Computer Training Research Articles

Hand-Written Analysis and Recognition Based on Deep Learning and Machine Learning Techniques

Abstract: This paper presents a comprehensive comparative analysis of Support Vector Machines (SVM), along with other machine learning models, and Convolutional Neural Networks (CNN) for Handwritten Analysis and Recognition. We investigate the performance of these models on a multi-class pairwise classification task using a large-scale handwritten dataset. Evaluation metrics such as accuracy, precision, and recall are utilized to measure their performance and compare them against the CNN model. The study aims to provide insights into the strengths and weaknesses of SVM, as well as other machine learning models, in comparison to CNN for handwritten analysis and recognition. SVM, known for its ability to handle complex non-linear relationships, offers good generalization and interpretability. However, it may face challenges in capturing intricate patterns in handwritten data. Other machine learning models, such as k-Nearest Neighbors (k-NN) and decision trees, also offer different advantages and limitations in this context. Through extensive experimentation, we compare the performance of SVM, k-NN, decision trees, and other relevant machine learning models with CNN on the same handwritten dataset. We analyze their accuracies, precision, and recall rates to evaluate their effectiveness in multi-class pairwise recognition of handwritten samples. Additionally, we discuss the computational requirements, training times, and interpretability associated with each model. The findings of this study provide valuable insights into the suitability of SVM and other machine learning models when compared to CNN for handwritten analysis and recognition tasks. The results shed light on the trade-offs between traditional machine learning approaches and deep learning architectures in this domain. These insights can guide researchers and practitioners in choosing the most appropriate model based on their specific requirements, computational resources, and interpretability needs.

Evaluation of Computer Training Programs Using Models Kirkpatrick to Increase the Competency Capacity of PerGunu Teachers

Computers have very multi-tasking functions. Computers can help complete various jobs, one of which is that they can be used in the teaching and learning process. With the sophistication of hardware and software on computers, teachers can present attractive, interesting, and fun learning methods and models. Unfortunately, not all teachers can make the most of computers as a solution in their teaching and learning process. One of them happened to teachers who were members of the Nahdlatul Ulama Teachers' Association (Pergunu) in Lampung Province. Therefore, the Regional Board of the Nahdlatul Ulama Teachers Association of Lampung Province has carried out computer training program activities for Gunu teachers. To determine the extent to which the success of this training can be felt by the participants, researchers conducted research on this computer training program using the Kirkpatrick model or what is known as the Kirkpatrick Four Levels Evaluation Model, namely evaluating the effectiveness of Kirkpatrick's training program by evaluating training programs based on four stages of evaluation, namely: 1) reaction, 2) learning, 3) behavior, and 4) results. This evaluation was carried out as an effort to improve computer training programs in the future so as to be able to increase the competency capacity of teachers in optimizing computers as a means of successful learning. The results of the research that has been carried out on this Computer Training Program, meanwhile, show that participants get the benefits to be applied in the learning process.

Peningkatan Kompetensi Guru SMK Melalui Pelatihan Komputer Perpajakan Berbasis Kasus

This service activity aims to improve the competence of accounting teachers in technical skills in the field of taxation. This is because SMK is a vocational education that is ready to provide skilled workers in their fields. In addition, the need for skilled labor in the field of taxation is currently a major need in the industrial world. The material taught in this training is in the form of a tax computer for filling out e-SPT PPh Article 21 and e-Faktur with a case approach. This training was attended by 39 people who are members of the Accounting Subject Teacher Conference (MGMP) throughout Yogyakarta. This activity is carried out for 2 (two) days. The first day is used for providing tax computer training, while the second day is used for exams which function to measure the level of understanding of the training participants. Based on the test results 39 people were declared competent or met certain qualifications so that it is hoped that the understanding and skills that have been obtained in this training can be taught back to students in their respective schools.

Pelatihan aplikasi Komputer Santri MDQH (Ma’had Darul Quran Wal-Hadist Almajidiah Asy-Syafi’ah) NWDI-Pancor

In public perception, religious scholars (Ulama) serve as highly trusted sources of information, particularly within Indonesian society. Therefore, it is imperative to ensure that Ulama possesses adequate digital literacy to convey accurate and reliable information. This study aims to provide computer training to the students (santri) of Ma'had MQDH NW Pancor, enabling them to master relevant computer technologies and applications. The training primarily focuses on internet literacy and Microsoft Office applications, employing a combination of lectures and hands-on practice. Ten participants from various levels of study were involved in the program, which took place at Laboratory 1, Faculty of Engineering, Hamzanwadi University. The results of the training revealed that 57.1% of the participants expressed high satisfaction levels, while 71.4% reported an increase in their understanding after completing the training. In conclusion, this computer training program has demonstrated positive benefits in enhancing the digital comprehension and skills of the students, preparing them to tackle digital challenges in the future.

PELATIHAN KOMPUTER DASAR TERHADAP KEMAMPUAN PENDIDIK PAUD DALAM PEMANFAATAN TIK

Obstacles to the use of ICT by PAUD educators are lack of access to ICT skills development, lack of knowledge about ICT, and lack of willingness of educators to utilize ICT. Knowing these conditions, Putera Batam University Lecturers took the initiative to organize a basic computer training program based on identifying the needs of PAUD educators in the area. The acceleration and expansion of Early Childhood Education (PAUD) services is one of the strategic policies rolled out by the Ministry of National Education. One of the strategies to improve the quality of PAUD educators that has been implemented so far is through the development of clusters. Microsoft Office application or better known as Office, of course almost everyone knows it, especially for those who usually deal with writing or reports. Members of the PAUD Cluster Work Center are no exception. For members of the PAUD Cluster Working Center, Microsoft Office 2019 is of course very helpful in their activities in preparing learning, carrying out learning activities, up to the evaluation and reporting stages of activities carried out.

Cognitive stimulation of executive functions through computational thinking

The healthy development of cognitive functions, including executive functions, has been shown to depend mainly on the experiences and learning opportunities of people, especially during childhood. Over the past few years, researchers have been studying the impacts of diverse types of interventions on children’s cognitive development in which computational thinking programs are a recent field. This pilot study evaluated the effect of computational thinking training based on the “Programming for Children” program on the executive functions of children aged 10 and 11 years: working memory, inhibition, and planning (N = 30). The results showed that children in the experimental group improved on tests of visuospatial working memory, cognitive inhibition, and sequential planning compared with the control group. However, tests of verbal working memory, memory strategy, and visual spatial planning did not show any observed changes. Although this was an exploratory study, and its findings should be interpreted cautiously due to the small sample size, the findings support the relevance and feasibility of conducting similar larger studies with larger samples.

Improving predictive models for rate of penetration in real drilling operations through transfer learning

The rate of penetration (ROP) is a key performance indicator in the oil and gas drilling industry as it directly translates to cost savings and emission reductions. A prerequisite for a drilling optimization algorithm is a predictive model that provides expected ROP values in response to surface drilling parameters and formation properties. The high predictive capability of current machine-learning models comes at the cost of excessive data requirements, poor generalization, and extensive computation requirements. These practical issues hinder ROP models for field deployment. Here we address these issues through transfer learning. Simulated and real data from the Volve field were used to pre-train models. Subsequently, these models were fine-tuned with varying retraining data percentages from other Volve wells and Marcellus Shale wells.Four out of the five test cases indicate that retraining the base model would always produce a model with a lower mean absolute error than training an entirely new model or using the base model without retraining. One was on par with the traditional approach. Transfer learning STL allowed for reducing the training data requirement from a typical 70 percent down to just 10 percent. In addition, transfer learning reduced computational costs and training time. Finally, results showed that simulated data could be used without real data or in combination with real data to train a model without trading off the model’s predictive capability.On top of our previous work Pacis et al. (2022) from a single transfer learning, we explored continuous transfer learning (CTL) in Alvheim field wells. Due to the inherent uncertainty and dynamics of drilling data, it was no surprise that continuous retraining further reduced the error than a single transfer learning paradigm. Moreover, we investigated the effect of drilled formations and input combinations on model performance.

Energy-Aware, Device-to-Device Assisted Federated Learning in Edge Computing

The surging of deep learning brings new vigor and vitality to shape the prospect of intelligent Internet of Things (IoT), and the rise of edge intelligence enables provisioning real-time deep neural network (DNN) inference services for mobile users. To perform efficient and effective DNN model training in edge computing environments while preserving training data security and privacy of IoT devices, federated learning has been envisioned as an ideal learning paradigm for this purpose. In this paper, we study energy-aware DNN model training in edge computing. We first formulate a novel energy-aware, Device-to-Device (D2D) assisted federated learning problem with the aim to minimize the global loss of a training DNN model, subject to bandwidth capacity on an edge server and energy capacity on each IoT device. We then devise a near-optimal learning algorithm for the problem when the training data follows the i.i.d. data distribution. The crux of the proposed algorithm is to explore using the energy of neighboring devices of each device for its local model uploading, by reducing the problem to a series of weighted maximum matching problems in corresponding auxiliary graphs. We also consider the problem without the assumption of the i.i.d. data distribution, for which we propose an efficient heuristic algorithm. We finally evaluate the performance of the proposed algorithms through experimental simulations. Experimental results show that the proposed algorithms are promising.

Optimizing Aggregation Frequency for Hierarchical Model Training in Heterogeneous Edge Computing

Federated Learning (FL) has been widely used for distributed machine learning in edge computing. The centralized aggregation server in FL is inclined to be the communication bottleneck and single point of failure. To solve the drawback, hierarchical model training frameworks like Hierarchical Federated Learning (HFL) and E-Tree learning have been proposed. Existing works lack solutions to quantitatively determine the aggregation frequencies of the edge devices at various layers for a hierarchical model training framework. Thus, in this paper, we propose a resource-based aggregation frequency controlling method, termed RAF, which determines the optimal aggregation frequencies of edge devices to minimize the loss function according to heterogeneous resources. Our proposed method can alleviate the waiting time and fully utilize the resources of the edge devices. Besides, RAF dynamically adjusts the aggregation frequencies at different phases during the model training to achieve fast convergence speed and high accuracy. We evaluated the performance of RAF via extensive experiments with real datasets on our self-developed edge computing testbed. Evaluation results demonstrate that RAF outperforms the benchmark approaches in terms of learning accuracy and convergence speed.

HardGBM: A Framework for Accurate and Hardware-Efficient Gradient Boosting Machines

Gradient boosting machine (GBM) is a powerful and widely used type of ensemble machine learning methods, among which the most famous one is XGBoost. However, the cost of running large GBMs on hardware could become prohibitive given stringent resources. Ensemble reduction on boosting ensemble is intrinsically hard because member models are constructed in a sequential order, where the training targets for latter ones depend on the performance of the former ones. In this work, a GBM reduction framework is proposed for the first time to tackle the problem. For the first time, the framework supports automatic hardware implementation of regression tree ensembles. Experiments on 24 datasets from various applications demonstrate that our method reduces overall area utilization by 81.60% (80.64%) and power consumption by 21.15% (19.06%), while exceeding or successfully maintaining the performance level comparing with the original XGBoost (LightGBM) ensembles. In comparative experiments, to attain approximately the same accuracy level as our framework or XGBoost, deep learning-based solutions require >52.7× footprints, 6.0× power consumption, and 1.4× training time. Equipped with tunable parameters, the framework is expected to seek a Pareto optimal front considering hardware resource limitation, accuracy and stability, and computation (training) efficiency.

Information technology for creating intelligent computer programs for training in algorithmic tasks. Part 2: Research and implementation

Information technologies, particularly artificial intelligence methods, involve more and more deeply into all spheres of human activity: science, technology, art, and education. Ukraine also has sufficient potential and needs to develop educational support, which is the subject of this paper. The work aims to demonstrate the components of information technology for the creation of Intelligent Tutoring Systems (ITS), which are involved in studying various engineering disciplines. The work includes methods of system analysis, mathematical and simulation modeling, technical diagnostics, and artificial intelligence. The proposed models and methods are implemented in ITS prototypes for teaching mathematics, programming, and the automatic control theory. The Intelligent Tutoring Systems were implemented in the educational process of KhAI University and other institutions in Ukraine, Great Britain, Austria, and China. Experimental studies have shown increased student learning success rates using ITS compared to traditional methods. Improved and adapted for computer training methods of technical diagnostics, Bayesian networks, and developed models of algorithmic tasks, the learning process and the learner are valuable from a scientific point of view. In a practical sense, the obtained results can be used to create new specialized ITSs and build an expandable common learning platform integrating the basic disciplines of a specific educational field.

Enhancing Digital Skills for the Deaf in Lahore: A Study of Training Program Effectiveness

This research aimed to explore the effectiveness of computer courses offered in different vocational institutes as perceived by persons with hearing impairment (PWHI). The research utilized a quantitative paradigm with a descriptive design. The target population comprised PWHI who had undertaken computer training. A convenience sampling technique was used to collect data from 180 PWHI enrolled in computer courses across four institutions in Lahore, Pakistan. SPSS software-enabled analysis of the data using frequency distributions for demographic information and respondent statements. Findings revealed most participants were satisfied with their acquired knowledge from the training courses, without facing substantial difficulties. Furthermore, respondents conveyed an understanding of the concepts and perceived the skills as beneficial for employment. Responses relating to instruction indicated most instructors used sign language during lessons and provided adequate time for supervised practice. Recommendations from the study advise enhancing course content to include skills to enable earning potential from freelancing.

Pelatihan Komputer Bagi Kader PKK Desa Kaibon

Family Welfare Empowerment or often shortened to PKK is a movement that has been built from the center to the village and its existence and benefits have been felt by the community. This PKK activity is reflected in the 10 PKK Main Programs, one of which isthe program focuses on health and special attention is paid to the health of mothers and children. These trained PKK cadres are not only seen from the way they handle maternal and child health but also must be supported by their ability to use computers in the recording and data collection process. However, not all PKK Cadres, especially in villages, are proficient in using computers, for example, PKK Cadres in Kaibon Village. In fact, many cadres are new to computers and cannot use standard programs such as Microsoft Excel and Microsoft Word. Kaibon Village PKK Cadres in operating computers to support good and smooth performance in terms of administration and data collection.Keywords: Computer Training, Kaibon Village PKK cadres

Communication Strategy of Jaya's Main Course and Training Institute in Attracting Students to Computer Training

Computer training aims to develop the ability of everyone, especially in the field of computers, where computers have become a big part of a technology that cannot be separated from human life activities, both work, school and other activities. This study aims to determine how the communication strategy of the main training course institution in attracting student interest in computer training. The research method used is descriptive qualitative research method with a research focus on communication strategies with the AIDA model, namely Attention, Interest, Desire and Action. Data collection techniques are done through observation and in-depth interviews with informants face-to-face so that it will produce clear data. The research shows that the communication strategy carried out is to use an advertising strategy which includes promotion through social media instagram, facebook, youtube and website and direct marketing. The most effective strategy is direct marketing which fosters relationships so that cooperation with several schools is established. In this case, students at the Utama Jaya Course and Training Institute (LKP) have increased. Based on the results of in-depth interviews conducted directly by researchers to the management of the institution, that the communication strategy that attracts the most prospective training students is direct marketing to schools where from this collaboration the school recommends to its students, especially vocational students majoring in Computer and Network Engineering (TKJ) and Software Engineering (RPL) to take computer training at the Main Jaya Training Course Institute (LKP) with the aim that students master the material to face computer practice exams.

Machine learning based prediction of phase ordering dynamics.

Machine learning has proven exceptionally competent in numerous applications of studying dynamical systems. In this article, we demonstrate the effectiveness of reservoir computing, a famous machine learning architecture, in learning a high-dimensional spatiotemporal pattern. We employ an echo-state network to predict the phase ordering dynamics of 2D binary systems-Ising magnet and binary alloys. Importantly, we emphasize that a single reservoir can be competent enough to process the information from a large number of state variables involved in the specific task at minimal computational training cost. Two significant equations of phase ordering kinetics, the time-dependent Ginzburg-Landau and Cahn-Hilliard-Cook equations, are used to depict the result of numerical simulations. Consideration of systems with both conserved and non-conserved order parameters portrays the scalability of our employed scheme.

VISUALISATION OF HISTOLOGICAL SPECIMENS AND TASKS AS A TECHNIQUE TO IMPROVE KNOWLEDGE, SKILLS AND ABILITIES OF MEDICAL STUDENTS

The purpose of this study is to explore the role of visualization of histological tasks in improving the quality of knowledge, abilities, and skills of medical students by summarizing the work experience of the Department of Histology, Cytology, and Embryology.
 In the field of higher medical education, the discipline of "Histology, Cytology, and Embryology" is crucial as it forms the foundation of students' understanding of the microscopic and functional morphology, structure, development, and regeneration of human cellular, tissue, and organ systems. However, first-year medical students often struggle with the vast amount of information presented to them.
 To optimize the teaching of this discipline and enhance the quality of students' knowledge, modern teaching methods and technologies are employed, including the use of microscopes with video apps, multimedia projectors, and educational resources that take advantage of digital technologies. The staff of the Department of Histology, Cytology, and Embryology has also created educational resources that broaden the possibilities of digital technologies for visualization of microscopic structures: collections of photomicrographs of histological preparations and electronograms, multimedia presentations, educational films, animated videos and Internet resources (computer training programs, electronic atlases, etc.).
 Students complete visualized tasks for the diagnosis of histological preparations by establishing correspondence between digital markings of structures on images of micro-samples and proposed lists of histological structures, or by diagnosing drugs through determining their colour and histological structures. These visual tasks expand the possibilities of studying histology, cytology, and embryology, eliminating randomness in assessing students' knowledge, abilities, and skills.
 In practical histology classes, moulages of histological preparations, structures of human embryonic development, facial development, teeth, and their supporting apparatus, thematic stands, and wet preparations are used to enhance the visualization of histological structures.
 By visualizing histological samples and tasks in practical classes, students can better assimilate educational material and acquire the skills necessary for differential diagnosis based on the analysis of microsamples. The experiences of the Department of Histology, Cytology, and Embryology can serve as an example of the positive impact of this approach on the quality of medical education.

Management of Teacher Professionalism Development: A Multi-Site Study of State Madrasah Aliyah in Langkat Regency

This study aims to analyze the management of teacher professional development at Madrasah Aliyah Negeri throughout Langkat Regency. This study used a multi-site qualitative method where data collection techniques used interviews, observation, and documentation. The results of this study indicate that: 1) Management of Teacher Professionalism Development in State Madrasah Aliyah throughout Langkat Regency is directed at efforts to develop teacher professionalism in the form of increasing expertise, and skills, broadening horizons, coaching carried out on the initiative and in collaboration with the regional government and the District Ministry of Religion. Langkat through planning, implementing, supervising, and evaluating activities contained in various forms of activities, including K-13 socialization training and the covid curriculum, computer training, MGMP, supervision, making efforts for teachers to attend seminars, workshops, also seeking to add to the collection of books -books in the library.

Pojok Literasi-Ku Berbasis Pendidikan Masyarakat Sebagai Upaya Meningkatkan Kesadaran Literasi di Desa Bumi Sari Kabupaten Nagan Raya Aceh

This service activity aims to increase the literacy awareness of the residents of Bumi Sari Village, Nagan Raya Regency, through a community education-based My Literacy corner. The method of implementing this service used mentoring and teaching. Partners in this service were the Nagan Raya Library Service, the Nagan Raya Education Office, the Environmental Care Alliance (APEL), the Communication, Informatics, and Statistics Service (KOMINFOTIK), and the Bumi Sari Village Government. The instrument used to measure the success of this community service activity was in the form of a questionnaire given to each person in charge of the literacy corner. This dedication resulted in implementing five literacy corner programs in Bumi Sari Village: Reading Corner, Writing Corner, Internet Introduction Corner, Computer Training Corner, and MS Introduction Corner Office. There was an increase in literacy among the people of Bumi Sari Village by 36.35%. This increase could be seen from the number of people before the activity, where only 34.41% understood literacy, then it increased to 70.76% after this community service activity. The response of the people of Bumi Sari Village, especially the parents, towards the literacy corner program was very good, judging from the number of children participating in learning in the literacy corner. This service activity could increase literacy awareness in Bumi Sari Village.

Perancangan Sistem Akuntansi Penggajian ‘Si Warna’ Di Pt Harapan Kurnia Textile Indonesia

The research conducted at PT Harapan Kurnia Textile Indonesia aims to produce accurate attendance information, fast, effective and efficient payroll management, provide financial reports for companies and payslips for employees appropriately. To identify the problems faced, primary and secondary data collection as well as qualitative descriptive approaches were used in this study. The method for developing software that uses waterfall with HTML to create pages on websites and databases is MYSQL. The results of this study show that the payroll management process has run well but still uses Microsoft Excel manually which is prone to errors. So to overcome these problems, an application design was made in the form of Flowmap, ERD and DFD and designed a payroll accounting system with web-based PHP programming language code. The conclusion of this study is that the color becomes a problem solver that will help HRD and financial staff in managing payroll to be faster, more precise and efficient and produce accurate payslips and financial statements. The advice given by the author is to conduct computer training payroll applications, perform system maintenance to prevent damage / errors in the system and back up data so that data remains safe if the computer is damaged.
 Keywords : Design, Accounting System, Payroll

A Study of Compressed Language Models in Social Media Domain

Transfer learning from large-scale language models is witnessing incredible growth and popularity in natural language processing (NLP). However, operating these large models always requires a huge amount of computational power and training effort. Many applications leveraging these large models are not very feasible for industrial products since applying them into power-scarce devices, such as mobile phone, is extremely challenging. In this case, model compression, i.e. transform deep and large networks to shallow and small ones, is becoming a popular research trend in NLP community. Currently, there are many techniques available, such as weight pruning and knowledge distillation. The primary concern regarding these techniques is how much of the language understanding capabilities will be retained by the compressed models in a particular domain? In this paper, we conducted a comparative analyses between several popular large-scale language models, such as BERT, RoBERTa, XLNet-Large and their compressed variants, e.g. Distilled BERT, Distilled RoBERTa and etc, and evaluated their performances on three datasets in the social media domain. Experimental results demonstrate that the compressed language models, though consume less computational resources, are able to achieve approximately the same level of language understanding capabilities as the large-scale language models in the social media domain.