Jurnal Informatika Progres

KLASIFIKASI TINGKAT KEMATANGAN LADA MENGGUNAKAN ENSEMBLE LEARNING BERDASARKAN CITRA WARNA KULIT

2025-09-01T12:58:08+00:00

Pepper fruit (Piper nigrum L.) is an agricultural commodity whose market value strongly depends on its ripeness level at harvest. Ripeness determination, which is still commonly performed through visual observation, tends to be inaccurate and subjective. This study aims to classify the ripeness level of pepper fruit based on skin color using an ensemble learning approach. The dataset consists of 1,996 pepper fruit images categorized into four ripeness levels unripe, semi ripe, ripe, and overripe. Color features were extracted from the HSV color model using color moment statistics including mean, standard deviation, and skewness. Random Forest and XGBoost models were combined using a soft voting method. The results show that the ensemble model achieved 98.25% accuracy, 98.30% precision, 98.27% recall, and 98.26% F1-score. The ensemble approach proved superior to single models by providing more accurate and stable classification of pepper fruit ripeness.

KLASIFIKASI PENYAKIT TANAMAN NILAM BERDASARKAN CITRA DAUN MENGGUNAKAN GLCM DAN SVM

2025-09-01T14:09:21+00:00

This study presents a classification model for detecting diseases in patchouli (Pogostemon cablin Benth) leaves using image processing techniques. The method combines Grey Level Co-occurrence Matrix (GLCM) for texture feature extraction and Support Vector Machine (SVM) for classification, optimised using the Particle Swarm Optimisation (PSO) algorithm. A total of 2,080 leaf images were collected and categorized into four classes: healthy, leaf spot, yellowing, and mosaic. Each image was augmented and converted to grayscale to enhance the dataset and reduce computational complexity. Four GLCM features—contrast, correlation, energy, and homogeneity—were extracted to represent leaf textures. The classification model achieved an accuracy of 89.74% using SVM alone, and improved to 97.12% when optimized with PSO. The results indicate that the integration of GLCM, SVM, and PSO provides an effective and accurate solution for early detection of patchouli leaf diseases, potentially supporting farmers in decision-making and improving crop productivity and quality.

IMPLEMENTASI DEEP LEARNING MENGGUNAKAN HYBRID SENTENCE-TRANSFORMERS DAN K-MEANS UNTUK PERBANDINGAN JURNAL

2025-09-01T14:45:37+00:00

This study addresses the challenge of identifying semantic relatedness between scientific journal articles by developing a classification system based on deep learning. The system applies an unsupervised learning approach using the Sentence-Transformers model and K-Means clustering to generate semantic similarity scores and categorical labels. Abstracts from journal PDFs are extracted and processed to determine similarity levels across four predefined categories. The optimal number of clusters was determined using Elbow Method, Silhouette Score, and Davies-Bouldin Index, resulting in k = 4. The system is implemented as a web-based application that allows users to upload two PDF files, compare them semantically, and receive both a similarity score and an AI-generated narrative explanation. Functional testing showed that all core features performed as expected. This system significantly reduces the time required to assess relatedness between journal articles, offering an efficient tool for academic research navigation.

IMPLEMENTASI K-MEANS DAN ANALISIS SENTIMEN KRITIK SARAN BERBASIS NLP PADA DATA MONEV BBPSDMP KOMINFO MAKASSAR

2025-09-03T04:37:29+00:00

Manual analysis of large-scale and unstructured textual feedback data is often inefficient and subjective, thereby hindering data-driven decision-making. This study aims to design and implement an integrated analytical workflow to automatically filter, cluster, and classify feedback data consisting of criticisms and suggestions. The research employs a hybrid approach that begins with TF-IDF-based data filtering, followed by dimensionality reduction using Latent Semantic Analysis (LSA), and topic clustering through K-Means clustering optimized with the Silhouette Score. The resulting cluster labels are then used as training data to build a Multinomial Naive Bayes classification model. The results show that this workflow successfully identified two main thematic clusters, namely "Criticism and Expectations" and "Suggestions and Compliments", and the classification model achieved an overall accuracy of 91%. Although class imbalance affected the recall of the minority class (47%), the model demonstrated high precision (95%) for that class. It is concluded that this hybrid approach effectively transforms raw data into structured insights, and utilizing clustering results as training data is an efficient strategy for automating feedback categorization, providing a reliable tool for institutional analysis.

PREDIKSI PEMAKAIAN AIR BULANAN DI PDAM KECAMATAN TAMALATE MENGGUNAKAN METODE AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA)

2025-09-03T07:51:39+00:00

Water consumption forecasting is a crucial aspect of efficient water resource management, particularly in urban areas with increasing demand. This study aims to predict the monthly water usage volume at the PDAM of Tamalate District using the Autoregressive Integrated Moving Average (ARIMA) method. The dataset consists of historical water usage data from January 2022 to December 2024, totaling 36 monthly observations. The analysis process includes stationarity testing using the Augmented Dickey-Fuller (ADF) test, model parameter identification through ACF and PACF plots, and performance evaluation using MAE, RMSE, and MAPE metrics. The results show that the best-performing model is ARIMA, which demonstrates high prediction accuracy, with a MAE of 26,049.80 m³, RMSE of 37,459.00 m³, and MAPE of 4.12%. This model is capable of generating predictions close to actual values and can be relied upon as a basis for PDAM’s water distribution planning. It is expected that this research will contribute to data-driven decision-making and support digital transformation in the public service sector.

IMPLEMENTASI HYBRID LEXICON-BASED DAN SVM UNTUK KLASIFIKASI ANALISIS SENTIMEN TERHADAP PELATIHAN BBPSDMP KOMINFO MAKASSAR

2025-09-03T09:04:43+00:00

The evaluation of government training programs is often hindered by manual analysis of unstructured qualitative feedback, making the process inefficient and subjective. This study aims to implement and evaluate a sentiment classification model using a hybrid Lexicon-Based and Support Vector Machine approach to analyze participants’ perceptions of the Vocational School Graduate Academy training organized by BBPSDMP Kominfo Makassar, as well as to compare the performance of a standard SVM model with a model optimized using Particle Swarm Optimization. This quantitative research employs 2,313 unstructured review data, which undergo text preprocessing, initial lexicon-based labeling, and TF-IDF feature extraction before being classified using an SVM with an RBF kernel. The results show that the SVM model optimized with PSO consistently outperforms the standard model across all four evaluation aspects, with the most significant accuracy improvement observed in the instructor category from 84.71% to 89.02% and in the assessor category reaching 91.46%. PSO optimization has proven effective in enhancing the model’s ability to identify negative sentiments, which represent the minority class. The hybrid approach with PSO optimization is capable of producing a more accurate and balanced classification system, with practical implications as an objective automated evaluation tool.

PENERAPAN ALGORITMA K-NEAREST NEIGHBOR DALAM ANALISIS PEMINJAMAN BARANG PADA DIVISI INVENTARIS TVRI MAKASSAR

2025-09-04T06:46:15+00:00

Inventory management in the TVRI Makassar Inventory Division is inefficient due to the lack of a predictive system, hampering proactive asset requirement planning. This study aims to apply the K-Nearest Neighbor (KNN) algorithm to analyze historical borrowing patterns, predict demand for goods three months in advance, and evaluate model accuracy. Using a quantitative approach, this study implements a systematic machine learning workflow, including data preprocessing, temporal feature engineering, class imbalance handling using the Synthetic Minority Over-sampling Technique (SMOTE), and hyperparameter optimization using GridSearchCV. The results show that the optimized KNN model achieved an overall accuracy of 80.18%, significantly outperforming the baseline model. Key findings revealed that the model's performance is contextual, with very high reliability (F1-Score > 0.95) on frequently borrowed assets, and is able to identify strong temporal demand patterns. It is concluded that KNN is effective for segmented inventory demand prediction and has the potential to serve as a basis for TVRI Makassar to adopt a proactive, data-driven inventory management strategy, enabling more efficient resource allocation.

IMPLEMENTASI MODEL BUSINESS TO BUSINESS PADA PEMASARAN PRODUK JAMU MENGGUNAKAN FRAMEWORK LARAVEL

2025-09-11T08:25:32+00:00

The advancement of digital technology presents great opportunities for business actors to enhance their competitiveness through more efficient marketing strategies. This study aims to implement the Business to Business (B2B) model in the marketing of traditional herbal products by Jamu Dapoer Niswah using the Laravel Framework and the Agile method. The system developed is a web-based platform with core features such as business partner registration, product catalog, ordering, order tracking, and contract management. The Agile method was chosen for its flexibility in responding to changing user needs through iterative development. This study adopts an iterative approach consisting of planning, implementation, testing, documentation, deployment, and maintenance stages. The implementation results show that the system is capable of accelerating the ordering process, improving transaction accuracy, and expanding the partner market reach. System testing was conducted using the black-box method and showed that all features function as intended. With this system, the marketing and distribution processes of herbal products can be carried out in a more structured and efficient manner, accessible to business partners at any time. This system is expected to become a digital solution that supports B2B marketing transformation in the traditional business sector.

IMPLEMENTASI METODE PROTOTYPE PADA SISTEM PAYMENT GATEWAY UNTUK SEPATU KOTAMA

2025-09-11T08:23:06+00:00

Digital era nowadays, online payment systems have become an essential component in supporting transaction efficiency, especially for enterpreneur such as Toko Sepatu Kotama. The store's current manual payment process causes several issues, including delayed confirmations, risk of data entry errors, and a lack of convenience for customers. Therefore, this study aims to develop a web-based payment gateway system that is secure, efficient, and easy to use. The development method used is the prototype model, which enables iterative processes involving users to ensure that the system meets actual needs. The system is built using the Laravel framework and MySQL, and integrated with Midtrans as the payment service provider. The interface is designed to be responsive for both desktop and mobile devices.Testing using Black Box Testing shows that all main features function properly, including registration, login, order placement, checkout, and payment processing. Admins can also effectively manage products, orders, and sales reports. The system's advantages include real-time payment notifications and direct integration with Midtrans. The system has proven to improve operational efficiency and user convenience. Future development may include automatic shipping tracking, multilingual support, and more diverse payment options such as credit cards and installment plans. It is also recommended to conduct broader user testing to gain more representative feedback from the market.

IMPLEMENTASI WEBGIS UNTUK NAVIGASI RUTE DAN MONITORING HYDRANT PEMADAM KEBAKARAN DI KOTA MEDAN

2025-09-21T09:22:33+00:00

Firefighting operations in Medan City still face significant challenges in locating and monitoring the condition of fire Hydrants: out of 118 Hydrant points, only 23 are documented as operational and just 6 are fully functional. This study aims to design and implement a web‑based WebGIS for the fastest route navigation to the nearest Hydrant and real‑time Hydrant condition Monitoring, employing an iterative Prototype methodology. The system is built using PHP & MySQL on the backend, Leaflet.js for interactive mapping and Routing, and AJAX for seamless data Updates. Functional testing (black‑box) confirms all modules Admin Login, Hydrant CRUD, “Nearest Hydrant” Routing, status Updates, and map Refresh operate as expected with average AJAX response times of 0.3 - 0.5 seconds and map load times under 1.5 seconds for 200 points. Field accuracy tests reveal a coordinate deviation of ± 3 meters, while Hydrant search time is reduced from an average of 2 minutes manually to approximately 15 seconds digitally. The responsive interface supports both desktop and mobile access, optimally serving Admin and Field Officer needs. Thus, the system demonstrably enhances firefighting response efficiency and Hydrant data accuracy in Medan City.

PENGEMBANGAN APLIKASI SMARTSCHOOL BERBASIS WEB UNTUK ADMINISTRASI SEKOLAH DASAR DENGAN AGILE SCRUM

2025-09-21T09:56:24+00:00

The development of information technology has encouraged many educational institutions to adopt digital systems in order to improve the efficiency and accuracy of school administration management. However, at SD Qurratu A’yun, administrative processes such as recording student data, teacher data, and financial transactions are still carried out manually, leading to various problems such as data loss, reporting delays, and inconsistent information. Based on a case study at SD Qurratu A’yun, a web-based school administration information system was developed as a digital solution. The system was designed using the Agile method with a Scrum approach, enabling iterative development that is responsive to user needs. The development process involved several stages: product backlog planning, Sprint planning, Sprint backlog development, daily Scrum meetings, Sprint review, and Sprint retrospective. The system is equipped with key features such as teacher, student, and class data management, transaction type management, cash transactions, and financial reports accessible by academic year. System testing showed that the developed solution successfully improved data organization, accelerated reporting processes, and provided ease of access to information in real time for administrators, treasurers, and principals. By applying the Agile (Scrum) method and building a web-based system, this research provides an optimal solution to the challenges of manual administration at SD Qurratu A’yun. The system not only streamlines workflows but also supports the sustainability of school operations in a more efficient and structured manner.

IMPLEMENTASI METODE HYBRID FUZZY JARO WINKLER DAN COSINE SIMILARITY PADA SISTEM PENCARIAN AYAT AL-QURAN BERBASIS TRANSLITERASI LATIN

2025-09-22T14:42:48+00:00

This research addresses the challenge of retrieving Qur’anic verses in Latin transliteration, which is hindered by the absence of a standardized orthography, leading to diverse spelling variations. The study aims to design and implement a hybrid information retrieval system that integrates Fuzzy Jaro-Winkler for lexical similarity and Cosine Similarity on fine-tuned DistilBERT embeddings for semantic relevance. The system workflow begins with preprocessing and normalization of the dataset, followed by initial candidate selection using Jaro-Winkler, and final reranking through semantic similarity scoring. Evaluation was conducted using black-box testing across scenarios including ideal queries, spelling variations, incomplete queries, and varying query lengths. Results show high accuracy for ideal (96%) and varied spelling queries (92%), with performance improving as query length increases, reaching 96% for four-word queries. The hybrid approach effectively bridges lexical and semantic gaps, outperforming single-method baselines, and demonstrates robustness in handling non-standard transliteration in Qur’anic text retrieval.

IMPLEMENTASI HYBRID CNN, FACIAL LANDMARK DAN LIVENESS DETECTION PADA SISTEM ABSENSI WAJAH

2025-09-23T13:33:24+00:00

This paper presents the implementation of a hybrid approach for face recognition attendance systems, combining Convolutional Neural Network (CNN), facial landmark detection, and liveness detection. The CNN model extracts facial features for identity recognition, while facial landmark detection captures dynamic movements such as eye blinking and mouth motion. Liveness detection ensures system robustness against spoofing attempts including photo and video replay. The system was developed using Python with OpenCV, MediaPipe, and TensorFlow, and tested under multiple spoofing scenarios. Results show a detection accuracy of 95.5%, with real-time performance and resilience against common spoofing threats.

PERANCANGAN SISTEM PENCATATAN STOK ROKOK REAL-TIME BERBASIS WEB MENGGUNAKAN METODE V-MODEL

2025-09-29T21:44:45+00:00

The rapid development of information technology has encouraged companies to shift from manual processes to digital-based systems in order to improve operational efficiency and data accuracy. In cigarette distribution companies, manual recording of stock and distribution activities often causes various problems such as delayed reporting, discrepancies between warehouse and store stock data, and difficulties in generating accurate and timely reports. Based on a case study in a cigarette distribution company, a Web-Based Cigarette Distribution and Stock Management System was developed as a digital solution. The system was designed using the V-Model methodology, which emphasizes a strong correlation between development stages and testing phases. The system’s main features include stock data management (CRUD), stock request submission by sales, request verification by admins, as well as monitoring and report generation by managers. The implementation utilized PHP & MySQL for the backend, Bootstrap for the user interface, and AJAX to enable real-time data updates. System testing using the Black Box Testing method demonstrated that all key functionalities operated according to user requirements. The system successfully improved distribution efficiency, ensured data accuracy through input validation, and provided centralized data management for easier monitoring and auditing. Consequently, this system enables faster, more transparent, and integrated cigarette distribution processes, thereby supporting the company’s work effectiveness and managerial decision-making.

SISTEM INFORMASI LAYANAN JASA SAVE MY SHOE BERBASIS WEB MENGGUNAKAN METODE AGILE SCRUM

2025-09-29T22:04:45+00:00

Save My Shoe is a shoe cleaning service business that faces challenges in managing transactions, recording customer data, and monitoring services due to the absence of an integrated information system and reliance on social media. This study aims to design and implement a web-based information system that supports operational processes and service delivery in a more efficient and structured manner. The system was developed using the Agile method with a Scrum approach, allowing each feature to be built incrementally and tailored to User needs. The system includes features such as registration, Login, service ordering, payment, service progress tracking, and transaction reporting, with User roles consisting of admin and customer. Based on testing using the Black Box Testing method, all system functionalities operated according to the defined requirements. This information system is expected to improve business management effectiveness, enhance service transparency, and support the sustainable digitalization of MSMEs.