JURTEKSI (jurnal Teknologi dan Sistem Informasi)

MAMDANI FUZZY LOGIC ANALYSIS FOR ANIMAL MEDICINE STOCK OPTIMIZATION

Mutia Desmarini — 2025-09-15

Abstract: The management of veterinary drug stocks at the Veterinary Clinic Technical Implementation Unit (UPTD) of the North Sumatra Province Plantation and Livestock Service faces obstacles in the form of discrepancies between supply and demand, resulting in excess stock and budget waste. Uncertain demand for drugs is a factor that complicates decision-making in stock provision. This study aims to optimize drug stock management using the Mamdani fuzzy logic method, which is capable of handling data uncertainty and modeling information linguistically. Three input variables are used, namely initial stock, demand, and number of visits, with the output being the final stock. The process involves fuzzification, inference based on IF–THEN rules, and defuzzification using the centroid method. The results show that the developed system has a good accuracy level with a MAPE value of 17.52%, which means that this model is effective in providing optimal and efficient drug stock recommendations in a veterinary clinic environment.

Keywords: fuzzy mamdani; optimization; animal drug stock.

ANALYSIS OF PSI METHOD IN DECISION SUPPORT SYSTEM TO SELECT THE FEASIBILITY OF COVID 19 PATIENT DATA SCANNER RESULTS

Iskandar Zulkarnain — 2025-09-23

Abstract: Hospitals play an important role in examining the scan results of patient data infected with the Covid 19 virus. However, there are problems when processing the scan results, namely that sometimes errors occur in the scan data, causing many failures and delays in sending data to the Health Office. The purpose of this study is to build a Desktop-based decision support system application that can facilitate hospitals in selecting the eligibility of the scan results of Covid 19 patient data. The urgency in examining the scan results of Corona patient data is a very pressing public health issue, because the long-term impact is very significant for patients. Thus, a scientific discipline is needed that can support the decision-making process, namely the Decision Support System using the Preference Selection Index (PSI) method. PSI is a simple and easy calculation method, based on statistical concepts without having to determine attribute weights. The results of this method are clear and firm values based on the level of strength of the rules applied. The results of the research conducted on the PSI process can be concluded that valid Covid 19 patient data is Recap File I with a value of 0.2042 which is declared valid and accepted.

Keywords: covid-19; decision support system; PSI

BATTERY LIFESPAN PREDICTION FOR MOTORCYCLES USING DOUBLE MOVING AVERAGE

Heru Syahputra — 2025-09-23

Abstract: The inability to accurately monitor the lifespan of motorcycle batteries can lead to sudden failures, disrupt user activities, and increase maintenance costs. This issue is exacerbated by the absence of a predictive system that can assist users and workshops in planning maintenance and managing battery inventory effectively. This study aims to develop a battery lifespan prediction model for motorcycles using the Double Moving Average (DMA) method. The model is built based on historical data from 12 motorcycle units, including usage frequency, duration, terrain conditions, and maintenance habits. Forecasting is conducted through two stages of moving averages followed by trend parameter calculations. Evaluation results show that the model has a high level of accuracy, with MAPE = 0.10, MAD = 1.68, and RMSE = 2.14, indicating very low prediction errors. In addition, DMA is also used to forecast product demand at PT Anugerah Karya Abiwara Kisaran to prevent stock shortages. The system is developed using Visual Studio 2010 and Microsoft Access and has proven effective in supporting maintenance planning and inventory control. With its high accuracy and efficiency, the results of this study provide tangible contributions to decision-making in battery maintenance and inventory management.

Keywords: battery; DMA; motorcycle; prediction.

Abstrak: Ketidakmampuan dalam memantau usia pakai aki sepeda motor secara akurat dapat menyebabkan kerusakan mendadak, mengganggu aktivitas pengguna, serta meningkatkan biaya perawatan. Permasalahan ini diperburuk oleh tidak tersedianya sistem prediktif yang membantu pengguna dan bengkel dalam merencanakan perawatan serta mengelola persediaan aki secara efisien. Penelitian ini bertujuan untuk mengembangkan model prediksi usia pemakaian aki sepeda motor dengan menggunakan metode Double Moving Average (DMA). Model dibangun berdasarkan data historis dari 12 unit sepeda motor yang mencakup frekuensi penggunaan, durasi, kondisi medan dan kebiasaan perawatan. Proses peramalan dilakukan melalui dua tahap perataan bergerak, yang kemudian diikuti dengan perhitungan parameter tren. Hasil evaluasi menunjukkan bahwa model ini memiliki tingkat akurasi yang tinggi, dengan nilai MAPE sebesar 0,10, MAD sebesar 1,68, dan RMSE sebesar 2,14, yang mengindikasikan tingkat kesalahan prediksi yang sangat rendah. Selain itu, metode DMA juga diterapkan untuk meramalkan permintaan produk pada PT Anugerah Karya Abiwara Kisaran guna mencegah terjadinya kekurangan stok. Sistem dikembangkan menggunakan Visual Studio 2010 dan Microsoft Access, serta terbukti efektif dalam mendukung perencanaan perawatan dan pengendalian persediaan. Dengan akurasi dan efisiensi yang tinggi, hasil penelitian ini memberikan kontribusi nyata dalam pengambilan keputusan terkait pemeliharaan aki dan manajemen inventori.

Kata kunci: baterai; DMA; prediksi; sepeda motor.

INTEGRATED AHP-TOPSIS DECISION SYSTEM FOR FAIR STUDENT PERFORMANCE EVALUATION

Rahmad Hafiz — 2025-09-24

Giving awards is essential to motivate students; however, selecting outstanding students at the junior high school level is often conducted manually and subjectively, which can lead to unfairness and prolonged processing time. This study develops a Decision Support System (DSS) that integrates the Analytical Hierarchy Process (AHP) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to support objective and transparent student selection. A quantitative descriptive approach was employed, with data collected through questionnaires, interviews, and documentation at two state junior high schools in Banjarmasin City. Seven assessment criteria were applied: attendance, behavior, uniform neatness, extracurricular participation, academic grades, competition achievements, and disciplinary records. AHP was used to determine the weight of each criterion, while TOPSIS ranked students based on these weights. The web-based system was developed using PHP and MySQL and evaluated using the Technology Acceptance Model (TAM). Results show that academic grades had the highest weight (28.5%), followed by attendance (22.3%) and competition performance (15.2%). The TAM evaluation yielded average scores of 4.32 for Perceived Ease of Use, 4.40 for Perceived Usefulness, 4.15 for Attitudes Towards Use, and 4.28 for Behavioral Intention to Use. The DSS produces accurate rankings, is well-received by users, and offers an efficient, fair, and replicable solution for data-driven educational governance in the digital era.

CLOUD-DRIVEN OPTIMIZATION OF LECTURER PERFORMANCE DOCUMENT DIGITALIZATION USING AGILE UNIFIED PROCESS

Rio Irawan — 2025-09-24

The development of digital technology encourages universities to improve effectiveness and efficiency in data management, particularly in recording and reporting faculty performance. Some lecturers still face difficulties in reporting their performance in the SISTER application due to challenges in locating documents scattered across various archives, which often leads to issues such as delays in reporting, low information accuracy, and lack of transparency of faculty performance documents for institutional needs. This study aims to optimize the digitalization of faculty performance documents based on cloud computing using the Agile Unified Process (AUP) approach, which is implemented in the development of a cloud-based system by utilizing Google Drive as the storage medium for digital faculty performance documents. The AUP methodology was chosen for its ability to combine flexible iterative and incremental principles, allowing the system to adapt quickly and continuously to user needs. Testing using Equivalence Partitioning, based on the functional and non-functional requirements of the system, has shown results in accordance with expectations.

CRITERIA ANALYSIS OF COURSE PARTICIPANTS USING K-MEANS: A CASE STUDY OF INET PALEMBANG

Muhammad Rasuandi Akbar — 2025-09-30

Abstract: INET Computer Palembang, as a computer training institution, faces difficulties in understanding participant characteristics due to variations in age, educational background, and chosen course packages. This study aims to analyze participant criteria and group them based on similarities using the K-Means Clustering algorithm. The data used were historical records of course participants from 2022 to 2025. The research process followed the CRISP-DM stages, starting from data cleaning and transformation, determining the optimal number of clusters using the Elbow Method, to evaluating cluster quality with the Davies-Bouldin Index. The implementation was carried out using Python and the scikit-learn library. The results show that the optimal number of clusters is k=5 with a Sum of Squared Errors (SSE) value of 1064.66 and a Davies-Bouldin Index (DBI) score of 0.820, indicating good cluster quality. The resulting clustering provides a structured profile of participants and demonstrates that K-Means is effective in segmenting course participants. These findings are expected to assist the institution in designing more targeted training programs.

Keywords: clustering; data mining; elbow method; k-means; computer course

ANALYSIS OF THE QUALITY OF "ONLINE EQUIVALENT" E-LEARNING USING WEBQUAL 4.0 AND IPA METHODS

Taufik Rahman — 2025-09-29

Abstract: The use of e-learning in non-formal education is increasingly important to support the improvement of access to learning, one of which is through the online platform. This study aims to analyze the quality of online services using WebQual 4.0 and Im-portance Performance Analysis (IPA) methods to evaluate the suitability between user expectations and perceptions. The research method used a quantitative approach by distributing questionnaires to active users, then analyzed using the WebQual Index to measure the overall quality of the system as well as the IPA to determine improvement priorities. The results showed that the quality of SeTARA Online was relatively good with a WebQual Index value of 0.798. However, there is still a gap between user expectations and satisfaction with a negative gap value of -0.238. The IPA analysis identified indicators in Quadrant I as priority improvements, especially in the aspects of service interaction and information presentation. These findings underscore the need for continuous development of features and technical support to optimize the user experience. The conclusion of this study suggests that there should be improvements in priority indicators to increase user satisfaction, as well as strengthen the effectiveness of online learning. Advanced research can expand variables, compare with other platforms, and combine quantitative and qualitative analysis methods for more comprehensive results.

Keywords: e-learning; importance performance analysis; quality of service; online equivalent; webqual 4.0

VEGECHAIN: SMART CONTRACT MARKETPLACE FOR VEGETARIAN SUPPLY CHAIN OPTIMIZATION

Eka Lia Febrianti — 2025-09-30

Abstract: The global transition towards sustainable food systems faces significant challenges in vegetarian food supply chains, including transparency issues, distribution inefficiencies, and quality verification problems. This research proposes VegeChain development, a decentralized marketplace ecosystem based on smart contracts designed to transform vegetarian food supply chains and accelerate Meatless, Balanced, Green (MBG) program adoption. Using mixed-method methodology integrating blockchain system design, stakeholder analysis, and economic simulation, this research develops a comprehensive technology framework combining blockchain transparency, smart contract automation, and sustainable tokenomics with novel mathematical models. The system implements dynamic pricing algorithms based on Automated Market Maker (AMM) mechanisms, multi-objective optimization for supply chain efficiency, and reputation-based consensus protocols. Simulation results demonstrate that VegeChain implementation can improve supply chain efficiency by 35%, reduce food waste by 28%, and increase consumer trust by 42% measured through validated stakeholder satisfaction surveys (n=456) using 5-point Likert scales with statistical significance p<0.001. Technical innovations include Byzantine Fault Tolerant consensus with 99.9% reliability, gas optimization achieving 67% cost reduction, and real-time quality verification algorithms with 98.7% accuracy.

Keywords: smart contracts; supply chain optimization; automated market makers; blockchain technology; sustainable tokenomics

CNN-BASED ADAPTIVE IDS WITH FEDERATED LEARNING FOR IOT NETWORK SECURITY

Sahren Sahren — 2025-09-30

Abstract: In the era of the Internet of Things (IoT), cyber threats are increasingly complex and dynamic, thus demanding an adaptive and intelligent network security system. This study proposes a Convolutional Neural Network (CNN)-based Intrusion Detection System (IDS) implemented through a Federated Learning (FL) approach in a Non-Independent and Identically Distributed (Non-IID) data environment. This approach allows the model to be trained in a distributed manner across multiple IoT devices without having to collect sensitive data to a central server, thereby maintaining data privacy while increasing the efficiency of the training process. The experiment used the CIC IoT 2023 dataset, which represents various modern IoT network traffic patterns. The results show that the proposed CNN–FL model achieves an overall accuracy of 0.99, with excellent performance in detecting various types of network traffic. The model obtains a perfect recall value (1.00) for normal traffic (Benign), as well as a very high F1-score for DDoS (0.99) and DoS (0.99) attacks. Stable and consistent performance across all five federation rounds demonstrates that this approach is a reliable, efficient, and accurate solution for detecting threats in distributed and privacy-preserving IoT networks.

Keywords: cnn; federated_learning; ids; non-iid; ciciot2023

Abstrak: Dalam era Internet of Things (IoT), ancaman siber semakin kompleks dan dinamis, sehingga menuntut sistem keamanan jaringan yang adaptif dan cerdas. Penelitian ini mengusulkan Intrusion Detection System (IDS) berbasis Convolutional Neural Network (CNN) yang diterapkan melalui pendekatan Federated Learning (FL) pada lingkungan data yang bersifat Non-Independent and Identically Distributed (Non-IID). Pendekatan ini memungkinkan model dilatih secara terdistribusi di berbagai perangkat IoT tanpa harus mengumpulkan data sensitif ke server pusat, sehingga mampu menjaga privasi data sekaligus meningkatkan efisiensi proses pelatihan. Eksperimen menggunakan dataset CIC IoT 2023, yang merepresentasikan berbagai pola lalu lintas jaringan IoT modern. Hasil penelitian menunjukkan bahwa model CNN–FL yang diusulkan mencapai akurasi keseluruhan sebesar 0.99, dengan performa yang sangat baik dalam mendeteksi berbagai jenis lalu lintas jaringan. Model memperoleh nilai recall sempurna (1.00) untuk lalu lintas normal (Benign), serta nilai F1-score yang sangat tinggi untuk serangan DDoS (0.99) dan DoS (0.99). Kinerja yang stabil dan konsisten di seluruh lima putaran federasi membuktikan bahwa pendekatan ini merupakan solusi yang andal, efisien, dan akurat untuk mendeteksi ancaman pada jaringan IoT yang bersifat terdistribusi dan menjaga privasi (privacy-preserving).

Kata kunci: cnn; federated_learning; ids; non-iid; ciciot2023

DEVELOPMENT OF A BLOCKCHAIN-BASED DECENTRALISED APPLICATION WITH NFT FOR LAND REGISTRATION

Rahmat Nugrohoning Gesang — 2025-09-30

Abstract: Land registration in Indonesia often encounters challenges in transparency, data integrity, and centralized bureaucracy. Manual and semi-digital systems remain vulnerable to manipulation and delays. The National Land Agency has initiated digitalization, but several challenges remain, particularly in ensuring transparency, efficiency, and security of land ownership data. Blockchain technology offers a potential solution through its decentralized and immutable characteristics. This study adopted a design and development method consisting of system analysis, requirements identification, architecture design, implementation, and black-box testing. The developed decentralized application (DApp) integrates smart contracts, NFTs, and IPFS to manage land certificates. Core functions such as minting, transfer, splitting, and self-custody were implemented and successfully tested, with all scenarios producing expected results. The findings demonstrate that blockchain integration can enhance security, reduce duplication, and streamline land administration. The study contributes a functional prototype with practical implications for modernizing land registration in Indonesia while identifying scalability and regulatory adaptation as areas for further research.

Keywords: blockchain; decentralized application; land registration; NFT; smart contract.

MACHINE LEARNING CONTENT-BASED FILTERING WOMEN EMPOWERING RECOMMENDATIONS ON YOUTUBE

Yuliana Yuliana — 2025-09-30

Abstract: YouTube is one of the most popular video streaming platforms, but it has constraints that can cause problems when clients have difficulty finding content according to their wishes. The main objective of this study is to increase user capacity in viewing content specifically in the field of women's empowerment. By using content-based filtering techniques, the system will analyze user preferences and interests through recommendations for women's empowerment content. The data source is via the YouTube API and is analyzed using PHP programming content-based filtering techniques. The system's recommendations provide a list of women's empowerment content with a user request display. The results of the research evaluation obtained a precision value of 62%, meaning that the recommendations match the topic being searched for, namely women's empowerment. The recall value of 84% indicates that the system has succeeded in finding relations from the database. The f1-score value of 72% indicates that there is a balance between precision and recall, meaning that a system is needed that is not only accurate but also complete. While the cosine value shows a score of 0.7071 approaching the maximum value (1.0). The recommendation of the content-based filtering method produces quite effective women's empowerment content.

Keywords: content-based filtering, recommendations, women Empowerment, youtube

DEVELOPMENT RICE PLANT DISEASE CLASSIFICATION USING CNN WITH TRANSFER LEARNING

Fachri Ayudi Fitrony — 2025-10-05

Abstract: The rice plant, Oryza sativa, is a major food source in Indonesia. This plant is processed into rice, a staple food for the Indonesian people. Rice growth is crucial to ensure the rice produced is of good quality. One part of the rice plant that is susceptible to disease is the leaves, which can inhibit growth and reduce rice quality. Therefore, early detection and accurate classification of rice diseases are crucial to minimize these negative impacts. This has driven the development of a Deep Learning model capable of high-performance automatic classification. This study aims to create a rice leaf classification model using the CNN algorithm and several transfer learning architectures such as ResNet101, VGG16, and Xception. A dataset of 859 rice leaf images collected from the Kaggle website was then processed using augmentation techniques to a total of 2,439 images, plus 215 smartphone photos for external data validation. Thus, the total dataset increased to 2,656 images, covering four categories: leafblast, brownspot, healthy, and hispa. The model was processed in two stages: on the initial dataset (Non-Augmented Dataset) and the Augmented Dataset. The best experimental results were obtained using the ResNet architecture, with a training accuracy of 96.17% and a validation accuracy of 95.22%. Based on the research results, the rice plant disease classification model using deep learning demonstrated good performance.

Keywords: convolutional neural network; deep learning; fine-tuning; image classification; resnet; rice plant

WEBGIS-BASED GEOGRAPHICAL INFORMATION SYSTEM FOR MAPPING BAKERY SHOPS IN KISARAN CITY

Yori Apridonal M — 2025-09-28

Abstract: Bakery MSMEs in Kisaran City play a significant role in the local economy, but their distribution data is still managed manually, making it difficult to access and analyze. This study aims to develop a WebGIS-based Geographic Information System to map and manage bakery MSME data digitally and in an integrated manner. The research methods include field surveys, spatial and non-spatial data collection, system design using UML, development with Leaflet.js and MySQL, and testing using the blackbox method. The results show that the resulting system is capable of displaying interactive maps with location details, business information, and an easy-to-use search feature. This system makes it easier for the government, business actors, and the public to access MSME information and supports data-based economic development planning. With the output in the form of publications in accredited journals, this research is expected to be an effective solution for MSME data management in other regions.

Keyword: bakery; mapping; MSME; WebGIS

Abstrak: UMKM toko roti di Kota Kisaran memiliki peran penting dalam perekonomian lokal, namun data persebarannya masih dikelola secara manual sehingga sulit diakses dan dianalisis. Penelitian ini bertujuan mengembangkan Sistem Informasi Geografis berbasis WebGIS untuk memetakan dan mengelola data UMKM toko roti secara digital dan terintegrasi. Metode penelitian meliputi survei lapangan, pengumpulan data spasial dan non-spasial, perancangan sistem menggunakan UML, pengembangan dengan Leaflet.js dan MySQL, serta pengujian menggunakan metode blackbox. Hasil penelitian menunjukkan sistem yang dihasilkan mampu menampilkan peta interaktif dengan detail lokasi, informasi usaha, dan fitur pencarian yang mudah digunakan. Sistem ini mempermudah pemerintah, pelaku usaha, dan masyarakat dalam mengakses informasi UMKM, serta mendukung perencanaan pembangunan ekonomi berbasis data. Dengan luaran berupa publikasi pada jurnal terakreditasi, penelitian ini diharapkan menjadi solusi efektif untuk pengelolaan data UMKM di daerah lain.

Kata kunci; pemetaan; toko roti; UMKM; WebGIS

DEVELOPMENT OF A QR CODE-BASED WEBAR TO DIGITIZE LOCAL WISDOM AS AN EFFORT TO INCREASE TOURIST ATTRACTION IN BORDER AREAS

Noviyanti P — 2025-09-30

Abstract: Current technological developments play a crucial role in driving the tourism industry, one of which is the utilization of technology. However, tourist attractions in border areas face challenges such as limited access to digital information and a lack of interactive media to present local wisdom such as history, customs, and culture. This study aims to develop and implement a QR Code-based Web-based Augmented Reality (WebAR) to digitize local wisdom at tourist attractions, making information more engaging and accessible to tourists. The research methodology adopted three approaches: UCD (User-Centered Design), Agile methods, and TAM (Technology Acceptance Model). This platform contains the local wisdom of two tourist villages in the border area, namely Sebujit Village and Jagoi Babang Village. The results of testing and evaluation using multiple linear regression and SEM-PLS methods on 45 respondents showed that the WebAR technology acceptance model was significant (F = 6.583; p < 0.001). User Attitude (ATU) is a key variable that significantly influences Intention to Use (BI) (β=0.429; p=0.001), while Ease of Use (PEOU) and Benefit (PU) indirectly influence BI through ATU. As additional validation, the classification test yielded an accuracy of 88.90% and an F1-score of 0.941, confirming that QR Code-based WebAR is effective and well-received as a digital information and promotion medium for local wisdom in border areas.

Keywords: border areas; local wisdom; qr code; tourist attractions; web augmented reality.

Abstrak: Perkembangan teknologi saat ini memiliki peran penting dalam mendorong industri pariwisata, salah satunya dengan pemanfaatan teknologi. Namun, objek wisata di daerah perbatasan menghadapi tantangan seperti keterbatasan akses informasi digital dan kurangnya media interaktif untuk menyajikan kearifan lokal seperti sejarah, adat istiadat, dan budaya. Penelitian ini bertujuan untuk mengembangkan dan mengimplementasikan Web-based Augmented Reality (WebAR) berbasis QR Code untuk digitalisasi kearifan lokal objek wisata, menjadikan informasi lebih menarik dan mudah diakses bagi wisatawan. Metodologi penelitian mengadopsi tiga pendekatan, UCD (User-Centered Design), metode Agile, dan TAM (Technology Acceptance Model). Platform ini memuat kearifan lokal dua desa wisata di daerah perbatasan, yaitu Desa Sebujit dan Desa Jagoi Babang. Hasil pengujian dan evaluasi dengan metode regresi linier berganda dan SEM-PLS pada 45 responden menunjukkan bahwa model penerimaan teknologi WebAR ini signifikan (F = 6.583; p < 0.001). Sikap Pengguna (ATU) menjadi variabel kunci yang berpengaruh signifikan terhadap Niat Penggunaan (BI) (β=0.429; p=0.001), sementara Kemudahan Penggunaan (PEOU) dan Manfaat (PU) memengaruhi BI secara tidak langsung melalui ATU. Sebagai validasi tambahan, uji klasifikasi menghasilkan akurasi 88,90% dan F1-score 0,941, yang menegaskan bahwa WebAR berbasis Kode QR efektif dan dapat diterima dengan baik sebagai media informasi dan promosi digital kearifan lokal di wilayah perbatasan.

Kata Kunci: daerah perbatasan; kearifan lokal; qr code; web augmented reality.

COMPARATIVE ANALYSIS OF MACHINE LEARNING ALGORITHMS FOR COSMETIC SALES PREDICTION ON TOKOPEDIA

Mutia Sahira — 2025-09-30

Abstract: The rapid growth of the cosmetics industry on e-commerce platforms has intensified competition, creating a critical need for effective, data-driven marketing strategies. This study aims to conduct a comparative analysis of machine learning algorithms to predict the sales categories (High, Medium, Low) of cosmetic products on the Tokopedia marketplace. Four classification models; Random Forest, XGBoost, Logistic Regression, and Naive Bayes were trained and evaluated on data collected via web scraping. The methodology incorporates the Synthetic Minority Over-sampling Technique (SMOTE) to address significant class imbalance and GridSearchCV for hyperparameter optimization to ensure a fair and robust comparison. The experimental results conclusively show that the Random Forest model achieved the best performance, yielding the highest F1-Score Macro Average of 0.75 and an accuracy of 85.3%. The superior model was subsequently implemented in a simple recommendation system to simulate optimal discount strategies, demonstrating its practical utility in providing actionable insights for business decisions.

Keywords: classification; comparative analysis; machine learning; sales prediction; SMOTE

Abstrak: Pertumbuhan pesat industri kosmetik pada platform e-commerce telah membuat persaingan ketat, sehingga menciptakan kebutuhan krusial akan strategi pemasaran yang efektif dan berbasis data. Penelitian ini bertujuan untuk melakukan analisis komparatif terhadap algoritma machine learning untuk memprediksi kategori penjualan (Tinggi, Sedang, Rendah) produk kosmetik di marketplace Tokopedia. Empat model klasifikasi, yaitu Random Forest, XGBoost, Regresi Logistik, dan Naive Bayes, dilatih dan dievaluasi menggunakan data yang dikumpulkan melalui web scraping. Metodologi penelitian ini menerapkan Synthetic Minority Over-sampling Technique (SMOTE) untuk mengatasi ketidakseimbangan kelas yang signifikan dan GridSearchCV untuk optimisasi hyperparameter guna memastikan perbandingan yang adil. Hasil eksperimen menunjukkan bahwa model Random Forest mencapai performa terbaik, dengan menghasilkan F1-Score Macro Average tertinggi sebesar 0,75 dan akurasi 85,3%. Model unggul ini kemudian diimplementasikan dalam sebuah sistem rekomendasi sederhana untuk menyimulasikan strategi diskon yang optimal, yang menunjukkan kegunaan praktisnya dalam memberikan wawasan yang dapat ditindaklanjuti untuk pengambilan keputusan bisnis.

Kata kunci: analisis komparatif; klasifikasi; machine learning; prediksi penjualan; SMOTE

DEVELOPMENT INTEGRATIVE MODEL FOR ACADEMIC INFORMATION SYSTEMS USING UTAUT, DELONE&MCLEAN, AND TTF

Asep Hilmi Mutakin — 2025-10-11

Abstract: The Academic Information System (SIAKAD) plays an important role in supporting the management of academic administration in higher education institutions, particularly for students. ABC University has implemented SIAKAD since 2018 to facilitate administrative ativities in line with its motto of a high technology campus. This study aims to measure the sucess of SIAKAD usage from the aspects of acceptance, satisfaction, suitability, and perceived benefits. The integration of the Unified Theory of Acceptance and Use of Technology (UTAUT), DeLone & McLean, and Task Technology Fit (TTF) models was carried out to obain a more comprehensive overview in assessing the success of SIAKAD. UTAUT explains the factors influencing the intention to use, DeLone & McLean emphasizes the relationship between system quality and both user satisfaction and net benefits, while TTF evaluates the fit between technology and user tasks. By combining these three models, the study addresses the limitations of each model and produces a more holistic approach in measuring acceptance, success, and the appropriateness of system use. The testing was conducted using SPSS and Structural Equation Modeling (SEM) analysis through AMOS.

Keywords: siakad; utaut; delone&mclean; penerimaan teknologi; sem