Jurnal Buana Informatika
Not a member yet
274 research outputs found
Sort by
Penerapan Optical Character Recognition untuk Pengenalan Variasi Teks pada Media Presentasi Pembelajaran
Full text linkMedia digital merupakan bentuk utama media pembelajaran yang banyak digunakan untuk kegiatan belajar mengajar di kelas saat ini. Media pembelajaran digital umumnya tersimpan dalam bentuk citra karena memiliki unsur visual di dalamnya. Salah satu kelemahan data dalam bentuk citra adalah seluruh isi di dalamnya dianggap sebagai gambar, sementara pada media pembelajaran juga terdapat unsur teks di dalamnya. Oleh karena itu, dibutuhkan metode OCR untuk membaca teks di dalamnya agar media tersebut dapat diolah lebih lanjut, misalnya untuk keperluan kategorisasi (indexing) atau untuk dibaca pada sistem lain seperti chatbot. Umumnya, metode OCR digunakan untuk mengenali tulisan dengan bentuk yang seragam pada sebuah citra. Sedangkan pada media pembelajaran, teks di dalamnya memiliki variasi yang berbeda-beda. Penelitian ini mencoba menerapkan metode OCR dengan menggunakan Tesseract untuk menguji 30 data media pembelajaran yang memiliki berbagai macam variasi teks dalam sebuah citra. Hasil pengujian menunjukkan tingkat akurasi pengenalan teks yang cukup baik, yaitu sebesar 91,11%
Evaluasi Performa Kernel SVM dalam Analisis Sentimen Review Aplikasi ChatGPT Menggunakan Hyperparameter dan VADER Lexicon
Full text linkChatGPT merupakan model bahasa kecerdasan buatan yang merespon pertanyaan dan pernyataan pengguna. ChatGPT memiliki manfaat dan kelemahan bagi pengguna. Hal ini menimbulkan komentar pada media sosial tentang manfaat dari ChatGPT. Penelitian ini membahas tentang analisis sentimen review aplikasi ChatGPT menggunakan SVM kernel linier, RBF, polinomial dan sigmoid. Pelabelan menggunakan VADER lexicon dan hyperparameter untuk menghasilkan parameter terbaik. Tujuan penelitian yaitu apakah aplikasi ChatGPT dapat memberikan manfaat dan membuktikan apakah kernel pada SVM dapat meningkatkan nilai akurasi. Diskenariokan persentase pembagian antara data uji dan data latih adalah 70:30, 80:30, dan 90:10. Setelah dilakukan preprocessing, kemudian dikelompokkan menjadi review positif dan negatif. Dilakukan hyperparameter terhadap parameter C dan Gamma sehingga menghasilkan nilai maksimal. Hasil eksperimen diperoleh akurasi tertinggi menggunakan SVM kernel RBF skenario 90:10 dengan nilai accuracy 92.72%, precision 92.44%, f1-score 96.10% dan AUC 88%
Gamified Distance Learning Application Design for Enhanced Student Engagement and User Experience
Full text linkDistance Learning in Indonesia is one of the learning methods that began to be applied during the Covid-19 pandemic. Yet students face some obstacles, such as lack of motivation, struggling with operating learning devices, difficulty maintaining focus, and student engagement during the learning process. Gamification offers a solution to these problems by significantly enhancing user motivation and engagement, as it has been tested in research to have a profound impact. Therefore, this study aims to design a mobile application for Distance Learning by implementing gamification. It employs qualitative and quantitative data, including 32 students' responses from questionnaires like UEQ-S, utilized for testing user interface, and UES-SF, employed for testing gamification elements. By implementing gamification in this design, an engagement score of 83% was obtained, and the overall UEQ-S result was 1.89 in the Excellent category
Sentiment Analysis of DKI Jakarta 2024 Election (Case Study: Anies Baswedan and Ridwan Kamil)
Full text linkThis study analyzes public sentiment toward two potential candidates for the 2024 Jakarta gubernatorial election, Anies Baswedan and Ridwan Kamil, using Twitter data. Applying the TextBlob model for text extraction and Naive Bayes for sentiment classification found that sentiment toward Anies Baswedan is mostly positive, 52.2%, while neutral sentiment dominates for Ridwan Kamil. The accuracy of the Naive Bayes model reached 80% for Anies Baswedan and 72% for Ridwan Kamil, with higher precision, recall, and F1-score for Anies' data. These results indicate that the model is more accurate in classifying sentiment toward Anies compared to Ridwan Kamil. The implications of these findings are important for political campaign strategies, where Anies can leverage the existing positive support, while Ridwan Kamil has an opportunity to strengthen public engagement through a more strategic approach, in line with the sentiment emerging on social media
Implementasi Data Mining untuk Estimasi Produksi Cabai menggunakan Metode Exponential Smoothing
Full text linkCabai merupakan komoditas hortikultura yang banyak dibudidayakan dan berpengaruh pada fluktuasi ekonomi di Kabupaten Sleman. Dalam upaya menstabilkan fluktuasi harga dan pertumbuhan ekonomi di Kabupaten sleman, maka perlu dilakukan estimasi produksi cabai untuk periode ke depan. Estimasi produksi cabai yang dilakukan dalam penelitian ini menggunakan tiga jenis metode Exponential Smoothing dengan kombinasi parameter alpha, beta, dan gamma. Penelitian ini bertujuan untuk mengembangkan model estimasi produksi cabai dengan menggunakan Single, Double, dan Triple Exponential Smoothing. Hasil penelitian ini menunjukkan bahwa Triple Exponential Smoothing adalah metode yang paling tepat digunakan untuk mengestimasi produksi cabai di masa mendatang, dengan persentase tingkat error sebesar 6.5%
Design and Implementation of Load Balancing for Quality of Service Improvement
Full text linkAt the Information Technology Faculty, Satya Wacana Christian University, load balancing systems are implemented where the web server serves 500 users. This is to prevent server overload or downtime during simultaneous access to the web server. Test results indicate significant differences in CPU usage, request time, and bandwidth between load balancing and single servers. The use of load balancing is more effective than relying on a single server, as evidenced by test results. The CPU usage with load balancing is significantly lower, with a difference of up to 45% compared to a single server. The request time with load balancing is also slightly better, with only 21.5ms compared to 42ms for a single server. However, the difference in bandwidth between load balancing and a single server is not very significant. The highest bandwidth recorded on a single server is 182kb/s, while with load balancing it reaches 165kb/s
Konfigurasi Model Prophet Untuk Prediksi Harga Saham Sektor Teknologi di Indonesia Yang Akurat
Full text linkSaham merupakan salah satu instrumen investasi yang sedang ramai dan digemari oleh masyarakat muda Indonesia. Untuk dapat meramal harga saham, dapat dilakukan analisis teknikal dengan menerapkan machine learning. Namun, untuk dapat menggunakan machine learning, diperlukan implementasi algoritma yang membutuhkan waktu panjang serta keterampilan tinggi. Maka dari itu digunakanlah model Prophet, model machine learning yang mudah untuk dikembangkan. Pengembangan dilakukan dengan menyesuaikan karakteristik data saham yang merupakan data bertipe time series. Eksperimen dilakukan untuk menemukan konfigurasi yang perlu dilakukan terhadap model dalam menghasilkan peramalan yang paling akurat. Melalui penelitian yang dilakukan, hasil terbaik yang didapatkan adalah model Prophet yang menggunakan dataset paling banyak dan juga melalui hyperparameter tuning. Hal ini dapat dibuktikan dengan visualisasi yang ada serta nilai error yang rendah, dimana MAPE (Mean Absolute Percentage Error) mempunyai nilai error sebesar 15%
Machine Learning for Environmental Health: Optimizing ConcaveLSTM for Air Quality Prediction
Full text linkThis study investigates the optimization of the ConcaveLSTM model for air quality prediction, focusing on the interplay between input sequence lengths and the number of LSTM units to enhance forecasting accuracy. Through the evaluation of various model configurations against performance metrics such as RMSE, MAE, MAPE, and R-squared, an optimal setup featuring 50 input steps and 300 neurons was identified, demonstrating superior predictive capabilities. The findings underscore the critical role of model parameter tuning in capturing temporal dependencies within environmental data. Despite limitations related to dataset representativeness and environmental variability, the research provides a solid foundation for future advancements in predictive environmental modeling. Recommendations include expanding dataset diversity, exploring hybrid models, and implementing real-time data integration to improve model generalizability and applicability in real-world scenarios
Optimizing Book Delivery Routes Using Genetic Algorithms: Case Study of Erlangga Publisher Yogyakarta Branch
Full text linkOptimizing Book Delivery Routes Using Genetic Algorithms: Case Study of Erlangga Publisher Yogyakarta Branch. This research aims to find the shortest route for book delivery using the Traveling Salesperson Problem (TSP) approach that is solved by a Genetic Algorithm (GA). The distance between the pair of locations will be known by using the longitude and latitude as the coordinates of the location (the place where books must be dropped and the trip continues). This network of the coordinates of locations is then viewed as TSP, which needs GA to solve the shortest path. Running the program for up to 100 iterations, this study resulted in the shortest route, 356 km in a whole route. Among the previous research, this research has its uniqueness, especially when the problem is viewed as a TSP, and when it comes to the crossover mechanism, it is quite rare. Moreover, the case of the Erlangga publisher is the first case that has used the GA
Identifikasi Kendaraan Beroda Menggunakan Algoritma YOLOv5
Full text linkThe importance of traffic density measurement in road planning has led to efforts in automation using object detection algorithms, particularly YOLO (You Only Look Once), which are replacing error-prone and time-consuming manual processes. However, challenges arise in dense traffic conditions, posing a challenge to vehicle detection accuracy. This research aims to compare the performance of vehicle detection between two YOLO approaches: multi-view layer detection and conventional detection, focusing on YOLOv5n, YOLOv5s, and YOLOv5m. The literature review encompasses Computer Vision, YOLO implementation, and related research to provide conceptual context. The research method details the steps of vehicle identification using YOLOv5, and the evaluation includes the performance of various YOLO variants and multi-view detection approaches. Thus, this study is expected to gain deeper insights into building an effective model and facilitating the selection of a suitable YOLO model for vehicle detection