Learning System Research Articles

Classification of food items in trash through images using a hybrid deep learning system

The issue that arouses concern is the issue of waste in general, especially waste in food. Deep learning has made its way into the aforementioned field. Pre-processing is done using a new filter derived from discrete wavelets that were derived from Legendre polynomials (LEGP) to achieve Multi Resolution Analyses (MRA), in which the image parameters are separated into proximity parameters and detail parameters, where the first is concentrated in LL, while the second is in HL, LH, HH to improve the input image of the waste in terms of noise removal and image compression. To begin the deep learning process, configure the New Convolutional Neural Network (NCNN) to configure 144 layer with the image size (224x224x3) with Google Net so that the waste resulting from leftover food is sorted. In this work, four foods were sorted: tomatoes, potatoes, bread, and rice, and the result was reached for accuracy for each upcoming image with the accuracy of the neural arrangement of rice (99.9337) to potato (100 %) to tomato (99.8959), and finally for bread (83.5492), with the accuracy of the neural network 94.12 % in 1 min 2 sec. The accuracy reached in this work with Discrete Legendre Wavelets Transform (DLEGWT) and training the new network for google net was 94.12 % in 1 minute and 2 seconds with the new filter (DLEGWT). A program was created that was translated into a new algorithm. It was shown that the new filter worked in two stages, which is the initial processing stage, included analyses image with removed the noise from the image, after which the image is compressed to reduce the space used by the image data. As for the second stage, the deep learning stage takes place with the new filter, so that a new convolutional neural network is created after Dividing the image into three channels RGB to complete the convolution process for each channel, so that the hidden layers are formed until it reaches full connectivity, so that the classification process is performed with triple accuracy. The proposed algorithm has proven its efficiency

Explainable lung cancer classification with ensemble transfer learning of VGG16, Resnet50 and InceptionV3 using grad-cam

Medical imaging stands as a critical component in diagnosing various diseases, where traditional methods often rely on manual interpretation and conventional machine learning techniques. These approaches, while effective, come with inherent limitations such as subjectivity in interpretation and constraints in handling complex image features. This research paper proposes an integrated deep learning approach utilizing pre-trained models—VGG16, ResNet50, and InceptionV3—combined within a unified framework to improve diagnostic accuracy in medical imaging. The method focuses on lung cancer detection using images resized and converted to a uniform format to optimize performance and ensure consistency across datasets. Our proposed model leverages the strengths of each pre-trained network, achieving a high degree of feature extraction and robustness by freezing the early convolutional layers and fine-tuning the deeper layers. Additionally, techniques like SMOTE and Gaussian Blur are applied to address class imbalance, enhancing model training on underrepresented classes. The model’s performance was validated on the IQ-OTH/NCCD lung cancer dataset, which was collected from the Iraq-Oncology Teaching Hospital/National Center for Cancer Diseases over a period of three months in fall 2019. The proposed model achieved an accuracy of 98.18%, with precision and recall rates notably high across all classes. This improvement highlights the potential of integrated deep learning systems in medical diagnostics, providing a more accurate, reliable, and efficient means of disease detection.

Application of Virtual Reality in Developing the Digital Twin for an Integrated Robot Learning System

Application of Virtual Reality in Developing the Digital Twin for an Integrated Robot Learning System

Machine learning and experimental analyses identified miRNA expression models associated with metastatic osteosarcoma

Osteosarcoma (OS), as the most common primary bone cancer, has a high invasiveness and metastatic potential, therefore, it has a poor prognosis. This study identified early diagnostic biomarkers using miRNA expression profiles associated with osteosarcoma metastasis. In the first step, we used RNA-seq and online microarray data from osteosarcoma tissues and cell lines to identify differentially expressed miRNAs. Then, using seven feature selection algorithms for ranking, the first-ranked miRNAs were selected as input for five machine learning systems. Using network analysis and machine learning algorithms, we developed new diagnostic models that successfully differentiated metastatic osteosarcoma from non-metastatic samples based on newly discovered miRNA signatures. The results showed that miR-34c-3p and miR-154-3p act as the most promising models in the diagnosis of metastatic osteosarcoma. Validation for this model by RT-qPCR in benign tissue and osteosarcoma biopsies confirmed the lower expression of miR-34c-3p and miR-154-3p in OS samples. In addition, a direct correlation between miR-34c-3p expression, miR-154-3p expression and tumor grade was discovered. The combined values of miR-34c-3p and miR-154-3p showed 90 % diagnostic power (AUC = 0.90) for osteosarcoma samples and 85 % (AUC = 0.85) for metastatic osteosarcoma. Adhesion junction and focal adhesion pathways, as well as epithelial-to-mesenchymal transition (EMT) GO terms, were identified as the most significant KEGG and GO terms for the top miRNAs. The findings of this study highlight the potential use of novel miRNA expression signatures for early detection of metastatic osteosarcoma. These findings may help in determining therapeutic approaches with a quantitative and faster method of metastasis detection and also be used in the development of targeted molecular therapy for this aggressive cancer. Further research is needed to confirm the clinical utility of miR-34c-3p and miR-154-3p as diagnostic biomarkers for metastatic osteosarcoma.

Selective consolidation of learning and memory via recall-gated plasticity.

In a variety of species and behavioral contexts, learning and memory formation recruits two neural systems, with initial plasticity in one system being consolidated into the other over time. Moreover, consolidation is known to be selective; that is, some experiences are more likely to be consolidated into long-term memory than others. Here, we propose and analyze a model that captures common computational principles underlying such phenomena. The key component of this model is a mechanism by which a long-term learning and memory system prioritizes the storage of synaptic changes that are consistent with prior updates to the short-term system. This mechanism, which we refer to as recall-gated consolidation, has the effect of shielding long-term memory from spurious synaptic changes, enabling it to focus on reliable signals in the environment. We describe neural circuit implementations of this model for different types of learning problems, including supervised learning, reinforcement learning, and autoassociative memory storage. These implementations involve synaptic plasticity rules modulated by factors such as prediction accuracy, decision confidence, or familiarity. We then develop an analytical theory of the learning and memory performance of the model, in comparison to alternatives relying only on synapse-local consolidation mechanisms. We find that recall-gated consolidation provides significant advantages, substantially amplifying the signal-to-noise ratio with which memories can be stored in noisy environments. We show that recall-gated consolidation gives rise to a number of phenomena that are present in behavioral learning paradigms, including spaced learning effects, task-dependent rates of consolidation, and differing neural representations in short- and long-term pathways.

Correction to: “Refiner: a reliable and efficient incentive-driven federated learning system powered by blockchain”

Correction to: “Refiner: a reliable and efficient incentive-driven federated learning system powered by blockchain”

Optimizing fairness and accuracy: a Pareto optimal approach for decision-making

AbstractIn the era of data-driven decision-making, ensuring fairness and equality in machine learning models has become increasingly crucial. Multiple fairness definitions have been brought forward to evaluate and mitigate unintended fairness-related harms in real-world applications, with little research on addressing their interactions with each other. This paper explores the application of a Minimax Pareto-optimized solution to optimize individual and group fairness at individual and group levels on the Adult Census Income dataset as well as on the German Credit dataset. The objective of training a classification model with a multi-objective loss function is to achieve fair outcomes without compromising utility objectives. We investigate the interplay of different fairness definitions, including definitions of performance consistency and traditional group and individual fairness measures, amongst each other coupled with performance. The results presented in this paper highlight the feasibility of incorporating several fairness considerations into machine learning models, which can be applied to use cases with multiple sensitive features and attributes that characterize real-world applications. This research is a valuable step toward building responsible and transparent machine learning systems that can be incorporated into critical decision-making processes.

Innovation through Artificial Intelligence in Triage Systems for Resource Optimization in Future Pandemics.

In this study, we propose the use of a machine learning system in emergency department triage during pandemics to detect patients at the highest risk of death and infection using the COVID-19 era as an example, where rapid decision making and comprehensive support have becoming increasingly crucial. All patients who consecutively presented to the emergency department were included, and more than 89 variables were automatically analyzed using the extreme gradient boosting (XGB) algorithm. The XGB system demonstrated the highest balanced accuracy at 91.61%. Additionally, it obtained results more quickly than traditional triage systems. The variables that most influenced mortality prediction were procalcitonin level, age, and oxygen saturation, followed by lactate dehydrogenase (LDH) level, C-reactive protein, the presence of interstitial infiltrates on chest X-ray, and D-dimer. Our system also identified the importance of oxygen therapy in these patients. These results highlight that XGB is a useful and novel tool in triage systems for guiding the care pathway in future pandemics, thus following the example set by the well-known COVID-19 pandemic.

Reading Strategies in the AI Age: Enhancing Comprehension and Engagement with Advanced Technologies

The advent of artificial intelligence (AI) has transformed various facets of daily life, including how individuals approach reading and comprehension. This article explores the evolving landscape of reading strategies in the AI age, evaluating the impact of AI tools and applications on reading habits, comprehension techniques, and educational outcomes. With the rapid advancements in AI technologies, traditional reading methods are being supplemented, and in some cases, entirely replaced by AI-driven solutions. The discussion emphasizes AI-driven recommendations, which tailor reading materials to individual preferences and comprehension levels, thereby fostering more personalized and efficient learning experiences. Additionally, this article delves into the integration of AI in educational settings, where adaptive learning systems and intelligent tutoring platforms provide customized guidance and real-time feedback, significantly enhancing student engagement and achievement. However, the incorporation of AI in reading and education also presents potential challenges, including issues of data privacy, the digital divide, and the ethical implications of relying heavily on technology for cognitive development. By critically examining these multifaceted impacts, this article aims to provide a comprehensive understanding of how AI is reshaping reading strategies and to highlight the considerations necessary for harnessing its potential responsibly. Keywords: Artificial Intelligence, Reading Strategies, Comprehension, Personalized Learning, Educational Technology Proceedings Citation Format Ademola, E.O. (2024): Reading Strategies in the AI Age: Enhancing Comprehension and Engagement with Advanced Technologies. Proceedings of the 38th iSTEAMS Multidisciplinary Bespoke Conference. 15th–19th July, 2024. University of Ghana, Accra, Ghana. Pp 105-124. dx.doi.org/10.22624/AIMS/ACCRABESPOKE2024P13

M is for Motivation or How To Support Student Autonomy in Asynchronous Online Courses

Drawing on his recent book, Autonomy-Supportive Teaching in Higher Education: A Practical Guide for College Professors (Rowman & Littlefield, 2023), Whitehead describes how to apply the evidence-based principles of autonomy-support into asynchronous online courses. These include, among others, being patient with students as they adapt to the learning system, designing surveys where students choose course topics, and creating opportunities for students to interact with one another.

Memory’s gatekeeper: The role of PFC in the encoding of congruent events

Theoretical models conventionally portray the consolidation of memories as a slow process that unfolds during sleep. According to the classical Complementary Learning Systems theory, the hippocampus (HPC) rapidly changes its connectivity during wakefulness to encode ongoing events and create memory ensembles that are later transferred to the prefrontal cortex (PFC) during sleep. However, recent experimental studies challenge this notion by showing that new information consistent with prior knowledge can be rapidly consolidated in PFC during wakefulness and that PFC lesions disrupt the encoding of congruent events in the HPC. The contributions of the PFC to memory encoding have therefore largely been overlooked. Moreover, most theoretical frameworks assume random and uncorrelated patterns representing memories, disregarding the correlations between our experiences. To address these shortcomings, we developed a HPC-PFC network model that simulates interactions between the HPC and PFC during the encoding of a memory (awake stage), and subsequent consolidation (sleeping stage) to examine the contributions of each region to the consolidation of novel and congruent memories. Our results show that the PFC network uses stored memory "schemas" consolidated during previous experiences to identify inputs that evoke congruent patterns of activity, quickly integrate it into its network, and gate which components are encoded in the HPC. More specifically, the PFC uses GABAergic long-range projections to inhibit HPC neurons representing input components correlated with a previously stored memory "schema," eliciting sparse hippocampal activity during exposure to congruent events, as it has been experimentally observed.

Máquinas de ensinar analógicas: as precursoras da inteligência artificial na aprendizagem

This study offers a brief history of the analogue teaching machines created almost a century ago, arguing that they can be considered as the true forerunners of artificial intelligence in education. To do so, it examines the premises of these old machines, pointing out that they closely resemble those currently associated with machine learning systems. These premises include the individualization of teaching, the reduction of the burden of laborious activities on teachers, and a better teaching experience by automating teaching. Current apprehensions about the arrival of AI systems also resemble those that plagued society when analogue machines first appeared - for example, the fear of teachers being replaced by robotic artefacts and the fear of a distancing between students and teachers rather than a closer relationship. By seeking out these connections, this study takes analyses the digital technologies used for learning, trying to help pave the way for a critical reflection on these uses based on valuing the human perspective over artificial intelligence.

Автоматизация процессов в системах дистанционного обучения: современные решения и перспективы

The role of automation in distance learning systems and its impact on educational processes is considered. Current trends in the field of distance education and the role of software robots in improving the effectiveness of the educational process are discussed. The analysis of popular SDS, such as Moodle and Google Classroom, and the prospects for their development using innovative technologies are presented. The article also discusses the problems and possibilities of automating processes in educational institutions.

EVALUATION OF E-LEARNING OUTCOMES OF MATHEMATICS STUDENTS USING THE COURSE COMPLETION METHOD

The current development of internet technology has changed the direction of the learning system at both elementary and tertiary levels by means of online, blended learning, and e-learning. This research aims to determine and analyze the learning outcomes of students using course completion during lectures using the eLearning method. The research subjects are Bina Darma University students. Data collection was carried out in this research through observation and taking primary data from the e-learning system. The instruments in this research include the independent variable (X), which is report completion, and the dependent variable (Y), which is elearning learning outcomes. The resulting research results are: 1. There is a positive influence from the activity task variable on the student learning outcome variable, which is known to be active in carrying out the tasks in e-learning. 2. There is a positive influence from the material variable on the student learning outcome variable, which is known to be active in opening the material. provided in e-learning. Multiple linear regression tests result from the course completion variable on student learning outcomes. It can be concluded that there is a positive influence on the learning outcomes of variable Y from variable X. Students can also share information or opinions on various matters relating to the subject matter or assignments given by the lecturer. Apart from that, lecturers can place learning materials and assignments that students must complete in certain places in e-learning for students to access.

Construction and Innovative Exploration of Personalized Learning Systems in the Context of Educational Data Mining

This study examines the current research on educational data mining, educational learning support services, personalized learning services, and personalized learning paths in education. The authors aim to integrate personalized learning concepts into traditional support services by drawing on the latest theoretical and practical research. Using multimodal data fusion techniques, the study conduct exploratory analyses on various data types, including learner academic performance, psychological assessments, learning behavior, and physiological information. This leads to the construction of a personalized education learning support service model. The model focuses on objectives such as monitoring learning behavior, identifying preferences, recognizing abilities, optimizing paths, and recommending resources. The goal is to provide learners with sustained support services throughout the personalized learning process, addressing individual needs, fostering enthusiasm, and maintaining long-term motivation.

On Vehicular Data Aggregation in Federated Learning

Vehicular federated learning systems will be beneficial to predicting traffic events in future intelligent cities. However, they might leak private information upon model updates. Hence, an honest but curious server could infer private information, such as the route of a vehicle. In this study, we elaborate on the nature of such privacy leakage caused by gradient sharing. With a simulated scenario, we focus on determining who is in danger of privacy threats and how successful a route inference attack can be. Results indicate that vanilla federated learning exposes intra-city and commuter traffic to successful location inference attacks. We also found that an adversarial aggregator server successfully infers the moving time of vehicles traveling during low-traffic periods.

Determination of Teacher Performance in Improving Education Quality in the Pandemic Era

Determining Teacher Performance to Improve the Quality of Education in the Pandemic Era. The problem that is used as the object of research is teacher performance which is seen to have decreased due to the change in the learning system from offline to online. This can cause the quality of education to decline. This research aims to analyze: 1). The Influence of Internal Factors on Teacher Performance. 2). The influence of internal factors on the quality of education. 3). The influence of external factors on teacher performance. 4). The influence of external factors on the quality of education. 5). The influence of teacher performance on the quality of education. 6). The influence of internal factors on the quality of education is mediated by teacher performance. 7). The influence of external factors on the quality of education is mediated by teacher performance. The data collection technique was using a questionnaire given to 85 respondents (teachers) from 3 schools, namely SMA Al Adzkar, SM AI Sinar Cendikia and Sekolah Alam Bintaro. The hypothesis used is Smart PLS ( Partial Least Square ). The research results show that: 1). The statistical value of internal factors on performance is 3.116 with a p value of 0.00 2 and a p value of less than 0.05, so it has a positive effect on teacher performance. 2). The statistical t value of internal factors on the quality of education is 3.277, the p value is 0.001 and the p value is more than 0.05, then positive influence on the quality of education. 3). The statistical t value of external factors on teacher performance is 2.561 with a p value of 0.011 or less than 0.05, so it has a positive effect on teacher performance. 4). The statistical value of external factors on the quality of education is 3.920 with a p value of 0.000 or a p value of more than 0.05, so it has a positive effect on the quality of education. 5). The statistical t value of teacher performance on the quality of education is equal to a p value of 0.022, so teacher performance has a positive effect on the quality of education. 6). The VAF value is 0.598, this shows that teacher performance is a partial mediator in the relationship between internal factors and the quality of education. 7). The VAF value is 0.631, this shows that teacher performance is a partial mediator in the relationship between external factors and the quality of education. The research conclusion is that, 1). There is a positive and significant influence of f actors internal on teacher performance. 2). There is a positive and significant influence of internal factors on the quality of education. 3). There is positive and significant influence external factors on teacher performance. 4) There is a positive and significant influence external factors on the quality of education. 5). There is a positive and significant influence on teacher performance to quality of education. 6) There is a positive and significant influence between internal factors on the quality of education with teacher performance as mediation. 7). There is a positive and significant influence between external factors on the quality of education and teacher performance as mediation.

Curriculum Portrait Independent Learning At The Center Of Excellence School (Studi Kasus di SMK Ma'arif NU Tirto Pekalongan)

In education, the curriculum is the main aspect of the learning process. The independent learning curriculum is categorized as a Center of Excellence school for vocational schools and a Driving School for high schools. However, this research focuses on vocational schools categorized as Center of Excellence schools. This study uses a qualitative method approach with field study research at SMK Ma'arif NU Tirto Pekalongan. The purpose of the research is to find out how the learning system in the Center of Excellence schools implements the Independent Learning Curriculum. Vocational Secondary Schools implementing the Merdeka Learning curriculum include the term P5BK (Pancasila Student Profile Strengthening and Work Culture Project) in each subject. Therefore, school principals, teachers, and students must be ready to implement the Independent Learning curriculum. For PK vocational schools in Pekalongan, the level of readiness is mature and according to the criteria of the Ministry of Education. This can be seen from the supporting factors, which include synchronization between teachers and students regarding the concept of the Independent Learning Curriculum, teachers as facilitators of student learning, and the encouragement of counseling guidance teachers for students and many young teachers who can utilize technology. In addition, there are several inhibiting factors, namely the difficulty of teachers in understanding the diversity of students' characters, the inadequate facilities and infrastructure of the automotive department, and the absence of conducive classroom conditions. This makes it difficult for teachers to adjust learning to the diversity of students' characters. However, the existing obstacles do not last long because at SMK Ma'arif NU Tirto Pekalongan, there is a collaboration with industry partners and assistance funds to complete the school's needs.

Learn with M.E.—Let Us Boost Personalized Learning in K-12 Math Education!

The traditional educational system, in certain aspects, limits personalized learning. This is mainly evident in the fact that average students, who do not have any learning difficulties, are required to solve the same tasks from the same textbook in the same order. Artificial intelligence and other smart learning tools present great opportunities for implementing a personalized learning system. Our previous surveys and literature reviews also show that educators see the greatest potential in personalized education for the assimilation of artificial intelligence into education. In this context, we have developed educational software called “Learn with M.E. as Math Educator”, which facilitates more personalized teaching of basic mathematical operations. This study presents the structure and operation of this application. We tested the usability of the software in several institutions. Our research target group consists of elementary school students, specifically those aged 11–15. This article provides a detailed overview of the accuracy and educational outcomes of the completed application. We evaluated the application and its effectiveness using both qualitative and quantitative methods. Our research design combined elements of educational technology development and effectiveness assessment. To evaluate student performance, we employed a control group methodology. Data were analyzed by comparing test results between students using the software and those receiving traditional instruction. We examined user satisfaction through survey questionnaires. Teachers’ opinions were gathered through structured interviews, and their responses were categorized using a SWOT analysis. The findings indicated that the use of the software significantly improved students’ mathematics performance compared to the control group. Students provided positive feedback on the software’s user interface, describing it as user-friendly and motivating. Teachers regarded the software as an effective educational tool, facilitating differentiated instruction and enhancing student engagement. The results suggest that digital educational tools, such as the developed software, can provide substantial added value in education.

Deep learning systems for forecasting the prices of crude oil and precious metals

Commodity markets, such as crude oil and precious metals, play a strategic role in the economic development of nations, with crude oil prices influencing geopolitical relations and the global economy. Moreover, gold and silver are argued to hedge the stock and cryptocurrency markets during market downsides. Therefore, accurate forecasting of crude oil and precious metals prices is critical. Nevertheless, due to the nonlinear nature, substantial fluctuations, and irregular cycles of crude oil and precious metals, predicting their prices is a challenging task. Our study contributes to the commodity market price forecasting literature by implementing and comparing advanced deep-learning models. We address this gap by including silver alongside gold in our analysis, offering a more comprehensive understanding of the precious metal markets. This research expands existing knowledge and provides valuable insights into predicting commodity prices. In this study, we implemented 16 deep- and machine-learning models to forecast the daily price of the West Texas Intermediate (WTI), Brent, gold, and silver markets. The employed deep-learning models are long short-term memory (LSTM), BiLSTM, gated recurrent unit (GRU), bidirectional gated recurrent units (BiGRU), T2V-BiLSTM, T2V-BiGRU, convolutional neural networks (CNN), CNN-BiLSTM, CNN-BiGRU, temporal convolutional network (TCN), TCN-BiLSTM, and TCN-BiGRU. We compared the forecasting performance of deep-learning models with the baseline random forest, LightGBM, support vector regression, and k-nearest neighborhood models using mean absolute error (MAE), mean absolute percentage error, and root mean squared error as evaluation criteria. By considering different sliding window lengths, we examine the forecasting performance of our models. Our results reveal that the TCN model outperforms the others for WTI, Brent, and silver, achieving the lowest MAE values of 1.444, 1.295, and 0.346, respectively. The BiGRU model performs best for gold, with an MAE of 15.188 using a 30-day input sequence. Furthermore, LightGBM exhibits comparable performance to TCN and is the best-performing machine-learning model overall. These findings are critical for investors, policymakers, mining companies, and governmental agencies to effectively anticipate market trends, mitigate risk, manage uncertainty, and make timely decisions and strategies regarding crude oil, gold, and silver markets.