cover
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
+6282161108110
Journal Mail Official
mib.stmikbd@gmail.com
Editorial Address
Jalan sisingamangaraja No 338 Medan, Indonesia
Location
Kota medan,
Sumatera utara
INDONESIA
JURNAL MEDIA INFORMATIKA BUDIDARMA
ISSN : 26145278     EISSN : 25488368     DOI : http://dx.doi.org/10.30865/mib.v3i1.1060
Decission Support System, Expert System, Informatics tecnique, Information System, Cryptography, Networking, Security, Computer Science, Image Processing, Artificial Inteligence, Steganography etc (related to informatics and computer science)
Articles 54 Documents
Search results for , issue "Vol 7, No 4 (2023): Oktober 2023" : 54 Documents clear
Performa Metode Convolutional Neural Network Pada Face Landmark Untuk Virtual Make Up Try On Dameethia Angeline; Erico Jochsen; Dyah Erny Herwindiati; Janson Hendryli
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6619

Abstract

Make up or facial makeup, is an activity to change the appearance from its original form with the help of make up materials and tools. Make-up tools are beauty tools that are commonly used by most women to beautify the appearance of their faces with many shade choices. The shade on the make-up tool is the color usually used in make-up. Examples of make-up tools that are most often used include eyeshadow, blush on, and lipstick. These make-up tools are sold widely online and offline in physical stores. However, usually a tester is also needed so that those who want to buy can try the shade that suits them. When buying online, they often find it difficult to choose the right shade, while testers in physical stores are sometimes considered less hygienic because they have been used by many people. The aim of this paper is to measure the performance of the Convolutional Neural Network (CNN) method using the ResNet-50 architecture on facial landmarks for creating virtual make up try ons which can be an alternative to this problem. The facial image data source used is from the Kaggle site called Facial Keypoints Detection. The testing process produces 78.99% accuracy while the training process produces 95.12% accuracy. The evaluation results of this model use Root Mean Squared Error (RMSE) of 2.2577 and Mean Absolute Error (MAE) of 1.5389.
Sistem Optimalisasi Pengadaan Alat Kesehatan Menggunakan Metode Fuzzy Time Series Febrina Sari; Soni Fajar Mahmud; Rudi Faisal
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6405

Abstract

Controlling the procurement of Medical Devices is an important matter for the Pharmacy industry to pay attention to in order to win highly competitive competition, therefore an appropriate model and strategy is needed so that the number of sales can increase, the right solution is to maintain optimal stock availability. this is in line with the Regulation of the Minister of Health of the Republic of Indonesia Number 35 of 2014, concerning pharmaceutical service standards at pharmacies for pharmaceutical preparations, medical devices and consumables which includes procurement and control. The purpose of this study is to assist pharmacies in optimizing the procurement of medical devices by applying the fuzzy time series Chen model, so that they can overcome stock emptiness and over stock, besides that the pharmacy has a system that can predict the optimal number of medical device product purchases for the next period. which has an impact on the ability to control stock. The results showed that the fuzzy time series method of the Chen model has very good performance. This can be seen from the value of the accuracy of the forecasting results which is calculated using the AFER (Average Forecasting Error Rate) formula with a value of 4%. The number of medical devices that will be provided for the January 2023 period is 15 pieces.
Penerapan Klasifikasi Algoritma C4.5 Dan Algoritma C5.0 Untuk Mengetahui Tingkat Kepuasan Mahasiswa Terhadap Website Sistem Informasi Terpadu Layanan Program Studi (SIPLO) Nurfitrayani Nurfitrayani; Islamiyah Islamiyah; Amin Padmo Azam Masa
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6433

Abstract

Integrated Information System for Study Program Services (SIPLO) is a website-based information system for academic services at the study program level, specifically designed for the Information Systems Study Program at Mulawarman University. Despite containing information that supports lectures, SIPLO's features and information have not met students' satisfaction, as indicated by data collected through interviews. Therefore, the objective of this study was to determine the level of student satisfaction with the SIPLO website.This study employed a data mining technique using the classification methods of the C4.5 algorithm and the C5.0 algorithm. The PIECES indicators, which include performance, information, economy, control, efficiency, and service, were used as attributes in the data mining application. The data utilized in the study consisted of questionnaires distributed to 182 students from the Information Systems Study Program at Mulawarman University in 2019, 2020, and 2021. The data was divided into a 80% training data and a 20% test data ratio. The research findings using the C4.5 algorithm revealed that the variables influencing student satisfaction are performance, control, information, efficiency, and service. Meanwhile, the C5.0 algorithm identified control, performance, efficiency, information, and service as the influential variables. Both algorithms yielded an accuracy value of 91.89%, precision value of 93.75%, recall value of 96.77%, F1-Score value of 95.24%, and an AUC value of 0.8172. These results indicate a good classification performance. 
Analisis Data Mining dalam Komparasi Average Linkage AHC dan K-Means Clustering untuk Dataset Facebook Live Sellers Jhiro Faran; Rima Tamara Aldisa
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6892

Abstract

Facebook Live is a social media platform owned by Facebook that allows users to broadcast videos directly or live stream via the internet. Users can share moments in real-time with friends, followers, or members of certain groups. The platform allows anyone with a Facebook account to create live video broadcasts from a mobile device or computer equipped with a webcam. Many Micro, Small and Medium Enterprises (MSMEs) use Facebook Live as a tool to sell products or services directly to their audience. This strategy is increasingly popular in direct marketing on social media, especially in countries such as China and Thailand. Sellers on Facebook Live, known as Facebook Live Sellers, broadcast live on the platform to introduce products or services. They explain all the features offered, answer questions from viewers, and encourage them to make a purchase immediately. To increase buyer interest, they often offer special offers or discounts. Facebook Live Sellers can also be considered a form of influencer marketing, where individuals or businesses build a loyal following and use their influence to promote products and services. Despite the potential benefits, Facebook Live Sellers also face challenges. They interact directly with potential buyers, who may sometimes be dissatisfied with the product offered or the way the seller promotes it. Therefore, evaluations such as comments, reactions (such as like, unlike, angry), and other interactions during broadcasts are important. This research aims to group potential buyers' reactions during Facebook Live broadcasts as a strategy to overcome several problems in direct sales via this platform. In addition, grouping by the number of likes and comments can help sellers identify the most active groups of buyers and have the potential to become loyal customers. The number of data samples was determined using the Solvin method so that the dataset that became the data sample was 341 data. The methods used for grouping are K-Means and AHC (Average linkage) with the final results showing that the amount of data grouped into three clusters by both methods is the same, with most of the data being in Cluster 0, namely 98.5% of the total data sample. . Cluster 1 has a small amount of data, namely 0.6%, while Cluster 2 has 0.9% of the data sample.
Pengenalan Potensi Racun dan Peningkatan Keamanan Pangan Dalam Jamur Menggunakan Convolutional Neural Network Ilham Rafiedhia Pramutighna; Arief Hermawan
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6372

Abstract

One promising advancement in the field of food agriculture is the cultivation of mushrooms. Mushrooms can be broadly classified into two groups: edible mushrooms and non-edible mushrooms. Edible mushrooms serve various purposes, including as food, medicine, and other applications, while non-edible ones can lead to poisoning. However, distinguishing between edible and non-edible mushrooms is a complex task. Even a slight error in selecting suitable mushrooms for consumption can have health repercussions for consumers. The progress in science and technology, particularly in digital image processing, aids in the classification of mushrooms. Image classification using Convolutional Neural Networks (CNNs) presents an alternative to address this issue. This research primarily focuses on identifying potential toxins in mushrooms using CNNs, aiming to contribute to a more efficient and accurate approach in classifying mushrooms fit for consumption. The results demonstrate that the model trained with data augmentation achieved the highest accuracy, with 96.53% for training data and 93.22% for validation data, accompanied by lower loss rates. This underscores that CNNs are an efficient and accurate approach in classifying mushrooms based on their genus. Furthermore, this study also discovered that parameters such as the number of epochs, batch size, optimizer, image size, and image augmentation influence the model training process.
Perbandingan Matriks Loss Pada Model Deep Learning Resnet50 dan Xception dalam Deteksi Objek Herimanto Herimanto
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6849

Abstract

The implementation of deep learning has expanded into various fields, not confined solely to the field of education, particularly in computer science. It has also integrated technology into various other domains, including geospatial, remote sensing, and even the medical field. This development has made a significant contribution to reshaping the way humans understand and tackle challenges across different sectors. In this context, deep learning is employed for object detection and classification. Despite the considerable progress facilitated by the application of deep learning, object detection remains a challenge that is not entirely resolved. Constraints such as variations in lighting conditions, angles of view, and object diversity make achieving high-accuracy object detection a difficult task. Therefore, further research is required to comprehend and compare the performance of various deep learning models in addressing this issue. This research focuses on the comparison of two deep learning models, namely ResNet50 and Xception, in terms of loss metrics when detecting an object, in this case, a chair. The models are provided with input images of chairs and predict whether the chairs are empty or occupied. The results obtained from this research indicate that the ResNet50 model has a lower total loss value of 0.19422098, while the Xception model has a total loss value of 1.1822930. The lower the loss value, the better the model's performance. Based on the comparison results, the author has developed a web application simulator using Flask, utilizing the model with the lowest loss, which is the ResNet50 model.
Optimasi SVM dan Decision Tree Menggunakan SMOTE Untuk Mengklasifikasi Sentimen Masyarakat Mengenai Pinjaman Online Rismawati Nurul Ikhsani; Ferian Fauzi Abdulloh
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6809

Abstract

With the development of technology, many applications and social media make it easier for users to do various desires, one of which is borrowing money online on the Online Loan Application with easy terms. The convenience provided causes many violations committed by irresponsible people, such as the breach of important information and data of online loan application users. This causes many people to express their comments and opinions on social media, especially on twitter. Sentiment analysis is conducted to see the tendency of public opinion to fall into negative, neutral, or positive sentiment. Furthermore, public opinion will be classified using two algorithms, namely the Support Vector Machine and Decision Tree algorithms. The aim of this research is to compare the performance of SVM and Decision Tree classification algorithms on the tendency of public opinion on twitter regarding online loans. Furthermore, optimization is carried out using SMOTE to optimize the accuracy of the two algorithms. The results obtained neutral sentiment as much as 78.96%, positive sentiment as much as 14.98%, and negative sentiment as much as 6.06%, people are more inclined to neutral sentiment. Then classification using SVM gets an accuracy of 87% and on Decision Tree gets an accuracy of 89%.. Then to optimize the performance results of the two algorithms, optimization using SMOTE is carried out. After SMOTE optimization, the accuracy produced by SVM is 99% and Decision Tree is 97%. Optimization using SMOTE proves that the SVM algorithm is better than Decision Tree. 
Klasifikasi Masalah Pada Komunitas Marah-Marah di Twitter Menggunakan Long Short-Term Memory Dian Sukma Hani; Chanifah Indah Ratnasari
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6755

Abstract

The rapid development of social media goes hand in hand with the increase in social media users. Among the social media platforms widely used in Indonesia, Twitter is one of the most popular. On Twitter, users are free to share every moment they experience or what they think. Many users use Twitter as a medium to express their emotions, as is what happens in angry communities. There are no special requirements to join as a community member other than getting admin approval. This community provides a place for its members to vent all kinds of anger they feel. This research classifies angry community tweets to find out the types of problems in these tweets. The results of this research can help in understanding communication and behavior patterns in angry communities, which can provide deeper insight into the social dynamics within them.Text data is retrieved via web scraping techniques, and then processed through a series of preprocessing steps, including unnecessary character removal, normalization, and tokenization. The classification uses the Long Short-Term Memory (LSTM) algorithm with six problem category classes, namely Study, Romance, Family, Career/Work, Person/Personal, and Swearing. After modeling, the model accuracy was 91.94%. The model was built using an embedding layer, Long Short-Term Memory (LSTM) layer, dense layer, and dropout layer which was run for 10 epochs. Model evaluation is carried out using metrics such as accuracy, precision, recall, and F1-score to measure model performance. The value resulting from the evaluation results using the confusion matrix is more than 50, this indicates that the LSTM model is able to classify the problem well.
Penerapan Metode Weighted Aggregated Sum Product Assessment (WASPAS) dengan Rank Order Centroid (ROC) Dalam Rekomendasi Barbershop Terbaik Wahyuni Wahyuni; Siti Lailiyah; Reza Andrea
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6769

Abstract

Barbershop is an innovation or development of a service previously known as a barber or barbershop. Barbershop operations are led by a barber or hairstylist, also known as a hairstylist, who has special skills in shaving and creating a variety of men's hairstyles. The barbershop business is experiencing rapid growth in this era. While there may be different variations of the name or brand, this business promises very attractive prospects in the long term. With so many barbershop options available, finding the best one can be a daunting task. In choosing a barbershop, there are several criteria to consider, such as the skill of the hairstylist, price, services offered, facilities provided, and level of cleanliness. To organize data and provide recommendations regarding the best barbershop, it is necessary to use an effective information system. The term "Decision Support System" (DSS) is often used to describe these information systems. The main objective of the DSS system is to improve the decision-making process and make it more effective and efficient by providing information, analysis and data modeling. The data needed to provide the best barbershop recommendations in Samarinda City were collected using the WASPAS (Weighted Aggregated Sum Product Assessment) and ROC (Rank Order Centroid) methods in this study. The replacement for the BS4, Sir Salon Barbershop, has the highest rating of 0.9815, making it a top barbershop recommendation.
Analisis Komparatif Kinerja Algoritma Machine Learning untuk Deteksi Stunting Syahrani Lonang; Anton Yudhana; Muhammad Kunta Biddinika
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 4 (2023): Oktober 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i4.6553

Abstract

Stunting is a serious problem caused by chronic malnutrition in children under five, causing stunted growth and having a negative impact on long-term health and productivity. Therefore, early detection of stunting is very important to reduce its negative impacts. Previous studies utilizing machine learning have proven the success of this method in various health applications, such as disease detection and the prediction of medical conditions. The results of this research are a comparative evaluation of five classifications, namely Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and Support Vector Machine (SVM), in classifying stunted toddlers. The dataset used contains four important attributes: age, gender, weight, and height of toddlers, as well as a binary class label that differentiates between toddlers who are stunted and those who are not. The evaluation results show that KNN at K = 3 produces the highest accuracy of 94.85%, making it the best model for classifying stunting in toddlers. Apart from accuracy, other metrics such as precision, recall, and F1-score are used to analyze the algorithm's ability to solve this problem. KNN stands out as the best model, with the highest F1-score of 89.47%. KNN also manages to maintain a balance between precision and recall, making it an excellent choice for treating stunting in toddlers. Apart from that, the use of the AUC metric from the ROC curve also shows the superiority of KNN in differentiating between stunted and non-stunting toddlers. With a combination of consistent evaluation results, both in terms of accuracy and other evaluation metrics, this research proves that KNN is the best choice for overcoming the task of classifying stunting in toddlers.