Jurnal ELTIKOM
Not a member yet
154 research outputs found
Sort by
K-Means Clustering Method For Customer Segmentation Based On Potential Purchases
Full text linkThe rapid growth in customer data has driven companies to develop smarter and more effective marketing strategies. One efficient approach is customer segmentation, which involves dividing a market or group of customers into smaller segments based on similar characteristics or behaviors. Customer segmentation improves understanding of customer needs, preferences, and behavior. This study uses customer segmentation based on purchase potential at Fast Moving Consumer Goods (FMCG). Analyzing potential purchases can help identify market opportunities, implement more effective pricing, target promotions, manage stock and distribution, and develop new products to enhance customer satisfaction. The most commonly used segmentation method is the K-Means Clustering algorithm, which groups data into homogeneous clusters. This study aims to segment customers based on potential purchases using the K-Means Clustering method. The customer dataset in FMCG stores was divided into three clusters using seven attributes: Sex, Marital Status, Age, Education, Income, Occupation, and Settlement Size. The results, calculated in Microsoft Excel, concluded after four iterations with three clusters: k1 (Cluster 1) with 535 customers having low purchase potential, k2 (Cluster 2) with 685 customers having high purchase potential, and k3 (Cluster 3) with 7810 customers having medium purchase potential
Detection of Bias in Machine Learning Models for Predicting Deaths Caused by COVID-19
Full text linkThe COVID-19 pandemic has significantly impacted global health, resulting in numerous fatalities and presenting substantial challenges to national healthcare systems due to a sharp increase in cases. Key to managing this crisis is the rapid and accurate identification of COVID-19 infections, a task that can be enhanced with Machine Learning (ML) techniques. However, ML applications can also generate biased and potentially unfair outcomes for certain demographic groups. This paper introduces a ML model designed for detecting both COVID-19 cases and biases associated with specific patient attributes. The model employs Decision Tree and XGBoost algorithms for case detection, while bias analysis is performed using the DALEX library, which focuses on protected attributes such as age, gender, race, and ethnicity. DALEX works by creating an "explainer" object that represents the model, enabling exploration of the model\u27s functions without requiring in-depth knowledge of its workings. This approach helps pinpoint influential attributes and uncover potential biases within the model. Model performance is assessed through accuracy metrics, with the Decision Tree algorithm achieving the highest accuracy at 99% following Bayesian hyperparameter optimization. However, high accuracy does not ensure fairness, as biases related to protected attributes may still persist
Enhancing Image Quality With Deep Learning: Techniques And Applications
Full text linkThe emergence of deep learning has transformed numerous fields, particularly image processing, where it has substantially enhanced image quality. This paper provides a structured overview of the objectives, methods, results, and conclusions of deep learning techniques for image enhancement. It examines deep learning methodologies and their applications in improving image quality across diverse domains. The discussion includes state-of-the-art algorithms such as Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), and Autoencoders, highlighting their applications in medical imaging, photography, and remote sensing. These methods have demonstrated notable impacts, including noise reduction, resolution enhancement, and contrast improvement. Despite its significant promise, deep learning faces challenges such as computational complexity and the need for large annotated datasets. outlines future research directions to overcome these limitations and further advance deep learning\u27s potential in image enhancement
Prediksi Cepat Gangguan Jaringan Tegangan Menengah Menggunakan Metode Knowledge Growing System (KGS)
No full textWith the increasing demand for electrical energy in the household and industrial sectors, reliability in the distribution of electrical energy is very important. Disturbance in electricity distribution is a routine problem that will always occur in the field. To improve the quality of service, readiness in overcoming distribution disturbances is needed, for example by knowing the disturbances that will occur in the field. This study was conducted to solve this problem by applying the Knowledge Growing System (KGS) method in predicting the type of electricity distribution disturbance that occurred in the PLN unit. In this study, the scope of the research object is limited to PLN units in the South Surabaya area. This prediction is done by recognizing the pattern of disturbances that occur every month based on data taken in 2020. This method was chosen because it is an intelligent agent that can generate its knowledge through observing certain phenomena so that it can produce its own knowledge in making predictions. In this study, 5 patterns of electrical disturbances were used at the location of the electricity distribution. From the results of calculations and analysis using the KGS method, it was found that the prediction of electrical distribution disturbances in the form of animal disturbances with the highest degree of confidence value (DoC) occurred at the Sukolilo substation of 34.77%. Predictions of other disturbances in the form of "material" disturbances occur in Rungkut, Waru, and Darmo Grand feeders with DoC values ​​of 28.33%, 29.72%, and 34.72%, respectively.Semakin meningkatnya kebutuhan energi listrik di sektor rumah tangga hingga industri menyebabkan energi listrik menjadi salah satu kebutuhan yang sangat penting dalam kehidupan sehari hari, sehingga keandalan dalam pendistribusian energi listrik harus sangat diperhatikan. Adanya gangguan-gangguan yang terjadi karena beberapa factor gangguan dapat menyebabkan terganggunya keandalan pada supply listrik. Situasi seperti ini sering terjadi di unit Perusahan Listrik Negara (PLN) di wilayah Surabaya Selatan, hal inilah yang melatar belakangi penelitian ini dibuat salah satunya dengan memprediksi gangguan distribusi listrik yang bertujuan untuk menguji metode Knowledge Growing System (KGS) dalam memprediksi masalah gangguan listrik dengan cara mengenali pola gangguan yang terjadi di setiap bulannya yang kemudian di akumulasi menjadi pola gangguan pertahun. KGS adalah agen cerdas yang dapat menghasilkan pengetahuannya sendiri tentang fenomena yang diamati dan menggunakan pengetahuan yang dihasilkan untuk membuat prediksi. Dengan memiliki pengetahuan tentang 10 pola gangguan listrik di lokasi distribusi listrik, KGS telah mampu memprediksi bahwa gangguan yang paling mungkin terjadi karena gangguan belum ditemukan/gangguan sesaat di Gardu induk Rungkut, Waru, Wonorejo, Sukolilo dan Ngagel dengan rata rata terjadi gangguan sebesar 26,91 %. Dengan prediksi yang cepat, unit PLN dapat mengembangkan rencana yang tepat untuk mengatasi gangguan dan memulihkan pasokan listrik dengan cepat
Avseed Battery: Environmentally Friendly Battery Innovation as Electrolytes in Dry Batteries
Full text linkEnergy is a fundamental force driving transformative processes across various domains. It is perpetual and adheres to the law of energy conservation, remaining indestructible. However, the challenge lies in the finite nature of conventional energy sources, which convert energy for intended purposes and cater to diverse needs, often escalating the overall cost. Among the commonly employed energy sources, batteries play a pivotal role. This study explores the viability of avocado seeds as potential electrolytes in dry batteries. The objectives are to assess the effectiveness of avocado seeds as electrolytes and to investigate the impact of solution concentration and composition on the generated electrical energy. A dry element battery, known for converting chemical energy into electrical energy through Redox (Reduction-Oxidation) electrochemical reactions, serves as the experimental focus. Using a quantitative approach with laboratory experiments, five treatments were administered, featuring different ratios (1:1, 1:2, 1:3), a negative control with avocado seeds, and a positive control with salt. The bio-battery effectiveness assessment revealed that the P4 composition (negative control with avocado seeds) exhibited the highest initial voltage of 3.4 V and an extended runtime of 156 hours. In summary, this research underscores the potential of avocado seeds as electrolytes in dry batteries, supported by observations of voltage levels and ignition times
Analysis of Hybrid Learning Sentiment among Information Systems Students using The Naïve Bayes Classifier
Full text linkHybrid learning, which combines online and face-to-face instruction, has gained significant attention. Particularly in the Faculty of Computer Science, student engagement in hybrid learning is a central concern that arises during implementation. Hybrid, or blended learning, integrates various teaching methods, such as face-to-face, computer-based, and mobile learning, and offers advantages by reducing the time required for meetings and information delivery. Sentiment analysis, a branch of text mining, aims to determine public opinion or sentiment on topics, events, or issues. This study surveyed 112 Information Systems students using an online questionnaire to assess their responses to hybrid learning, classified as positive, negative, or neutral using the Naïve Bayes classifier. The research stages included data collection, preprocessing, Naïve Bayes model training, model evaluation, and sentiment analysis. The study aimed to analyze hybrid learning’s impact on students\u27 learning experiences and assess the accuracy of the Naïve Bayes method in classifying sentiments regarding this impact. The results indicated that the initial test had an accuracy of 60.87% without using the SMOTE up-sampling operator, while the second test achieved 80.65% accuracy with the operator
Systematic Literature Review: Identifying Key Variables and Measuring Maximum Loan Limits
Full text linkThis systematic literature review aims to identify key variables and measurement methods for determining maximum credit loan limits, following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The complexity of setting an optimal credit limit to manage credit risk effectively presents a significant challenge. Establishing an efficient maximum loan limit is essential to mitigate credit risk, as an overly high limit increases default potential, while an excessively low limit restricts the financial institution\u27s growth. This study identifies key variables and measurement methods, including Machine Learning techniques, Neural Networks, and traditional statistical approaches. Machine Learning models, such as Random Forest and Gradient Boosting, often surpass traditional methods in handling large, unstructured datasets due to their capacity for modeling complex, non-linear relationships. Conversely, traditional methods like logistic regression may be more suitable for smaller datasets, offering better interpretability and ease of use. The results indicate that systematic variable identification and the use of appropriate measurement methods can enable financial institutions to manage credit loan risk more effectively, supporting the development of sound credit policies
A Systematic Literature Review: Performance Comparison of Edge Detection Operators in Medical Images
Full text linkMedical images play a crucial role in the diagnosis of diseases. To make the diagnosis more accurate, the image should usually be enhanced first using image processing methods such as segmentation and edge detection stages. However, the complexity and noise that may arise in these images pose challenges in edge detection. Therefore, to portray the characteristics of edge detection operators, this research presents a systematic literature review of the performance of various edge detection operators in medical images, focusing on literature published between 2019 and 2023. After the selection process, 41 papers out of the initial 112 collected papers were chosen for further review. The study evaluates edge detection operators e.g., Canny, Sobel, Prewitt, Roberts, and Laplacian of Gaussian (LOG) on medical images such as X-rays, MRI, CT scans, ultrasound, Pap smears, and others. In the analysis, the accuracy, computational time, and response to noise of each operator are compared. The results indicate that despite longer computational times, Canny emerges as the most accurate operator, especially in Pap smear and CT scan images. The LOG operator offers high accuracy in MRI images with more efficient computational time. Evaluation of operator reliability against noise confirms the superiority of Canny. Furthermore, the future potential of edge detection in medical services is also reviewed. For instance, Canny, known for accurate and noise-resistant edges, enhances detection in complex CT-Scan and X-ray images. Meanwhile, LOG, handling artifacts with lower computational time, improves edge clarity in medical images. Potential applications include enhanced diagnosis, efficient patient monitoring, and improved image clarity in future medical services
Diabetic Retinopathy Severity Level Detection Using Convolution Neural Network
Full text linkDiabetic retinopathy is a common complication of diabetes mellitus, leading to damage and blockage of retinal blood vessels. Early and accurate detection of diabetic retinopathy severity levels is crucial for timely treatment and prevention of blindness. Diagnostic methods rely on manual examination and human interpretation, resulting in slower and less efficient treatment processes. As a branch of artificial intelligence, computer vision offers a potential solution to analyze retinal images quickly and accurately. The developed system employs image processing techniques and a CNN-based classification model to detect and classify the severity levels of diabetic retinopathy. By providing an automated and efficient approach, the system aims to assist doctors and optometrists in making informed decisions and reducing subjectivity in diagnosis. Early detection through this system can facilitate prompt treatment and improve patient outcomes. The developed system achieves promising results through experimentation and testing with various datasets, with accuracy ranging from 80% to 97%. This project\u27s integration of artificial intelligence, machine learning, and image processing technologies demonstrates their potential in healthcare applications, particularly in diabetic retinopathy diagnosis
Optimal Power Flow using An Optimally Tuned Pattern Search Algorithm
Full text linkOptimal power flow (OPF) is a critical optimization application in power system planning and operation. Numerous studies employ metaheuristic techniques to address OPF problems of varying complexity. However, these techniques often suffer from slow convergence due to their dependence on the quality of initial solutions. To overcome this limitation, initial solutions must be optimally tuned to achieve good outcomes with faster convergence. This paper proposes an optimally tuned pattern search (OPS) algorithm to solve OPF problems in medium and large power systems. The tuning process, performed using the classical interior point method (IPM), provides optimal initial control variable values for the standard pattern search (PS) algorithm. The proposed technique is applied to three test systems: IEEE 30-bus, IEEE 57-bus, and IEEE 118-bus systems. The OPF problem is formulated to minimize four objectives: total active power loss, total generator fuel cost, total generator emission, and total deviation in load bus voltage magnitude. The performance of the OPS algorithm is evaluated based on objective function values and computation times and is compared with IPM and two popular metaheuristic techniques, particle swarm optimization (PSO) and genetic algorithm (GA). Results indicate that the OPS algorithm\u27s performance varies across test systems but generally balances optimization performance with computational efficiency