Sinkron : jurnal dan penelitian teknik informatika
Not a member yet
1261 research outputs found
Sort by
Optimizing Iron Price Forecasting with Linear Regression Analysis and RapidMiner
No full textCompetition in companies often occurs in price, advertising and promotion, and quality. Price is very influential on competition in a business. The price of a product is one of the things that influences buyers to want to buy a product or not; therefore, price is very important to determine. There are two objectives in this study; the first objective is to predict the right iron price to be used in the following year so that it can be used to increase the competitiveness of the company. The second objective is to determine the attributes that affect the price. This research uses a linear regression algorithm to predict prices and measure the attributes' relationship using the RapidMiner tool. RapidMiner is software that functions as a learning tool in data mining science in which various data processing models are ready to be used easily. From the test results on the training data, an accuracy value of 95% was obtained with a threshold value of 30, which stated that the results were accurate. Then, the factors that affect the price produce factors from the size variable (mm) and unit (kg); between the two variables that affect the price, there are results from the variables that most affect the price, namely size (mm). For the performance of the linear regression model calculated using the root mean square error (RMSE) produces a value of 199,291
Analysis of Student Excellence Classes in Data Mining Using the KNN Method
No full textExcellent classes are programs designed to maximize the academic and non-academic potential of students and girls, with the aim of improving their overall achievement. This program aims to provide more intensive learning and a curriculum tailored to students' needs and abilities, so that they can develop their talents and competencies optimally. In order to evaluate the effectiveness of the superior class program and to identify students who are most suitable for the program, this research was conducted using the K-Nearest Neighbors (KNN) method in data mining. The research process includes several critical stages, namely determining relevant data, designing a machine learning model, testing the model to ensure its effectiveness, and evaluating the model to assess the accuracy and reliability of the results. This research used sample data consisting of 92 male and female students, where the results of the analysis showed that 42 of them met the criteria to enter the superior class, while 50 other students did not. These criteria are determined based on various factors, including academic achievement, participation in extracurricular activities, and other individual characteristics assessed through the KNN method. The accuracy results obtained from the model evaluation show excellent performance, confirming that the approach used is effective in classifying students based on their potential to excel in superior class programs. The conclusion of this research shows that the use of the KNN method in data mining can accurately identify students who will benefit most from superior class programs. Thus, this approach offers a valuable tool for educational institutions to optimize student potential and raise overall standards of achievement
Deep Learning for Exchange Rate Prediction Within Time Constrain
No full textThe implementation of an open economic system in Indonesia since 1969 has significant impact to the national economic growth. The high demand and supply of goods from within the country involved in international trade demonstrate a close correlation between export and import activities with the exchange rate of the rupiah. Economic stability is measured through the stability of the rupiah exchange rate against foreign currencies. The balance between demand and supply in the global market is considered crucial for creating a stable economy. History has recorded the Indonesian economic crisis in 1998, where the exchange rate of the rupiah against the US dollar drastically raises and causing challenges to the domestic production cost. This research aiming to make predictions using data science approach based on historical (time series) data. GRU, LSTM, and RNN algorithm being assess to perform the prediction. Results show that RNN algorithms generally outperform GRU and LSTM in making the prediction, particularly with limited time series data. Although RNN is typically superior, in one instance, GRU achieved slightly higher accuracy (0.047% difference) for the CNY to IDR pair over five years. Furthermore, the research highlights the substantial impact of batch size on algorithm accuracy, considering external factors such as interest rates. These findings offer valuable insights for economic decision-making and policy formulation
Implementation of Exploratory Data Analysis and Artificial Neural Networks to Predict Student Graduation on-Time
No full textAlmost all universities in Indonesia face the problem of a low number of students graduating on time. This will affect higher education accreditation. For this reason, universities must pay attention to the timely graduation of their students. The way that can be taken is to predict students' graduation on time. This research aims to predict students' timely graduations using a combination of exploratory data analysis and artificial neural networks. Exploratory data analysis is used to study the relationship between features that influence students' on-time graduation, while artificial neural networks are used to predict on-time graduation. This research goes through method stages, starting with determining the dataset, exploratory data analysis, data preprocessing, dividing training and test data, and applying artificial neural networks. From the research, it was found that Work features and GPS features greatly influence graduation on time. Students who study while working are less likely to graduate on time compared to students who do not work. Students who have an average GPS above 3.00 for eight consecutive semesters will find it easier to graduate on time. Meanwhile, Age and Gender features have no effect on graduating on time. With a percentage of 50% training data and 50% test data, epoch 100, and learning rate 0.001, the best network model was obtained to predict graduation on time with an accuracy rate of 69.84%. The research results also show that the amount of test data and the learning rate can influence the level of accuracy. Meanwhile, the number of epochs does not affect the level of accuracy
Designing Claim Systems in Health Insurance Companies with Microservices and Event-Driven Architecture Approach
No full textThrough digital transformation, insurance companies, especially in the health sector, are increasingly adopting modern technologies to enhance efficiency and service quality. Health insurance allows individuals or families to mitigate the financial risks associated with high and unexpected medical expenses. One crucial area is insurance claim, where a fast and accurate process is key to customer satisfaction. This study proposes the design and architecture of an insurance claim system using a microservices and event-driven approach. This approach enables insurance companies to break down applications into separate components, facilitating scalability, flexibility, and easier maintenance. Additionally, with an event-driven approach, the system can quickly respond to changes and events in the business environment. A comprehensive analysis shows that implementing microservices and event-driven architecture in the insurance claim system can enhance overall system performance, scalability, and resilience. For insurance companies, adopting microservices and event-driven architecture can lead to increased operational efficiency, reduced time to market for new products, and improved customer experiences through faster claim processing. Policyholders will benefit from quicker claim resolutions and a more transparent and responsive claim process. This study provides valuable insights for health insurance companies looking to upgrade their IT infrastructure to meet future challenges. The findings from this research will be documented to support the development of insurance business technology, specifically for health insurance claims in Indonesia
Implementing Moving Average Forecasting System for Apparel Sales: Predicting Inventory Needs with Enhanced Accuracy
No full textForecasting the supply of goods is one of the company's planning strategies to increase sales. However, there are several obstacles in forecasting the supply of goods in one of the boutiques in Jember Regency such as manual sales data collection, namely by recording clothing sales data in the sales book. So that there can be errors in predicting the supply of goods in the future. The purpose of this study is to apply a clothing sales forecasting system using the moving average method to forecast the supply of goods. This study applies the waterfall model to build a system with stages of analysis, design, implementation and testing. Analysis will be carried out by collecting data related to system requirements through observation, interviews and literature studies. While at the design stage there are usecase diagrams and system flow diagrams. Furthermore, the implementation stage was carried out in boutiques in Jember Regency by piloting the boutique owners. System testing uses black box testing to ensure there are no system functional errors. The findings show that the system in the form of a website can be run properly and can be accessed as long as there is an internet network. In addition, our system is already running well based on the results of black box testing. So that this system can be used by companies as forecasting considerations in providing inventory of goods
Hybrid Inverse Weed Optimization Algorithm with Math-Flame Optimization Algorithm
No full textIn this work, two Meta-Heuristic Algorithms were hybridized, the first is the Invers Weed optimization algorithm (IWO), which is a passing multiple algorithm, and the second is the Moth-flame Optimization Algorithm (MFO). Which depend in their behavior on the intelligence of the swarm and the intelligence of society, and they have unique characteristics that exceed the characteristics of the intelligence of other swarms because they are efficient in achieving the right balance between exploration and exploitation. So the new algorithm improves the initial population that is randomly generated, A process of hybridization was made between the IWO and MFO Algorithm to call The new hybrid algorithm (IWOMFO). The new hybrid algorithm was used for 16 high-scaling optimization functions with different community sizes and 250 repetitions. The Algorithm showed access to optimal solutions by achieving the value Minority () for most of these functions and the results of this algorithm are compared with the basic algorithms IWO, MF
Comparison of K-Means and K-Medoids Clustering Algorithms for Export and Import Grouping of Goods in Indonesia
No full textInternational relations affect the economic growth of each country, which can affect the economic growth of each country. As a result, global economic growth is necessary, which means that the global economy has a greater capacity to produce goods and services. Exports and imports are very important to drive economic growth. but if exports and imports are not balanced, it will have a bad impact if the value of imports is greater than exports, export prices abroad will definitely fall. An analysis comparing export and import categories is needed to determine which goods are most imported and exported in Indonesia in 2021-2023. This study uses a quantitative methodology and machine learning methods, namely k-means and k-medoids algorithms. These two methods will be compared to determine which is the most effective for export and import data of goods in Indonesia in 2021-2023. The results of the study were obtained by K-Means more effectively in handling data on the grouping of exports and imports of goods in Indonesia in 2021-2023. The dataset shows the results of the evaluation of K-Means using DBI of 0.59, while the results of the evaluation using K-Medoids show a result of 1.7868. Because the evaluation value of K-Means has low computing performance compared to K-Medoids. The largest amount of the value and weight of exports and imports of goods in Indonesia is in C1 where in the HS code [27], namely Mineral fuels with a total export value of goods in 2021 to 2023 of 134,999,470,522 US. Meanwhile, the total export weight of goods from 2021 to 2023 in mineral fuel goods is 1,505,006,250,327 Kg or around 1,658,985,413 tons and the total import weight is 186,446,782,134 Kg or around 205,522,397 tons
Using Real-Time Ray Tracing in Game Action-Adventure ANGKARA "The Rise of Asura"
No full textThe pressing problems facing the games industry in Indonesia are first, the lack of developers who can meet the needs of the local market and the lack of adaptation of local content in games. Second, how to improve the optimization of the Ray Tracing process? To overcome these challenges, this research aims to design and build an action-adventure game model using Unreal Engine 5. By developing the Real-Time Ray Tracing (RTRT) feature. The game design will also utilize the latest technology such as Ray Tracing, and artificial intelligence to create an immersive and realistic gaming experience. The method used in this research is GDLC (Game Development Life Cycle). The results of research using black box testing with components, character responsiveness, basic attacks, interaction with the environment, defense, enemy behavior, and optimization of Ray Tracing state that it has been validated
The Classification of Avocado Ripeness Levels Using CNN Method
No full textThis research aims to develop a model for classifying the ripeness level of avocados using the Convolutional Neural Network (CNN) algorithm. The dataset comprises images of avocados categorized into three classes: unripe, ripe, and overripe. The CNN model is trained to classify the images into one of these three categories. The results indicate that the developed model can classify avocado images with high accuracy. The primary tool used for developing and implementing this method is MATLAB R2022a. The CNN algorithm is utilized to recognize and classify the ripeness level of avocados. This process involves several image processing steps, starting with preprocessing, image enhancement, and segmentation to isolate the avocado area. The dataset used in this research consists of 452 images distributed in 3 classes (unripe with 142, ripe with 66, and rotten with 244), with 80% used for training and 20% for testing. After 10 accuracy tests, the results indicate an accuracy rate of 90%. Additionally, features extracted from the images include color, shape, size, and texture characteristics, such as Mean, Standard Deviation, Kurtosis, Skewness, Variance, Entropy Value, Maximum Pixel, and Minimum Pixel. This research contributes to the field of agricultural technology by providing a robust method for the automatic classification of avocado ripeness. The findings are expected to facilitate accurate and efficient recognition of avocado ripeness, thereby supporting agricultural practices and market operations. Future research could explore the use of data augmentation techniques to further improve the accuracy and generalization of this model