BIOS: Jurnal Teknologi Informasi dan Rekayasa Komputer
Not a member yet
75 research outputs found
Sort by
Implementasi Manajemen Bandwidth Hierarchical Token Bucket (HTB) Menggunakan Metode Network Development Life Cycle (NDLC)
No full textSMKN 1 Klakah is one of the educational institutions in Kab. Lumajang requires an internet network to support the learning process for teachers or students. With adequate internet, it can make it easier for teachers and students to access learning materials. There are times when using the SMKN 1 Klakah Internet network causes poor Internet performance when users access the Internet simultaneously. In addition, the download and upload volume for each user is not distributed evenly, so bandwidth management is required. One method that can stabilize the distribution of bandwidth is the Hierarchical Token Bucket (HTB) method. The research was conducted based on the Network Development Life Cycle (NDLC) model with 6 stages, namely: analysis, design, simulation prototyping, implementation, monitoring and management. Resulting in research that the implementation of the HTB method for bandwidth management in the SMKN 1 Klakah Computer Network Lab was successfully implemented. It was proven that when testing the bandwidth it was in accordance with the specified limit. Also when the QoS test was carried out it was in the good category, as evidenced by the QoS test results on the delay parameters with a value of 5,6 ms, jitter with a value of 4,04 ms, throughput with an average value of 0,921 Mbit, and packet loss with a value of 0%
Implementasi Algoritma Fuzzy C-Means untuk Pengelompokkan Provinsi di Indonesia Berdasarkan Kualitas Perguruan Tinggi
No full textThe education system in Indonesia is very large and complex with low quality. The quality of Indonesian education can improve, one way is by having equal distribution of education in every province in Indonesia. This equality can be one solution to improve the quality of graduates in Indonesia. This equality can be done by grouping Indonesian provinces with low quality education. One grouping method that can be used is the Fuzzy C-Means algorithm, which is a clustering technique that is determined by the degree of membership in each data point in one cluster. The grouping process was carried out using 136 data on Higher Education Gross Enrollment Rates in 34 provinces from 2019-2022. The data was processed using the Fuzzy C-Means algorithm and then a search for optimal clusters was carried out using the help of the Partition Coefficient Index. Based on testing from 2 to 10 clusters, the optimum cluster is 2 clusters, with a Partition Coefficinet Index value of 0.83491. In the optimum cluster, we get cluster 1 with 20 provinces and cluster 2 with 14 provincial groups. Characteristics resulting from data from 2019 to 2022, cluster 1 has provincial members with the lowest higher education APK compared to cluster 2, especially in cluster 1 members, namely Kep province. Bangka Belitung which has the lowest higher education APK is 2019 with 14.27 APK, 2020 with 14.73 APK, 2021 with 15.23 APK, 2022 14.85 APK
Klasifikasi Tingkat Kecemasan Atlet Sebelum Bertanding Menggunakan Algoritma K–Nearest Neighbor (KNN) Berbasis Website
No full textAnxiety experienced by an athlete before a match often affects their performance, so it is important for the coach to know the athlete's anxiety level before competing in order to provide appropriate mental training and make decisions that will affect the outcome of the match. However, not all coaches can know the level of anxiety of athletes; therefore, it is necessary to build a web-based system to classify the anxiety level of athletes before competing. The system can be built using one of the data mining methods, namely KNN (K-Nearest Neighbour), where this method can be used to classify the anxiety level of athletes based on a dataset of 364 futsal athlete data participating in the Mechanical Futsal Competition, which will be classified into 3 anxiety categories, namely low, medium, and high, from 17 attributes. From the tests carried out on the dataset using the confusion matrix method using the ratio of testing data: 80:20 training data with K = 5, accuracy, precision, and recall values of 100% were obtained. So we successfully built a website that can be used by a coach to classify athletes based on their anxiety level
Implementasi Machine Learning Untuk Prediksi Penyakit Jantung Menggunakan Algoritma Support Vector Machine
No full textHeart disease is currently a disease that has taken over many human lives. Data shows that more than 17 million people have died from heart disease. The high number of deaths, therefore, requires special handling to treat and prevent heart disease. In the development of technology, diagnosis of heart disease can be done with the help of information technology, one of which is through machine learning. This study aims to implement machine learning through the SVM algorithm to predict heart disease. The model formed by SVM produces an evaluation value indicated by an accuracy value of 0.85, a precision of 0.93, a recall of 0.76, and an f-1 score of 0.83. This model is used as training data to predict heart disease which is then successfully used to create a system through the Streamlit library which can be easily accessed via the website
Analisis Kualitas Website e-RKAM Menggunakan Metode WebQual 4.0
Full text linkTechnological developments in the field of education have encouraged the implementation of the e-RKAM application to facilitate the preparation and reporting of School Operational Assistance funds digitally in madrasas. However, there are still some technical obstacles such as servers that often go down, errors in some menus, and delays in producing the latest report output (update), which has a significant impact on the quality of application services. This study aims to analyze the quality of the e-RKAM website using the WebQual 4.0 method, which includes three variables: usability quality, information quality, and service interaction quality, as well as its influence on user satisfaction. The results of the analysis showed that these variables together contributed 44.7% to user satisfaction, while the remaining 55.3% was influenced by other factors outside this study. The T-test showed that only the quality of information had a significant influence on user satisfaction (sig. value 0.000), while the quality of usability and the quality-of-service interaction had no significant effect (sig. values of 0.066 and 0.844, respectively). These findings highlight that the model used has not been fully comprehensive in explaining the variation in user satisfaction, so more research is needed to explore additional factors that may have an effect
Implementasi Ensemble Learning Metode XGBoost dan Random Forest untuk Prediksi Waktu Penggantian Baterai Aki
No full textIn motor vehicles, including cars, the battery plays an important role, namely as a place to store electrical energy and as an electric voltage stabilizer when the engine is turned on. In general, motorized vehicle users do not know the condition of the battery in their vehicle. Even though the use of battery batteries that are already in poor condition can interfere with vehicle performance. In battery replacement services such as after-sales service, the process of checking and replacing battery batteries takes a relatively long time. This can be caused by high service volume, lack of worker reliability, lack of responsiveness to the complexity of the inspection. This research aims to build a prediction model for battery battery replacement time quickly. To meet these needs, a Machine Learning approach can be used. Machine Learning uses historical replacement data to make predictions of replacement time. Machine Learning algorithms that can be used for prediction are XGBoost and Random Forest. This research uses ensemble learning techniques to combine the two models. Based on the evaluation results, it can be concluded that the model built with ensemble learning has better prediction results than a single model. Evaluation results with MSE on the ensemble bagging model have the lowest error values of 145,448. The MAPE, MAE, and RMSE evaluations on the ensemble boosting model have the lowest error values of 11.56 %, 43.80 and 38,760
Analisis Sentimen Program Jaminan Kesehatan Nasional Menggunakan Multiclass Support Vector Machine
No full textOptimizing the implementation on National Helath Insurance which requires the use of BPJS participant cards in various public services is one of the government policies that is widely discussed and has garnered many opinions in the community. Public opinion is expressed through social media, one of which is through Twitter. The aim of this research is to classify public opinion regarding the new regulations of the National Health Insurance Program as a form of government policy to implement Presidential Instruction Number 1 of 2022 using Twitter data. Public opinion as many as 1.179 tweets were labeled positive, negative and neutral sentiments, then TD-IDF wighting was carried out and analyzed using the multiclass SVM algorithm with the One Against All approach. The results of the analysis showed that Multiclass SVM with a linear kernel was able to classify with an accuracy level of 81% where the classification pf positive sentiment was17 (7.6%), negative sentiment was 115 (48.7%) and neutral sentiment are 104 (44.1%). This shows that public sentiment is dominated by negative sentiment or disagreement with the new regulations of the National Health Insurance Program
Evaluasi Kepuasan Pengguna SIMPUS di Puskesmas Singotrunan Banyuwangi
Full text linkSingotrunan Banyuwangi Health Center started implementing the puskesmas management information system since 2010. There are still various obstacles in implementing SIMPUS, including incomplete and concise information content, no notification forms, system report Formats that are not up to date, the system still loads frequently at service, access to the report menu can only be done after 12 noon, there is no help menu in the system. The study aimed to assess user satisfaction using EUCS by descriptive study. The object was the SIMPUS application in all Singotrunan Banyuwangi Health Centers. The research subjects were all SIMPUS users at the puskesmas with a total of 21 respondents. The data analyze was performed by calculating the score and then collecting it into a criterion value. The study found that the criteria of good value with the proportion of the content aspect was 79.8%, the Accuracy aspect was 68.5%, the Format aspect was 73.7%, the timeliness aspect was 73.9%, the ease of use aspect was 75%. This states that SIMPUS is running well and necessary to be maintained, but there were still some deficiencies so it is suggested that it is necessary to improve the system through SIMPUS development by making the information content more concise and complete, adding warnings to forms, updating the report Format on SIMPUS, improving the system for minimize loading, provide flexible hours of access to process reports and add help menus or guides
Penerapan Data Mining Klasifikasi Lahan Tanam Buah Alpukat dengan Algoritma Naïve Bayes
No full textThe avocado plant is a plant that came into Indonesia in the 18th century. It originated in Central America under the Latin name Persea Americana Mill. Avocado plants have many different varieties and the majority grow fertile in the tropics. Nevertheless, there are differences in growing needs between different types of avocado crops when planted on different crops. As in this study where in the observation of the research on the growth differences between avocado plants of type miki and shepard on the grown land of KTH Pedunung Lestari Welfare Village Purworejo Prefecture Pungging district of Mojokerto. In this case, it is necessary to determine exactly what type of avocado plants are suitable to be planted on the land of KTH Pedunung Lestari Sejahtera. The research was conducted using the Naïve Bayes algorithm method. Based on observations, interviews with sources and library studies, the most influential variables are ground height (Mdpl), Temperature (°C), Rainfall (mm/day) and Soil type. In this study, the results were obtained on the land of KTH Pedunung Lestari Sejahtera with a land height of 250 Mdpl, temperature 18°C, rainfall 25 mm/day and humus soil type, more suitable for planting avocado type miki than shepard type. Based on the calculations on miki avocado, the value of "Yes" is 0.75 and "No" is 0,25, while shepard type has a value of “Yes” of 0 and “No” of 1. The value of accuracy is 50%, Precision is 43% and Recall is 100%
Implementasi YOLOv5 untuk Deteksi Objek Mesin EDC: Evaluasi dan Analisis
Full text linkThe Electronic Data Capture (EDC) machine is essential for facilitating non-cash transactions, yet its efficient detection remains a challenge. This study explores the implementation of the You Only Look Once (YOLOv5) algorithm to enhance EDC machine detection. The objective is to improve accuracy and efficiency in detecting EDC machines in various environments, thereby enhancing transaction security and efficiency. The research methodology involved acquiring a diverse dataset from social media platforms and the internet, comprising 396 images after augmentation. Using Roboflow, the dataset was annotated and divided into training, validation, and testing sets. The YOLOv5 model was trained on Google Colab, achieving a Precision of 97.1%, Recall of 86.4%, and mean Average Precision (mAP50) of 92.0% on the validation set. The results demonstrate that YOLOv5 effectively detects EDC machines with high accuracy across different scenarios, validating its robustness in real-world applications. This research suggests that YOLOv5 can significantly improve transaction security and efficiency in retail and service industries. The implications of this research are substantial for industry stakeholders and decision-makers, offering a reliable solution to enhance transaction security and streamline non-cash payment processes. By integrating YOLOv5, businesses can optimize operational efficiency and customer service, paving the way for broader adoption of advanced computer vision technologies in commercial application