick and choose whatever classes they wanted to attend, and they may use online learning as often as they wished. Academic performance was evaluated as summative score, spot exam score and pre-post-test improvement. Results: More frequently students used online learning platform, the less they attended lecture classes (P = 0.035). High proactive online learners (High PO) who were irregular attendee (IrA) had significantly lower summative scores (P = 0.026), spot exam score (P = 0.012) and pre-post-test improvement (P = 0.036). In the meanwhile, conditional attendees (CoA) who only attended classes with attendance check had significantly higher summative score (P = 0.025) and spot exam score (P = 0.001) if they were in the High PO group. Conclusions: The benefit and drawbacks edges of using an online learning platform were demonstrated in our research. Based on this double-edged sword effect, we believe that online learning is a valuable learning strategy, but students must carefully plan their study schedule to gain the “benefit edge” meanwhile avoiding its “drawback edge”.

Keywords: academic performance, assessment, attendance, online learning, preclinical medical students

Procedia PDF Downloads 128

Authors: Pavan K. Rallabandi, Kailash C. Patidar

Abstract:

In this paper, we present an optimization technique or a learning algorithm using the hybrid architecture by combining the most popular sequence recognition models such as Recurrent Neural Networks (RNNs) and Hidden Markov models (HMMs). In order to improve the sequence or pattern recognition/ classification performance by applying a hybrid/neural symbolic approach, a gradient descent learning algorithm is developed using the Real Time Recurrent Learning of Recurrent Neural Network for processing the knowledge represented in trained Hidden Markov Models. The developed hybrid algorithm is implemented on automata theory as a sample test beds and the performance of the designed algorithm is demonstrated and evaluated on learning the deterministic finite state automata.

Keywords: hybrid systems, hidden markov models, recurrent neural networks, deterministic finite state automata

Procedia PDF Downloads 354

Authors: B. Ramesh Naik, T. Venugopal

Abstract:

This paper presents a novel approach which improves the high-level semantics of images based on machine learning approach. The contemporary approaches for image retrieval and object recognition includes Fourier transforms, Wavelets, SIFT and HoG. Though these descriptors helpful in a wide range of applications, they exploit zero order statistics, and this lacks high descriptiveness of image features. These descriptors usually take benefit of primitive visual features such as shape, color, texture and spatial locations to describe images. These features do not adequate to describe high-level semantics of the images. This leads to a gap in semantic content caused to unacceptable performance in image retrieval system. A novel method has been proposed referred as discriminative learning which is derived from machine learning approach that efficiently discriminates image features. The analysis and results of proposed approach were validated thoroughly on WANG and Caltech-101 Databases. The results proved that this approach is very competitive in content-based image retrieval.

Keywords: CBIR, discriminative learning, region weight learning, scale invariant feature transforms

Procedia PDF Downloads 147

Authors: Sresha Yadav, Ishwar Kumar

Abstract:

Second language learning is an important area of research in the language and linguistic domains. Literature suggests that several factors impact second language learning, including age, motivation, objectives, teacher, instructional material, classroom interaction, intelligence and previous background, previous linguistic experience, other student characteristics. Previous researchers have also highlighted that classroom atmospherics has a significant impact on learning as well as on the performance of students. However, the impact of classroom atmospherics on second language learning is still not known in the existing literature. Therefore, the purpose of the present study is to explore whether classroom atmospherics has an impact on second language learning or not? And if it does, it would be worthwhile to explore the nature of such relationship. The present study aims to explore the impact of classroom atmospherics on second language learning by dwelling into the existing literature to explore factors which impact second language learning, classroom atmospherics which impact language learning and the metrics through which such learning impacts could be measured. Based on the findings of literature review, the researchers have adopted a clustering approach for categorization and positioning of various measures of second language learning. Based on the clustering approach, the researchers have approach for measuring the impact of classroom atmospherics on second language learning by drawing a student sample consisting of 80 respondents. The results of the study uncover various basic premises of second language learning, especially with regard to classroom atmospherics. The present study is important not only from the point of view of language learning but implications could be drawn with regard to the design of classroom atmospherics, environmental psychology, anthropometrics, etc as well.

Keywords: classroom atmospherics, cluster analysis, linguistics, second language learning

Procedia PDF Downloads 425

Authors: Yinheng Li

Abstract:

The remarkable advancements in large language models (LLMs) have brought about significant improvements in natural language processing tasks. This paper presents a comprehensive review of in-context learning techniques, focusing on different types of prompts, including discrete, continuous, few-shot, and zero-shot, and their impact on LLM performance. We explore various approaches to prompt design, such as manual design, optimization algorithms, and evaluation methods, to optimize LLM performance across diverse tasks. Our review covers key research studies in prompt engineering, discussing their methodologies and contributions to the field. We also delve into the challenges faced in evaluating prompt performance, given the absence of a single ”best” prompt and the importance of considering multiple metrics. In conclusion, the paper highlights the critical role of prompt design in harnessing the full potential of LLMs and provides insights into the combination of manual design, optimization techniques, and rigorous evaluation for more effective and efficient use of LLMs in various Natural Language Processing (NLP) tasks.

Keywords: in-context learning, prompt engineering, zero-shot learning, large language models

Procedia PDF Downloads 51

Authors: Yi-Hsiang Pan, Chen-Hui Huang, Ching-Hsiang Chen, Wei-Ting Hsu

Abstract:

The purposes of the study were to explore the students' learning effect of physical education curriculum between merging Teaching Personal and Social Responsibility (TPSR) with sport education model and TPSR with traditional teaching model, which these learning effects included sport self-efficacy, sport enthusiastic, group cohesion, responsibility and game performance. The participants include 3 high school physical education teachers and 6 physical education classes, 133 participants with experience group 75 students and control group 58 students, and each teacher taught an experimental group and a control group for 16 weeks. The research methods used questionnaire investigation, interview, focus group meeting. The research instruments included personal and social responsibility questionnaire, sport enthusiastic scale, group cohesion scale, sport self-efficacy scale and game performance assessment instrument. Multivariate Analysis of covariance and Repeated measure ANOVA were used to test difference of students' learning effects between merging TPSR with sport education model and TPSR with traditional teaching model. The findings of research were: 1) The sport education model with TPSR could improve students' learning effects, including sport self-efficacy, game performance, sport enthusiastic, group cohesion and responsibility. 2) The traditional teaching model with TPSR could improve students' learning effect, including sport self-efficacy, responsibility and game performance. 3) the sport education model with TPSR could improve more learning effects than traditional teaching model with TPSR, including sport self-efficacy, sport enthusiastic,responsibility and game performance. 4) Based on qualitative data about learning experience of teachers and students, sport education model with TPSR significant improve learning motivation, group interaction and game sense. The conclusions indicated sport education model with TPSR could improve more learning effects in physical education curriculum. On other hand, the curricular projects of hybrid TPSR-Sport Education model and TPSR-Traditional Teaching model are both good curricular projects of moral character education, which may be applied in school physical education.

Keywords: character education, sport season, game performance, sport competence

Procedia PDF Downloads 405

Authors: Suriya

Abstract:

Rapid socio-economic development and urbanization have led to an increasingly serious air pollution problem in Ulaanbaatar (UB), the capital of Mongolia. PM₂.₅ pollution has become the most pressing aspect of UB air pollution. Therefore, monitoring and predicting PM₂.₅ concentration in UB is of great significance for the health of the local people and environmental management. As of yet, very few studies have used models to predict PM₂.₅ concentrations in UB. Using data from 0:00 on June 1, 2018, to 23:00 on April 30, 2020, we proposed two deep learning models based on Bayesian-optimized LSTM (Bayes-LSTM) and CNN-LSTM. We utilized hourly observed data, including Himawari8 (H8) aerosol optical depth (AOD), meteorology, and PM₂.₅ concentration, as input for the prediction of PM₂.₅ concentrations. The correlation strengths between meteorology, AOD, and PM₂.₅ were analyzed using the gray correlation analysis method; the comparison of the performance improvement of the model by using the AOD input value was tested, and the performance of these models was evaluated using mean absolute error (MAE) and root mean square error (RMSE). The prediction accuracies of Bayes-LSTM and CNN-LSTM deep learning models were both improved when AOD was included as an input parameter. Improvement of the prediction accuracy of the CNN-LSTM model was particularly enhanced in the non-heating season; in the heating season, the prediction accuracy of the Bayes-LSTM model slightly improved, while the prediction accuracy of the CNN-LSTM model slightly decreased. We propose two novel deep learning models for PM₂.₅ concentration prediction in UB, Bayes-LSTM, and CNN-LSTM deep learning models. Pioneering the use of AOD data from H8 and demonstrating the inclusion of AOD input data improves the performance of our two proposed deep learning models.

Keywords: deep learning, AOD, PM2.5, prediction, Ulaanbaatar

Procedia PDF Downloads 13