Telematika
Not a member yet
319 research outputs found
Sort by
Pengembangan Website untuk Sistem Pengontrolan Reservoir Air Bersih Berbasis IoT pada Bandung City View I
Full text linkPerumahan Bandung City View I (BCV I) sering menghadapi masalah meluapnya air di reservoir atas, terutama saat penggunaan air warga rendah pada malam hari. Untuk mengatasi masalah ini, dikembangkan sebuah sistem pengontrolan reservoir air berbasis IoT yang dapat beroperasi secara kontinu selama 24 jam. Sistem ini menggunakan perangkat keras yang aman dan dapat diakses melalui web dan aplikasi mobile untuk mempermudah proses pemantauan dan pengontrolan reservoir air secara jarak jauh dan real-time. Pengembangan website untuk sistem ini melibatkan penggunaan HTML, CSS, dan JavaScript di bagian frontend, serta Laravel framework di bagian backend. Firebase Realtime Database digunakan untuk penyimpanan data dan sinkronisasi real-time. Pengujian sistem menunjukkan bahwa website yang dikembangkan dapat mengontrol pompa dan memonitor kondisi reservoir dengan efektif, mengurangi pemborosan air, dan menurunkan biaya operasional. Hasil ini mengindikasikan bahwa sistem pengontrolan reservoir berbasis IoT yang diakses melalui website dapat menjadi solusi yang efisien dan efektif dalam mengelola sumber daya air di perumahan BCV I
Forecasting Sea Surface Salinity in the Eastern Madura Strait Using a 1D Convolutional Neural Network
Full text linkTujuan: Penelitian ini bertujuan untuk memprediksi salinitas permukaan air laut pada perairan Selat Madura bagian Timur menggunakan 1D CNN dan menguji daripada performa model arsitektur 1D CNN yang dibuat. Berdasarkan hasil prediksi yang diperoleh, diharapkan mampu memberi informasi ke masyarakat terkait kondisi salinitas permukaan Selat Madura bagian Timur beberapa hari ke depan.Perancangan/metode/pendekatan: Hal pertama yang perlu dilakukan adalah memprediksi tiap parameter sebelum memprediksi salinitas permukaan. Penelitian ini menggunakan metode 1D CNN, dengan parameter kecepatan arus eastward, arus northward dengan 3 kedalaman berbeda, dan salinitas pada 2 kedalaman berbeda.Hasil: Berdasarkan penelitian ini diperoleh model 1D CNN mampu memprediksi salinitas dengan sangat baik, dengan MAPE sebesar 2.86% pada nilai dropout 0.8 dan batchsize 64. Adapun hasil prediksi untuk 6 hari ke depan, dari 17 Januari 2023 pukul 19.00 hingga 23 Januari 2023 pukul 07.00 dengan rentang waktu per 12 jam adalah mengalami penurunan dengan angka terendah menyentuh 33.313 PSU.Keaslian/ state of the art: Pada penelitian ini menggunakan parameter prediksi, metode, dan diperoleh hasil yang berbeda dengan penelitian sebelumnya
Prediction And Detection Of Type II Diabetes Mellitus Using The K-Nearest Neighbor Algorithm
No full textPurpose: High blood sugar causes Mellitus (DM), a metabolic disorder. DM affects human metabolism and causes many complications, such as heart disease, kidney problems, skin disorders, and slow healing. Therefore, using machine learning algorithms to implement an automatic diabetes diagnosis system is crucial for predicting DM.Design/methodology/approach: This research created a DM disease prediction system using machine learning with the K-Nearest Neighbor algorithm. The National Institute of Diabetes and Digestive and Kidney Diseases, Hospital Frankfurt, Germany, and the results of health surveys and medical research are the sources of two separate datasets used in the Kaggle platform data. The stages in Machine Learning include data merging, data cleaning, and data splittingFindings/result: This research produces the best prediction model at a ratio of 70:30, with the lowest MSE value on testing data, 0.217. With K Folding Cross-validation, it makes an average accuracy of 73.88%.Originality/value/state of the art: This research creates a prediction model for diabetes mellitus type 2 using two different datasets with 9 features. It makes a Machine Learning model using the KNN algorithm by importing the KneighborClassifier and evaluating it using the MSE (Mean Square Error) matrix and K Folding cross-validation to determine modelling accurac
Evaluasi Kualitas Sistem Informasi Akademik dengan Standar ISO/IEC 25010 (Studi Kasus: Universitas ABC)
Full text linkTujuan: Penelitian ini bertujuan untuk mengevaluasi kualitas sistem informasi akademik di Universitas ABC berdasarkan standar ISO/IEC 25010.Perancangan/metode/pendekatan: Penelitian ini menggunakan pendekatan pengujian yang terstruktur sesuai dengan kriteria-kriteria yang terdapat dalam ISO/IEC 25010.Hasil: Pengujian menggunakan standar ISO/IEC 25010 menunjukan bahwa tujuh dari delapan standar yaitu: (a) functional suitability berjalan 100% dengan nilai satu; (b) performance efficiency mendapatkan grade B; (c) compatibility mendapat skor 100%; (d) usability dengan skor rata-rata 78.6%; (e) reliability dengan skor 100%; (f) maintainability memenuhi aspek identifikasi Land; dan (g) portability memperoleh skor 100% telah memenuhi standar, sedangkan security pada level dua (medium) belum memenuhi standar ISO/IEC 25010.Keaslian/ state of the art: Penggunaan seluruh standar ISO/IEC 25010 memberikan analisis yang lebih luas dan komprehensif dalam melakukan evaluasi kualitas sistem informasi akademik di Universitas ABC. Menggunakan metode observasi dan pengujian langsung dengan berbagai alat pengujian memberikan informasi yang objektif dengan kondisi sistem yang sebenarnya
Classification of apple maturity based on color using the K-Nearest Neighboor (KNN) method
Full text linkPurpose: The aim of this research is to provide support to apple fans and farmers in determining the choice of fruit that is ripe and ready to be consumed, using indicators of outer skin color as a basis for classification.Design/methodology/approach: The approach uses the K-Nearest Neighbor (KNN) method to classify the level of ripeness of apples based on skin color. KNN is used as a classification method. This approach utilizes the similarity of skin color with training data to determine the level of maturity. The evaluation results showed an accuracy of 90%, making it an effective approach for identifying the ripeness level of apples.Findings/result: From the results of the system evaluation of 206, it shows an accuracy level of 90% with a sensitivity of 80% and a specificity of 67% as measured by the Hold Out Estimation model.Originality/value/state of the art: This research uses test data/testing data originating from Kaggle and Google as well as several photos taken directly. In total, 206 images of apples were used
Implementation of Mel-Frequency Cepstral Coefficient As Feature Extraction Method On Speech Audio Data
Full text linkSounds cannot be directly processed by machines without a feature extraction process being carried out first. Currently, there are so many choices of feature extraction methods that can be used, so determining the right feature extraction method is not easy. One method of feature extraction on sound signals that is often used is Mel-Frequency Cepstral Coefficient (MFCC). MFCC has a working principle that resembles the human hearing system, which causes it to be widely used in various tasks related to recognition based on sound signals. This research will use the MFCC method to extract characteristics on voice signals and Support Vector Machine as a method of emotion classification on the RAVDESS dataset. MFCC consists of several stages, namely Pre-emphasize, Frame Blocking, Windowing, Fast Fourier Transform, Mel-Scaled Filterbank, Discrete Cosine Transform, and Cepstral Liftering. The type of test design that will be carried out in this research is parameter tuning. Parameter tuning is carried out with the aim of obtaining parameters that produce the best accuracy in the machine learning model. The parameters that will be tuned include the α value in the Pre-Emphasis process, frame length and overlap length in the Frame Blocking process, the number of mel filters in the Mel-Scaled Filterbank process, the number of cepstral coefficients in the Discrete Cosine Transform process and the C value in SVM. The best accuracy in males of 85.71% was obtained with a combination of filter parameter pre-emphasize of 0.95, frame length of 0.023 ms, overlap of adjacent frames of 40%, number of mel filters in the mel-scaled filterbank process of 24 mel, number of cepstral coefficient of 24 coefficient and the value of \u27C\u27 in SVM of 0.01. The best accuracy in women of 92.21% was obtained with a combination of filter parameters pre-emphasize of 0.95, frame length of 0.023 ms, overlap of adjacent frames of 40%, the number of mel filters in the melscaled filterbank process of 24 mel, and the number of cepstral coefficient of 13 coefficient and \u27C\u27 value in SVM of 0.01. From the two test results of tuning parameters between men and women, there are similar parameter values in all test parameters, except for the number of cepstral coefficients. The number of cepstral coefficient in men is 24 coefficient while the number of cepstral coefficient in women is 13 coefficient. Based on the research conducted, there are the following conclusions, the combination of MFCC and SVM methods can be used for emotion classification based on input data in the form of voice intonation with an accuracy of 85.71% in men and 92.21% in women. The difference in accuracy obtained between male and female models is due to the different data used. Male models are trained with male voice data and female models are trained with female voice data, this is done because men and women have different voice frequency ranges
Performance Analysis of FastAPI Framework on Lost Circulation Handling Management Application in Oil Well Drilling
Full text linkPurpose: This study aims to conduct a load testing using JMeter and then analyze the performance of the FastAPI framework on the backend of the lost circulation handling management application in oil well drilling. The developed API receives input in the form of drilling parameter data (daily drilling report) from drilling engineers to be processed by a machine learning model (prediction and classification) through the FastAPI framework. The developed API returns processing data in JSON format.Methodology: Performance measurement is done by conducting load testing simulations using the help of JMeter software. Load testing scenarios are created by varying the number of users and ramp-up time, as well as the method of loading the machine learning model used (normal or preemptive loading). The parameters measured in the test scenario are average execution time, maximum execution time, error percentage, and request throughput.Findings: Load testing on a FastAPI-developed API demonstrated that for compute-heavy tasks like machine learning inference, increasing the number of processor cores and using preemptive model loading led to significantly better performance improvements than changes in processor clock speed or switching from HDD to SSD. Even when simulating a higher user load than initially expected (up to 250 users/threads), FastAPI maintained good response times and a low error rate, remaining below 20%.Originality/value/state of the art: This study result is an information about the performance of the FastAPI framework in the application of lost circulation handling management in oil well drilling in the deployment phase, not only up to the model testing phase as in previous studies.
Application of the Technological Acceptance Model(TAM) Approach to the Influence of Public Perceptions Using Digital Wallets
Full text linkPurpose: This research aims to conduct a study of the perceptions of digital wallet users and how users\u27 reactions influence the benefits of digital wallets in the big city of JakartaDesign/methodology/approach: The research method applied is a quantitative method. The population of this research is digital wallet users in the city of Jakarta. The number of samples applied was 121 respondents according to the purposeful sampling method. The data testing methods applied are convergent validity, discriminant validity, composite reliability and Cronbanch alpha. Data calculations apply Smart PLS 3 software. The results of this study show that trust and perceived risk do not influence user preferences for using digital walletsFindings/result: The results of this research constantly support a number of previous studies related to TAM where perceived usefulness and perceived ease of use play a direct and indirect role in interest in using digital wallets. So the community\u27s perceived usefulness is a variable that has a prominent influence on the preferences of digital wallet users in the city of Jakarta.Originality/value/state of the art: The steps taken from the start of the study to its conclusion were designed to use the TAM approach to determine how the public felt about this particular study. This study uses quantitative research methods and yields two models: an inner model, or structural model, that includes path analysis through Smart PLS 3 data analysis, and an outer model, or measurement model, that includes composite reliability, conbranch alpha, discriminant validity, and convergent validity.
Maturity Level Analysis of Furniture MSME Business Processes in Yogyakarta
Full text linkAbstrakTujuan: Penelitian ini bertujuan untuk mendeskripsikan tingkat kematangan proses bisnis UMKM Furnitur di Yogyakarta berdasarkan pengukuran BPOMM.Desain/metodologi/pendekatan: Penelitian ini mengumpulkan data melalui observasi dan wawancara dengan pemilik UMKM mebel di Kota Yogyakarta. Hasil pengumpulan data kemudian dianalisis menjadi elemen pengukuran BPOMM.Temuan/Hasil: Tingkat kematangan proses bisnis UMKM Furnitur di Yogyakarta dan saran perbaikan proses bisnis tersebut.Orisinalitas/nilai/keadaan terkini: Penelitian ini menunjukkan tingkat kematangan proses bisnis UMKM Furnitur di Kota Yogyakarta berdasarkan pengukuran BPOMM beserta saran perbaikan proses bisnisnya
Rubber Leaf Image Classification Using Artificial Intelligence Methods as an Effort to Improve Plantation Production Results
Full text linkPurpose: Rubber is one of the plantation commodities that contributes positively to the trade surplus in the agricultural sector. Seeing the positive trend in global rubber consumption and production, demand is expected to continue increasing in the future. To enhance rubber productivity, rubber processing technology can be used to make it more efficient, thus increasing the amount of latex extracted from the sap and reducing waste materialDesign/methodology/approach: One technology that can be developed to increase the productivity efficiency of rubber plants is by using Artificial Intelligence. This technology is expected to be implemented in the rubber plantation sector, specifically in the automatic recognition of rubber leaves.Findings/result: The measurement and performance analysis of the rubber leaf image classification algorithm based on Artificial Intelligence has also been evaluated, showing near-perfect accuracy on training data (99.86%) and very good performance on validation data (97.43%), with a very low validation loss (0.0873), indicating that the model has learned well by the last epochOriginality/value/state of the art: The population in this study consists of image data from various tree leaves, including 10 types of rubber leaves and non-rubber leaves