Conventional Learning Methods Research Articles

Effective Active Learning Method for Spiking Neural Networks.

A large quantity of labeled data is required to train high-performance deep spiking neural networks (SNNs), but obtaining labeled data is expensive. Active learning is proposed to reduce the quantity of labeled data required by deep learning models. However, conventional active learning methods in SNNs are not as effective as that in conventional artificial neural networks (ANNs) because of the difference in feature representation and information transmission. To address this issue, we propose an effective active learning method for a deep SNN model in this article. Specifically, a loss prediction module ActiveLossNet is proposed to extract features and select valuable samples for deep SNNs. Then, we derive the corresponding active learning algorithm for deep SNN models. Comprehensive experiments are conducted on CIFAR-10, MNIST, Fashion-MNIST, and SVHN on different SNN frameworks, including seven-layer CIFARNet and 20-layer ResNet-18. The comparison results demonstrate that the proposed active learning algorithm outperforms random selection and conventional ANN active learning methods. In addition, our method converges faster than conventional active learning methods.

Bi-Directional Ensemble Feature Reconstruction Network for Few-Shot Fine-Grained Classification.

The main challenge for fine-grained few-shot image classification is to learn feature representations with higher inter-class and lower intra-class variations, with a mere few labelled samples. Conventional few-shot learning methods however cannot be naively adopted for this fine-grained setting - a quick pilot study reveals that they in fact push for the opposite (i.e., lower inter-class variations and higher intra-class variations). To alleviate this problem, prior works predominately use a support set to reconstruct the query image and then utilize metric learning to determine its category. Upon careful inspection, we further reveal that such unidirectional reconstruction methods only help to increase inter-class variations and are not effective in tackling intra-class variations. In this paper, we introduce a bi-reconstruction mechanism that can simultaneously accommodate for inter-class and intra-class variations. In addition to using the support set to reconstruct the query set for increasing inter-class variations, we further use the query set to reconstruct the support set for reducing intra-class variations. This design effectively helps the model to explore more subtle and discriminative features which is key for the fine-grained problem in hand. Furthermore, we also construct a self-reconstruction module to work alongside the bi-directional module to make the features even more discriminative. We introduce the snapshot ensemble method in the episodic learning strategy - a simple trick to further improve model performance without increasing training costs. Experimental results on three widely used fine-grained image classification datasets, as well as general and cross-domain few-shot image datasets, consistently show considerable improvements compared with other methods.

Penerapan Metode Take and Give dalam Meningkatkan Keterampilan Sosial Peserta Didik pada Kurikulum Merdeka

The Independent Curriculum provides freedom to teachers and students in the learning process by focusing on the development of holistic competencies that include cognitive, affective, and psychomotor aspects. This take and give learning method is in line with the principles of the Independent Curriculum. This method emphasizes active and collaborative interaction between learners. This study aims to explore the application of the take and give method in learning the Qur'an Hadith at MTs Al-Azhar Sempu Banyuwangi to improve students' social skills. The social skills in question include communication, cooperation, empathy, and problem-solving, which are often less accommodated in conventional learning methods. This research uses a qualitative approach with a case study type. Data collection is carried out through observation, interviews and documentation, then the data is analyzed with the stages of data collection, data condensation, data display, and conclusions drawing/verifying. Checking the validity of the data using triangulation of sources and methods. The results of the study show that the take and give method is effective in improving students' social skills through improving communication skills, cooperation, and empathy, as well as applying the values of the Qur'an and Hadith in daily life. The contribution of this research to the development of a more holistic learning method and relevant to the principles of the Independent Curriculum. The take and give method is expected to be an innovative method to improve students' social skills.

Pengaruh Aplikasi Matematika Interaktif pada Perkembangan Kemampuan Pemecahan Masalah Matematika

This study aims to determine whether there is an effect of using interactive math applications on the development of mathematical problem solving skills in students. This study was conducted by collecting data from students who used interactive math applications as a tool in learning mathematics, and compared with the control group who used conventional learning methods. Data were collected through math problem solving ability tests before and after the use of interactive math applications. The results showed that the use of interactive math applications significantly improved students' mathematical problem solving ability compared to the control group that used conventional methods. In addition, interactive math applications also help students in understanding math concepts better and increase their interest in this subject.

Penerapan Media Pembelajaran Animasi Menggunakan Animiz untuk Meningkatkan Hasil Belajar Siswa pada Mata Pelajaran Bahasa Indonesia Kelas X SMAN 1 Ciawigebang

This research aims to utilize animation learning media developed using Animiz software, to improve students' academic achievement in Indonesian Language subjects in class X of SMAN 1 Ciawigebang. The approach used in this study is quantitative with experimental methods, especially using Quasi Experiment design. The research design applied is a nonequivalent control group design. This study adopts a Quasi Experiment design, in which two groups of classes are the subjects of the research: X-G class as the experimental class and X-J class as the control class. The experimental class uses Animiz media, while the control class follows conventional learning methods. Student learning outcome data is collected through pretest and posttest which involves observation and tests. The data analysis technique uses the t-test. The results of the study show that the application of Animiz Animation Maker animation learning media has succeeded in increasing students' understanding of poetry material in Indonesian Language subjects in class X of SMAN 1 Ciawigebang

Application Of Wingeom-Assisted Think Pair Share Model to Students' Mathematical Communication Skills

The cooperative learning model of Think Pair Share assisted by WinGeom provides an opportunity for students to develop their mathematical communication skills during the learning process. This study aims to determine the effect of implementing the Think Pair Share cooperative learning model assisted by WinGeom on mathematical communication skills among junior high school students in Geometry. The research uses a quasi-experimental design with pretest-posttest 1, 2, and 3. The population of this study consists of grade VIII students at SMP Negeri 7 Serang City. The sample includes students from classes VIII-A and VIII-B class was selected randomly. Data collection techniques used mathematical communication skills tests. This study compares Think Pair Share assisted by WinGeom with conventional learning methods. ANOVA was used to analyze the effectiveness of the two learning methods. ANOVA test results for each posttest indicate that Think Pair Share assisted by WinGeom has a different effect compared to conventional learning. The study results show a significant difference between the average mathematical communication skills of students in the experimental class and the control class. This means that the cooperative learning model of Think Pair Share assisted by WinGeom has an impact on junior high school students' mathematical communication skills. In the future, this learning method could be a solution for teachers worldwide to improve students' learning achievements, particularly in mathematical communication skills. Keywords: Think pair share model, WinGeom, Mathematical Communication, Cooperative

Optimal Feature Picking for Intrusion Detection on The Basis of Explainable Artificial Intelligence

Traditional security mechanisms for intrusion detection are strong, and the use of machine learning algorithms to detect intrusion into IoT networks has increased. As a result, the efficiency of the IoT network has increased and an effective model can be developed to detect attacks on the network. UNSW-NB15 is an IoT-based publicly available dataset for traffic data that contains general activity and malicious activity. Using this dataset, the key features were selected and these features fed different classifiers for training and classifying the attack in the network. Ann model performance levels of accuracy, the ML classifier, and accompanying algorithms for constructing a digital security system based on packet flow in-network, and features that really can monitor anomalous botnet behaviors proven to be highly efficient. The decision-making model for the classification was visualized using proven Explainable AI (XAI) approaches utilizing Scikit-Learn, LIME, ELI5, and SHAP libraries to boost explainability in classification predictions. The findings showed that XAI is realistic and profitable since security expertise and specialists stand to benefit greatly from combining conventional machine learning methods with Explainable AI (XAI) methodologies.

Training-free approach to constructing ensemble of local experts

Conventional machine learning methods assume that a central server holds the entire training dataset for training a global model. However, in many real-world situations, datasets are collected from individual local nodes in distributed environments. Gathering these datasets often poses challenges owing to various practical constraints such as data security and privacy issues, storage limits, and transmission costs. When data distributions differ among local nodes, local experts trained on individual local nodes exhibit poor generalization performance for new instances. In this study, we propose a training-free method for constructing an ensemble of local experts without accessing datasets. For a query instance, we apply uncertainty quantification and out-of-distribution (OOD) detection to calculate the uncertainty and OOD scores of the predictions from individual models. We then determine the weights of the models by assigning a lower weight to those with higher uncertainty and OOD scores. Using these weights, the predictions of the individual models are aggregated to obtain a final prediction. In contrast to existing methods that impose specific application requirements, the proposed method can be applied in situations in which only individual predictions from local models are available for use when making a prediction for a query instance. We evaluate the effectiveness of the proposed method using image classification benchmark datasets.

A Transfer-Learning-Like Neural Dynamics Algorithm for Arctic Sea Ice Extraction

Sea ice plays a pivotal role in ocean-related research, necessitating the development of highly accurate and robust techniques for its extraction from diverse satellite remote sensing imagery. However, conventional learning methods face limitations due to the soaring cost and time associated with manually collecting sufficient sea ice data for model training. This paper introduces an innovative approach where Neural Dynamics (ND) algorithms are seamlessly integrated with a recurrent neural network, resulting in a Transfer-Learning-Like Neural Dynamics (TLLND) algorithm specifically tailored for sea ice extraction. Firstly, given the susceptibility of the image extraction process to noise in practical scenarios, an ND algorithm with noise tolerance and high extraction accuracy is proposed to address this challenge. Secondly, The internal coefficients of the ND algorithm are determined using a parametric method. Subsequently, the ND algorithm is formulated as a decoupled dynamical system. This enables the coefficients trained on a linear equation problem dataset to be directly generalized to solve the sea ice extraction challenges. Theoretical analysis ensures that the effectiveness of the proposed TLLND algorithm remains unaffected by the specific characteristics of various dataset. To validate its efficacy, robustness, and generalization performance, several comparative experiments are conducted using diverse Arctic sea ice satellite imagery with varying levels of noise. The outcomes of these experiments affirm the competence of the proposed TLLND algorithm in addressing the complexities associated with sea ice extraction.

Utilizing Quizzes on the Duolingo to Promote Students Vocabulary Mastery

This research aims to find out the effectiveness of Duolingo games to help students' vocabulary mastery and describe students' perceptions of learning English using the Duolingo application. One of problems found in learning English is a lack of vocabulary that can make it difficult for students to understand texts, speak, listen, and write effectively. Limited student emphasis on acquiring vocabulary in class. In addition, schools that still use conventional learning methods make students bored. This research was conducted using qualitative methods using questionnaires and interviews to determine student perceptions. This research was conducted in one of Islamic senior High School in Tasikmalaya involving 33 students from class X participants. The results of this study indicate that students give a positive perception about the Duolingo application because most students (79%) agree more with learning English vocabulary using the Duolingo application as a medium of learning vocabulary. Apart from it, there are several advantages that can be drawn from using the Duolingo application as a medium for learning English vocabulary: 1) the Duolingo application make the students easy to understand the material; 2) the Duolingo application made the students more enthusiastic in learning English vocabulary; 3) the Duolingo application can eliminate boredom in learning; and 4) the Duolingo application encourage the students for generating new idea in learning.

A novel multitask learning algorithm for tasks with distinct chemical space: zebrafish toxicity prediction as an example

Data scarcity is one of the most critical issues impeding the development of prediction models for chemical effects. Multitask learning algorithms leveraging knowledge from relevant tasks showed potential for dealing with tasks with limited data. However, current multitask methods mainly focus on learning from datasets whose task labels are available for most of the training samples. Since datasets were generated for different purposes with distinct chemical spaces, the conventional multitask learning methods may not be suitable. This study presents a novel multitask learning method MTForestNet that can deal with data scarcity problems and learn from tasks with distinct chemical space. The MTForestNet consists of nodes of random forest classifiers organized in the form of a progressive network, where each node represents a random forest model learned from a specific task. To demonstrate the effectiveness of the MTForestNet, 48 zebrafish toxicity datasets were collected and utilized as an example. Among them, two tasks are very different from other tasks with only 1.3% common chemicals shared with other tasks. In an independent test, MTForestNet with a high area under the receiver operating characteristic curve (AUC) value of 0.911 provided superior performance over compared single-task and multitask methods. The overall toxicity derived from the developed models of zebrafish toxicity is well correlated with the experimentally determined overall toxicity. In addition, the outputs from the developed models of zebrafish toxicity can be utilized as features to boost the prediction of developmental toxicity. The developed models are effective for predicting zebrafish toxicity and the proposed MTForestNet is expected to be useful for tasks with distinct chemical space that can be applied in other tasks.Scieific contributionA novel multitask learning algorithm MTForestNet was proposed to address the challenges of developing models using datasets with distinct chemical space that is a common issue of cheminformatics tasks. As an example, zebrafish toxicity prediction models were developed using the proposed MTForestNet which provide superior performance over conventional single-task and multitask learning methods. In addition, the developed zebrafish toxicity prediction models can reduce animal testing.

Random forests for the analysis of matched case–control studies

BackgroundConditional logistic regression trees have been proposed as a flexible alternative to the standard method of conditional logistic regression for the analysis of matched case–control studies. While they allow to avoid the strict assumption of linearity and automatically incorporate interactions, conditional logistic regression trees may suffer from a relatively high variability. Further machine learning methods for the analysis of matched case–control studies are missing because conventional machine learning methods cannot handle the matched structure of the data.ResultsA random forest method for the analysis of matched case–control studies based on conditional logistic regression trees is proposed, which overcomes the issue of high variability. It provides an accurate estimation of exposure effects while being more flexible in the functional form of covariate effects. The efficacy of the method is illustrated in a simulation study and within an application to real-world data from a matched case–control study on the effect of regular participation in cervical cancer screening on the development of cervical cancer.ConclusionsThe proposed random forest method is a promising add-on to the toolbox for the analysis of matched case–control studies and addresses the need for machine-learning methods in this field. It provides a more flexible approach compared to the standard method of conditional logistic regression, but also compared to conditional logistic regression trees. It allows for non-linearity and the automatic inclusion of interaction effects and is suitable both for exploratory and explanatory analyses.

Implementasi Metode Contextual Teaching And Learning (CTL) untuk Meningkatkan Hasil Belajar Siswa Kelas IV pada Mata Pelajaran Pendidikan Kewarganegaraan di SDN 09 Rangkang

This study aims to explore the effect of the Contextual Teaching and Learning (CTL) method on improving student learning outcomes in Civic Education subjects in grade IV of SD Negeri 09 Rangkang. The background of this study is the low student learning outcomes caused by the use of conventional learning methods that are less effective. The method used is classroom action research (CAR) with two cycles, involving 30 students as subjects. Data were collected through observation, tests, and student worksheets. The results showed that the application of the CTL method can significantly increase student active participation and their learning outcomes. The average student score increased from 70 in cycle I to 85 in cycle II. In conclusion, the CTL method is effective in improving student learning outcomes and can be a good alternative for teachers in teaching complex materials. This study recommends that teachers apply the CTL method consistently to improve the quality of learning in the classroom. Thus, the CTL method can be a practical solution to the problem of low learning outcomes in elementary schools and help create a more interesting and interactive learning environment for students.

Optimalisasi Kemampuan Menulis Teks Eksplanasi dan Keaktifan Siswa Kelas 11 Melalui Model Pembelajaran Problem Based Learning

Purpose – This study aims to explore the effect of Problem-Based Learning Model (PBL) on the learning outcomes of students in grade XI of SMAN 4 Medan in writing an explanatory text in Indonesian language subject. By focusing on the effectiveness of PBL in improving students' writing skills, this study provides theoretical contributions to the development of learning theories as well as practical benefits for teachers and schools in improving the quality of learning.Method – The study used a quasi-experimental design with an experimental group and a control group. The approach was Problem Based Learning (PBL) Model. The research was conducted at SMAN 4 Medan involving students of grade XI. Data were collected through an explanatory text writing test conducted before (pretest) and after (posttest) the application of PBL. The analysis was conducted by comparing the posttest results between the experimental group using PBL and the control group using conventional learning methods. The results were analyzed to assess the effectiveness of PBL in improving students' writing ability.Findings – The results showed that the Problem-Based Learning Model (PBL) significantly improved students' learning outcomes in writing explanatory texts. Proven by the average value of posttest in the experimental group which is higher than the control group, which proves the effectiveness of PBL in the context of Indonesian language learning in class XI SMAN 4 Medan. Research Implications – The Problem-Based Learning (PBL) model enhances students' writing skills in explanatory texts at the senior high school level. This study, limited to one school, does not reflect broader educational variations in Indonesia. It focused solely on PBL's impact on writing skills without a thorough exploration of other influencing factors. Future research should expand the sample size and include diverse contextual elements to more robustly verify these results.

THE EFFECTIVENESS OF DIGITAL TEACHING MATERIALS TO IMPROVE LITERARY LITERACY SKILLS IN PRIMARY SCHOOLS

This research is motivated by the challenge of improving the literary literacy skills of elementary school students, which are often considered boring using conventional learning methods. The purpose of this study was to analyze the effectiveness of digital teaching materials in improving literary literacy skills, as well as the interests and motivations of 5th grade elementary school students in Majalengka Regency, West Java.. The research method used was quasi-experimental with a pre-test and post-test design in two groups: the Experimental Group which used digital teaching materials and the Control Group which used conventional methods. The research sample consisted of 60 grade 5 students, with each group consisting of 30 students. Data collection techniques include literary literacy tests, interest and motivation questionnaires, and classroom observations. Data analysis was done using statistical tests to compare changes in pre-test and post-test scores between the two groups. The research results showed that the use of digital teaching materials significantly improved the literary literacy skills of students in the Experimental Group compared to the Control Group. The average literary literacy score in the Experimental Group increased from 45 to 72, while in the Control Group, it only increased from 43 to 55. Students' interest and motivation also increased more significantly in the Experimental Group. In conclusion, digital teaching materials effectively improve students' literary literacy skills, interest, and motivation. The implication is that teachers and schools are advised to integrate technology into literature learning, and policymakers need to support using digital teaching materials in the national curriculum to improve the quality of education.

The Influence of the Team Games Tournament (TGT) Learning Model Assisted by Wordwall on the Scientific Literacy of Class V Elementary School Students

This study aims to present the results of a study that compares the effectiveness of two learning methods, namely the Team Games Tournament (TGT) assisted by Wordwall and conventional learning methods, in improving the science literacy skills of grade V elementary school students. The research method uses a quasi-experimental method by using an experimental class applying TGT assisted by Wordwall, while the control class uses conventional learning methods. Data obtained from both groups of students were analyzed to evaluate the improvement of science literacy skills. The results showed that the experimental class experienced a significant improvement in science literacy skills compared to the control class. Wordwall-assisted TGT approach creates a more interactive, collaborative, and engaging learning environment for students, which encourages active participation and increases their motivation to learn. On the other hand, conventional methods show some students experiencing a decline in ability, highlighting the need for a more varied and interactive approach to learning. These results underscore the importance of innovation in learning methods to improve the quality of science education in primary schools, with great potential to apply the Wordwall-assisted TGT model widely in this context.

Deep reinforcement learning for stock prediction

Due to its chaotic and high volatility character, as well as many other uncertainties from reality, predicting the price of stock market is always a challenging goal to achieve. Due to those characteristics mentioned above, it could be regarded as a classification problem, and then many methods using different machine learning tools could be well applied to solve the challenging problem. Within these methods, deep neural network is a popular and highly noticed one in recent years. This is mainly because of its unique advantages compared to the more conventional machine learning methods, the highly complex nonlienearity and deep nonlinear topologies, to appropriately describe the complex situations. Later, after adding advantages of reinforcement learning to enable the model of the advantages to improve feature dimensions, the deep reinforcement learning method is well proposed to improve the performance. Deep reinforcement learning is a method to combine the advantages of deep learning and the advantages of reinforcement learning, and this paper will discuss its characters and advantages, and finally talk about its limitations and future development.

Effectiveness of Using Macbox Media to Improve Students’Grammar Mastery

Grammar as essential aspect of English component helps students to recognize pattern and avoid errors. Alternative media is needed to make grammar learning more successful since students find it challenging to master grammar when it is taught using conventional method. This research aimed to find out about how effectively Macbox media works to help students in improving their grammar mastery. True-experimental research design was used in this research by using class X MA Matholi'ul Huda Bugel students as the research population. The two classes used as sample are 36 students of XG as the experimental class and 38 students of XH as the control class. Both classes received pre-test, treatment and post-test in order to gather research data. Then, SPSS 25.0 was utilized to examine the data outcomes. The findings of the research showed that the differences were seen in the two classes' average scores. For pre-test, the experimental class's average pre-test score was 48.47, while the control class' was 48.02. For post-test, the experimental class’s average score was 75,41 and the control class’ was 62,76. The significance value for both classes was revealed by the independent sample t-test analysis result, which were 0.000 (sig (2-tailed) < 0,05). It means that there are the significant differences between both classes. The result and discussion show the use of Macbox media is significantly effective in improving students’ grammar mastery. It means that the use of Macbox media is more effective in improving students' grammar mastery compared to the use of conventional learning method

Automatic damage detection and data completion for operational bridge safety using convolutional echo state networks

Structural response data might be partially missing during acquisition and transmission by a structural health monitoring (SHM) system, affecting subsequent structural damage identification. Conventional deep learning methods require a substantial amount of data and have low training efficiency. With this in mind, this paper proposes a hybrid approach for data reconstruction and damage identification using an echo state network embedded with convolutional layers. The network has a training-free reservoir layer and focuses on historical information and multi-channel spatial features. As data redundancy occurs in most measured structural vibration responses, the extended Frobenius norm-principal component analysis method is applied to select valuable samples for updating the network, which greatly enhances the training efficiency. Experimental and numerical data from a physical bridge model is used to verify the data reconstruction capability and damage identification accuracy of the proposed method, and its performance is further compared with several existing deep learning methods.

Eligibility traces in an autonomous soccer robot with obstacle avoidance and navigation policy

As the thematic literature of machine learning suggests, reinforcement learning falls between the two methods of supervised learning and unsupervised learning. In this method, the learning agent receives a reward or punishment from the environment according to its action. Therefore, the learning agent interacts with the environment through trial and error and learns to choose the optimal action to achieve its goal. In the meantime, the eligibility traces are considered as one of the main mechanisms of reinforcement learning in receiving delayed rewards. In the use of conventional reinforcement learning methods, when the learning agent achieves a goal, only the value function of the last state-action pair changes, but all the trace states are affected based on the eligibility traces. In other words, the delayed rewards are distributed throughout the trace. Like the ant pheromone effect, this method can increase learning speed empirically to some extent. A soccer robot on the field encounters moving obstacles such as balls, rival robots, and home robots, and fixed obstacles such as gates and flags, so its environment is in a very dynamic state. Therefore, due to the dynamic environment, it is an important issue for autonomous soccer robots to avoid obstacles and should be considered in real play. The main idea of this study is to determine the appropriate traces for the robot to move towards the ball and ultimately to score with the approach of avoiding obstacles in simulating a real soccer match. The desired results obtained from a game played indicate a high level of online experience and decision-making power in the face of new situations.