Komputasi: Jurnal Ilmiah Ilmu Komputer dan Matematika

IOT-based Soil Moisture Monitoring and Water Pump Control System for Ornamental Plants

Attalarik Zidan, Yunita Rahma , Sufiatul Maryana — Thu, 31 Jul 2025 00:00:00 +0000

Ornamental plants cultivation is a type of plant cultivation that is growing rapidly and also has great potential in Indonesia. In Bogor Regency itself, many farmers started developing this ornamental plant after the Pandemic Era. The unique tropical pattern has become a surge in market demand, one of which is the Syngonium type. The challenges for farmers in cultivating ornamental plants come from the environment itself, the country's tropical climate causes Indonesia to have high rainfall and unpredictable heat. There are several types of plants whose growth process is very sensitive to the level of soil acidity. One solution to dealing with this problem factor is to apply intensification. This research aims to create a tool that is a monitoring and control system for water pumps to provide direct information about soil moisture as well as regular irrigation. The system was created using a capacitive soil sensor as input data for 3 soil conditions, namely Dry, Moist and Wet, then sent by the NodeMCU into the database to be displayed via the website and application, all of which are connected to the Arduino cloud, and received by a water pump for periodic irrigation. The result of this research is a prototype system for monitoring soil moisture and controlling water pumps for ornamental plants based on IoT which is expected to help farmers produce good planting media as well as good irrigation in the plant care phase.

Numerical Verification of the Order of Accuracy of a Runge-Kutta-Fehlberg Method in Solving an SEIR Model

Lintang Kinanthi Widi, Sudi Mungkasi — Thu, 31 Jul 2025 00:00:00 +0000

A Runge-Kutta-Fehlberg method (RKF45) is considered to solve an SEIR (Susceptible-Exposed-Infected-Recovered) mathematical model for infectious disease transmission. This numerical method is of the Runge-Kutta class, and has theoretical order of accuracy four and five depending on the involved formulations. We limit our problem to the RKF45 formula having theoretical order five. It is important to understand the behavior of RKF45 in solving an initial value problem, especially when implemented for real problems. In this paper, we take an SEIR mathematical model in our case, because this model is applicable in the prediction of infectious disease transmission in real life. Our research method is quantitative. We record the errors of numerical simulations. From the error values, we determine the numerical order of accuracy. Our research results show that RKF45 produces numerical order of accuracy five for the time step is sufficiently small. Therefore, the numerical order of accuracy matches with the theoretical one.

Shallot Growth Stage Monitoring with Multispectral Imagery using MobileNetV2

Yanuar Firmansyah, Nugraha Priya Utama — Thu, 31 Jul 2025 00:00:00 +0000

Shallots are a strategic horticultural commodity, playing a pivotal role in the Indonesian food supply as a staple food. It is anticipated that the demand for shallots will increase in the future. Nonetheless, the production level remains inadequate, necessitating the importation of shallots. The success of the production process is contingent upon the monitoring of the crops by the farmers themselves. The current manual approach to this process has proven to be inadequate, necessitating the development of agricultural technology. Remote sensing has the potential to become a valuable technology in agriculture and represents a feasible methodology for accurate land mapping. However, its application in Indonesia remains constrained by the heterogeneity of the mapped areas and the quality of the images obtained. Furthermore, Indonesia's agricultural land is extensive and heterogeneous. This research presents the use of multispectral cameras for the precise monitoring and mapping of agricultural land. The imaging system employs a multispectral camera attached to a UAV to perform land mapping. RGB imagery and vegetation indices, including NDVI, NDRE, and OSAVI, are calculated to ascertain the crop status within a single shallot planting period. The shallot growth stage prediction is proposed using MobileNetV2 architecture. The proposed method demonstrated an accuracy of 84.78% using RGB and 44.78% using a combination of vegetation indices.

Cybercrime Awareness Phishing Email Using Secure Socket Layer as an Interpretation of Information Security

Thu, 31 Jul 2025 00:00:00 +0000

Abstract. Email phishing is a fraudulent technique used by cybercriminals to
obtain sensitive information through emails that appear legitimate but have
malicious purposes. Information security has become a crucial aspect in the
increasingly developing digital world, especially along with the increasing threat
of cybercrime, such as email phishing. In this context, implementing Secure Socket
Layer (SSL) is one solution to increase awareness and mitigate risks against
phishing attacks. SSL, which is a security protocol for data encryption, can help
protect digital communications by authenticating servers and encrypting data
sent, thereby reducing the potential for misuse of personal data sent via email.
This research aims to examine the role of SSL in providing protection against email
phishing and increasing user awareness regarding the importance of information
security in the use of digital technology. Through an information security
interpretation approach, this study suggests that SSL can serve as an effective
barrier in reducing the risk of phishing attacks and increasing the level of user
trust in the emails received. Therefore, it is important for individuals or
organizations to understand and apply information security principles such as the
use of SSL to maintain the integrity and confidentiality of data exchanged online

The Similarity of COVID-19-Related Profiles and Pandemic Conditions in ASEAN Countries

Elis Khatizah — Thu, 31 Jul 2025 00:00:00 +0000

Abstract

Abstract This paper examines the potential relationship between countries’ profiles and COVID-19
conditions in 10 ASEAN countries, focusing on new confirmed cases and deaths. We initially clustered
the countries based on demographic profiles and then observed the pandemic situation within each cluster.
The samples of genome sequence, particularly the Omicron variant collected from each country, are also
clustered to investigate the spread of the virus. Furthermore, we created a simple algorithm to identify
countries with profiles similar to ASEAN nations. Our findings show that countries with similar
demographic profiles and geographically proximate are more likely to experience similar pandemic
conditions even though they have no identical genome sequence of the virus. Five of the 10 ASEAN
countries have similar demographic profiles to countries within the ASEAN region and as assumed,
experienced similar COVID-19 situations. Conversely, the remaining ASEAN countries, which share
profiles with countries outside the region, demonstrate fairly different COVID-19 conditions, particularly
with regard to the timing of the spread of the virus. With comparable resources at hand, insights from
countries with similar profiles provide valuable information for effective pandemic management efforts
in the future.
Keywords: ASEAN profiles; clustering; COVID-19; demographic; pandemic management

Implementation of Fuzzy Search and Audio Player for Web-Based Quranic Verse Retrieval

Thu, 31 Jul 2025 00:00:00 +0000

This study aims to develop a web-based Quranic search system integrating fuzzy search and audio player features to address typographical errors in user inputs and enhance accessibility. Employing the Prototype Development Model, we implemented the Levenshtein Distance algorithm via Fuse.js for tolerance-based text matching and integrated an interactive audio player for verse recitation. Testing with typo variations revealed optimal performance at an 80–85% similarity threshold, achieving 92% user satisfaction in handling common misspellings (e.g., "Albaqara" → "Al-Baqarah"). The system’s client-side processing ensured rapid responses (<1s), while Progressive Web App (PWA) architecture enabled offline functionality. Key contributions include: (1) a typo-resistant Quranic search interface, (2) empirical validation of threshold settings, and (3) holistic user engagement through audio-text integration. This work bridges gaps in prior studies that focused solely on exact matching or audio features, offering a novel synergy of adaptive search and multimedia interaction for Islamic digital tools.

IMPROVEMENT MARKET BASKET ON TRANSACTION SALES USING APRIORI ALGORITHM ( Case study : Workshop The Sari (Mulya )

Nurul Aziz, Sri Setyaningsih, Dini Suhartini — Thu, 31 Jul 2025 00:00:00 +0000

This research refers to the improvement of Market Basket with Apriori Algorithm to analyze Bengkel Las Sari Mulya data. At the implementation stage, Apriori Algorithm is used to find unique values from the data carefully and accompanied by UAT (User Acceptance Test ) testing for tests that can be used as evidence that the software has been accepted with an average value of 4.7 that this application is very satisfying for users and has met the needs of those who request it. To find out the response of the Workshop owner to the application. The results of the analysis show an association between Goods such as Fences and Canopies, by determining a Support value of 15% and Confidence of 20%, resulting in an association rule value with a Support value of 24% and Confidence of 73%. In conclusion, customers who buy fence goods tend to also buy Canopies in one transaction. These results can be used for marketing strategies and stocking goods, for example, with frequently sold goods such as fences and canopies, you can stock goods such as: 2x4 Galvanized Iron, 4x4 Galvanized Iron, 4x8 Galvanized Iron, 5x10 Galvanized Iron, spandek, Alderon, Because from these results, fences and canopies can be found, goods that are often sold and it can be seen that both use the materials used in making goods, the owner will easily make decisions regarding the plan each month.

Say Key: Apriori Algorithm, Transaction Data, Welding Workshop.

Application of Proportional Integral Derivative system as speed control on Mobile Robot Line Follower

Raka Pramudiya Ramadani, Arie Qur'ania, Agus Ismangil — Thu, 31 Jul 2025 00:00:00 +0000

The development of the world of robotics in Indonesia is experiencing very rapid development compared to a few years ago, the use of PID Control-based technology is growing in the field of robotics. This line follower robot uses IR Line follower sensor to detect the black line and PID control to regulate the movement of the robot automatically, so that the robot can follow the line that has been made with direction. The Arduino Uno microcontroller in this robot design is the main part of the input and output control system. In this robot design, Arduino Uno is used as a microcontroller to control sensors that function as sensing / detection tools for robots. The working system of this robot is: how to run it by activating the power button on the back of the robot. The number of sensors and the accuracy of the PID control value can affect the comparison of the robot's travel time value from the starting point to the end of the test with two paths that have been tested, the results obtained 4 IR LF Sensors with an average travel time of 6.7 seconds on the first path and 13.5 seconds on the second path, better than 2 or 3 IR LF Sensors which have a difference of 0.64 seconds on the first path and 0.7 seconds on the second path. The above values can change according to the influence of component specifications, circumstances and robot environment during testing.

Recent Advances in Deep Learning for MRI-Based Brain Tumor Identification: A Systematic Review (2020 - 2025)

adly maulana, Suherman — Thu, 31 Jul 2025 00:00:00 +0000

The utilization of deep learning (DL) technology in brain MRI image analysis has seen significant advancements over the past five years. This study presents a systematic review of literature from 2020 to 2025, evaluating DL progress in automated tumor lesion segmentation, tumor type classification, genetic biomarker prediction, and treatment response monitoring. Various DL architectures, such as nnU-Net and ensemble models, dominate segmentation tasks, while transformer-based methods and foundation models are emerging as new pathways for large-scale medical image management. However, technical challenges including cross-institutional MRI protocol variations, underrepresentation of pediatric data, and model bias remain primary concerns. Initiatives like BraTS and federated learning approaches offer potential solutions to enhance DL model validity and scalability. This review highlights future directions for developing more adaptive, accurate, and ethical DL systems to support individualized and sustainable brain tumor diagnosis and management.

Application of Principal Component Analysis on Factors Causing Inflation in West Kalimantan

Asri Rahmawati, Yuyun Eka Pratiwi, Onelia Rochmah — Thu, 31 Jul 2025 00:00:00 +0000

Inflation is an important indicator in assessing the economic stability of a region. Inflation fluctuations in West Kalimantan are influenced by various economic and structural factors. This study aims to identify the main factors causing inflation in West Kalimantan using Principal Component Analysis (PCA). Secondary data for the 2024 timeframe was obtained from West Kalimantan's Central Statistics Agency (BPS). Economic variables that are suspected of influencing inflation are analyzed using PCA to be reduced to new dominant factors. The main components obtained are then interpreted economically to understand the structure of the causes of inflation. The results of the analysis show that the cumulative proportion of the two components reaches 90%, so the two main components are sufficient to represent the main structure of the data. This means that most of the information from the original variables can be effectively reduced into two main components.

Keywords: Inflation; PCA; reduction; economic factors.

Interpreting sign language images into Text-Based using YOLOv7

Thu, 31 Jul 2025 00:00:00 +0000

Sign Language recognition remains a significant challenge due to variations in hand gestures, occlusions, and environmental factors. This study introduces a YOLOv7-based sign language recognition system designed to interpret American Sign Language (ASL) images into text in real time, enhancing communication accessibility for individuals with hearing and speech impairments. The primary objectives are to develop an accurate detection model, improve real-time gesture interpretation, and optimize system performance for accessibility. The study follows the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology, including data collection, preprocessing, model training, evaluation, and deployment. The system was developed using YOLOv7, Python, and OpenCV, with an ASL dataset sourced from Roboflow and formatted in YOLOv7 PyTorch. Image preprocessing techniques such as normalization, resizing, and data augmentation were applied to enhance detection accuracy. The application integrates a real-time gesture recognition system, where detected ASL signs are instantly translated into text. Results demonstrate high detection accuracy for most ASL letters, but certain gestures such as J, Z, S, and T posed challenges due to similar hand shapes and motion-based characteristics. Optimization efforts included dataset expansion, refined annotations, and hyperparameter tuning to improve model precision. The system significantly enhances real-time ASL recognition, offering a scalable, AI-powered assistive tool for the deaf and hard of hearing community. This research contributes to machine learning-driven accessibility solutions, bridging the communication gap between sign language users and non-signers.

Smart Waste Management System Using Computer Vision

Thu, 31 Jul 2025 00:00:00 +0000

Waste management in a university setting is a complex task that involves handling diverse waste streams in classrooms, laboratories, and canteen areas, proper segregation of wastes, and disposal while ensuring compliance with environmental regulations and sustainability goals. In response to the inefficiencies in handling adequate waste segregation at Guimaras State University (GSU), this study presents the development of a Smart Waste Management System using Computer Vision. The system aims to develop a model to classify waste according to waste type, handle automatic segregation, and provide stakeholders with real-time updates. Adopting the Cross Industry Standard Process for Data Mining (CRISP-DM) is part of the approach, including business understanding, data understanding, data preparation, modeling, evaluation, and deployment phases. The complications of waste management processes have been identified through interviews with utility personnel at Guimaras State Unversity – Mosqueda Campus. A prototype bin was developed using hardware electronic technologies and third-party platforms such as Edge Impulse and Blynk to handle waste identification for proper disposal and segregation. The study’s primary results include a substantial decrease in the complexity of the waste management process, as well as promoting adequate waste segregation. By utilizing hardware technologies and training a machine learning model, the system was able to manage waste identification, automated waste disposal operations, and offer real-time updates to the stakeholders. Overall, the Smart Waste Management System using Computer Vision offers efficiency and sustainability in the waste management process at GSU.

Keywords: smart waste management; computer vision; arduino; esp32 camera; esp32