1,720,973 research outputs found
Online Instructional Consultation (OICON) Model for Higher Education Institution
Full text linkThe purpose of this research is to solve the problems of the existing telementoring program which are (a) the miscommunication due to lack of nonverbal cues, (b) the need or competency in written communication and technical skills, and (c) the issue regarding recording, retrieving, and playback of consultation recorded document. The main objective of this research is to develop an appropriate online consultation model for higher education institution. The specific objectives of this research are to identify suitable multimedia components to be implemented in the online instructional consultation (OICon) model, to develop a prototype, and to test and evaluate the acceptance of online instructional consultation (OICon) prototype by students and lecturers in higher education institution. OICon model was established based on the identification of multimedia communications components and features that were adapted and adopted from the existing online financial consultation, tele-medicine consultation model as well as major e-consultation components for public policy consultation. In addition, recommendations from the IT and Educationist experts were also taken into consideration. This model was then transformed into a prototype and tested on 40 students and 8 lecturers. Eleven hypotheses are derived from 7 factors of TAM with actual system variable excluded. The hypotheses relationships among these 4 factors (Perceived Usefulness (PU), Perceived Ease of Use (PEOU), Attitude, Behavioural Intention (BI) are supported except that PEOU does not have positive relationship with attitude as predicted. Perceived Importance of Communication Components and Features have positive relationship with PEOU. Users are relatively positive towards the implementation of multimedia communication tools for consultation in higher education institution. Users perceived the communication components as important if the components are easy to use. They agreed that they will use the OICon prototype in the future in term of PU and Attitude
Optimizing Energy Prediction in Smart Home Area Networks and Buildings Using Artificial Neural Networks and Machine Learning Techniques
Full text linkSmart home area networks (HANs) and buildings have become increasingly popular in recent years, with the integration of various smart devices into these networks. However, managing energy consumption in these networks is a major challenge. In this paper, we propose a hybrid artificial neural network-based energy prediction model to predict energy consumption of smart devices in HANs and smart buildings. Our proposed model utilizes a combination of artificial neural networks (ANNs) and machine learning (ML) techniques to predict energy consumption in smart HANs and buildings. The ANN component of the model is used to model the complex relationships between different variables, while the ML component is used to improve the accuracy of the predictions. To evaluate the performance of our proposed model, we collected data from a smart building and a smart HAN. Our results show that the proposed model outperforms traditional prediction methods, with an average prediction error of less than 3%. The proposed model can be used to optimize energy consumption in smart HANs and buildings, by providing accurate predictions of energy consumption. This can help to reduce energy costs and improve the overall energy efficiency of these networks. Additionally, the proposed model can be easily adapted to other types of smart networks, such as smart cities and industrial networks
Early Detection and Diagnosis of Chronic Kidney and Breast Cancer Using Multi-level Machine Learning: A Hybrid Prediction Model
Full text linkIn this study, a multilevel machine learning approach is proposed for the early detection and diagnosis of chronic kidney disease (CKD) and breast cancer. The proposed hybrid prediction model uses a combination of supervised and unsupervised machine learning techniques, including Long Short-Term Memory (LSTM) and random forest algorithms, to improve the early detection and diagnosis of these diseases. The model also includes a feature selection process to extract the most relevant features from the data. The performance of the proposed model was evaluated on a dataset of patient information and compared with other machine learning models and traditional diagnostic methods. The results show that the proposed model outperforms traditional diagnostic methods and other machine learning models in terms of accuracy, sensitivity, and specificity in the early detection and diagnosis of CKD and breast cancer. The proposed multilevel machine learning approach provides an effective way to improve the early detection and diagnosis of CKD and breast cancer and has the potential to be used in clinical practice to improve patient outcomes
The Impacts of the use of Thematic & Chronologic Multi-modal Information Representation on Sequential and Global Students’ Historical Understanding
No full textThis study examined the two different modes of multi-modes information presentations that affected sequential and global learners' history understanding: thematic and chronological. A total of 134 secondary schools’ students were enrolled (69 learning in chronological mode, 65 learning in thematic mode). Before the start of the treatment session, students were given a pre-test. The results showed that multimodal information presentation did not have a significantly great impact on historical learning or between pupils who learn in chronological and thematic ways. The chronological frame of reference technique, which reflected an interactive timeline, was reported to have supported students' sequential learning in chronological mode. Students who learned in thematic mode, on the other hand, had greater gains in points of historical understanding than students who learned in chronological mode. Students who exhibited a significant difference in their assessment of historical understanding proved the effectiveness of multimedia information presentation in acquiring history-related abstract ideas
Online instructional consultation for higher education institutions in Malaysia: The system architecture
Full text linkEvolution of communication technologies at present gives impetus to researchers and practitioners by simply put the computer-mediated communication tools on their telementoring application without really understand its potential benefits. This causes several issues and challenges confronting existing telementoring program which are: (a) miscommunication due to lack of nonverbal cues, (b) the need of competency in written communication and technical skills, and (c) document recording, retrieving and reviewing. Thus, in this paper, the focus is the design and development of Online Instructional Consultation (OICon) system for student and lecturers of higher education institutions. Hybrid modes of synchronous and asynchronous communication that provide nonverbal and verbal cues are incorporated in Online Instructional Consultation (OICon) system to compensate the identified issues and challenges. We also focus on recording, retrieving and reviewing of recorded consultation document and management is also emphasized.The multimedia communication components from e-consultation model of financial and telemedicine context are adapted. The general structure, modules of OICON system, multimedia communication components, and communication server are illustrated and the potential benefits of OICon system are presented
Systematic Review: Advances in Machine Learning Frameworks for Predicting Patent Infringements
Full text linkThe rise of patent infringement cases has spurred the demand for innovative solutions in intellectual property (IP) management. This systematic review explores advancements in machine learning (ML) frameworks for predicting patent infringements, focusing on algorithm performance, data balancing, and feature selection. By evaluating Random Forest, Support Vector Machines (SVM), Logistic Regression, and hybrid ensemble models, we provide insights into their strengths and limitations. Key findings highlight the critical role of data preprocessing techniques, such as Synthetic Minority Oversampling Technique (SMOTE) and Recursive Feature Elimination (RFE), in improving model accuracy. Furthermore, ethical and practical considerations, including scalability and bias mitigation, are discussed. The review concludes by proposing a roadmap for integrating advanced ML techniques into proactive IP protection strategies
A Machine Learning-Based Predictive Framework for Patent Infringement Detection: Enhancing Intellectual Property Protection Through Hybrid Ensemble Models
Full text linkPatent infringement poses significant risks to innovation and economic growth. Traditional intellectual property (IP) protection methods are often reactive, expensive, and inefficient for large-scale patent management. This study introduces an optimized machine learning framework designed to predict patent infringements proactively. The research evaluates the performance of Random Forest, Support Vector Machines (SVM), and Logistic Regression on a curated dataset enriched with patent citations, legal status, and family size. The study employs Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance and Recursive Feature Elimination (RFE) for feature selection. A novel hybrid ensemble model integrating Random Forest and SVM is developed, achieving 75% precision, 95% recall, and an F1-score of 84%, outperforming baseline models. The findings contribute to IP management by offering a scalable predictive framework that minimizes litigation costs and enhances proactive infringement detection
Comparative Analysis of Linear and Nonlinear sEMG Methods for Detecting Muscle Fatigue During Dynamic Biceps Curls
No full textMuscle fatigue, a key concern in sports science, rehabilitation, and occupational health, influences performance, injury risk, and provides insights into muscle functionality and endurance. Surface electromyography (sEMG) has emerged as a vital tool for non-invasively tracking muscle electrical activity and gauging health. As its application for muscle fatigue assessment grows, identifying the most accurate analytical methods is essential. Current sEMG analyses employ both linear and nonlinear metrics to measure fatigue onset and progression, yet research is ongoing to determine which method is most effective in the context of dynamic contractions. The study was aimed to evaluate the efficacy of established linear and nonlinear methods in measuring muscle fatigue caused by dynamic contractions through surface electromyography (sEMG) signals. A group of twelve healthy individuals completed biceps curls at a consistent pace of one repetition per four seconds, which constituted 75% of their 10-repetition maximum. Concurrently, sEMG signals were captured from the biceps brachii muscle at 1000 Hz. To assess the sEMG signals during the initial, middle, and final sets of 10 repetitions, three linear metrics—mean frequency, median frequency, and spectral moment ratio (SMR)—along with two nonlinear approaches, namely sample entropy and detrended fluctuation analysis (DFA), were utilized. The study's outcomes indicated notable shifts in the SMR values and the two DFA-derived scaling exponents across the exercise sets. These results indicated that SMR, sample entropy, and DFA are effective in gauging muscle fatigue, with sample entropy and DFA demonstrating heightened sensitivity to the fatigue levels when compared to the linear metrics
A Data Augmented Method for Plant Disease Leaf Image Recognition based on Enhanced GAN Model Network
No full textThe identification of plant disease leaves based on deep learning is the key to control the development and spread of plant diseases. In this paper, the existing problems of traditional classification and recognition of plant disease leaves and the limitations of deep learning-based plant disease leaf training are analysed. An enhanced GAN model network based on the Wasserstein GAN loss function has been developed to address the limited training images of plant disease leaves. The self-attention layer is added into the self-encoding structure of the generating network. The effectiveness of data generated by the encoder is increased after the self-attention layer is added after the convolution. Finally, the model's training process is stabilised using the depth gradient punishment method. Three types of corn disease photos and 100 health images from the PlantVillage dataset were used as data sets in the experiment. An AWGAN model was applied to generate around 3000 images. Several data improvement techniques were applied to augment the same datasets. Comparative tests are conducted using AlexNet, VGG16, and ResNet18. The results indicate that the proposed AWGAN model is capable of generating sufficient images of maize leaf diseases with apparent lesions, making it a viable solution for data augmentation of plant disease images. The training model's recognition accuracy is significantly increased. The proposed awGAN-based image identification method for plant leaf disease efficiently resolves the over-fitting problem in the small sample training set. The model recognition accuracy in the ResNet18 network achieves 98.4%
- …
