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 68 Documents
Search results for , issue "Vol 8, No 1 (2024): Januari 2024" : 68 Documents clear
Pengembangan Chatbot Kesehatan Mental Menggunakan Algoritma Long Short-Term Memory Fajarudin Zakariya; Junta Zeniarja; Sri Winarno
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7177

Abstract

Mental health has now become a crucial aspect of contemporary society, especially in Indonesia. This reflects the emotional, psychological, and social well-being of individuals, encompassing the ability to cope with stress in daily life. A comprehensive understanding of mental health has become highly important for the community to prevent the occurrence of mental health problems or disorders. The objective of this research is to design a chatbot as an information and solution hub for maintaining mental health, with the hope that the development of this chatbot can help reduce the risk of mental health-related issues. In the development process of this chatbot, the author applies the AI Project Cycle and utilizes a deep learning approach for the chatbot model. The development involves the Flask platform, and to achieve high accuracy, the model employs the Long Short-Term Memory (LSTM) architecturea type of recurrent neural network (RNN) specifically designed to handle long-term dependency issues common in complex mental health contexts. LSTM enables the model to store and access long-term contextual information, which can be highly beneficial in providing accurate solutions and understanding emotional condition changes. The trained LSTM model demonstrates an accuracy of 93%, validation accuracy of 82%, a loss of 0.3%, and validation loss of 1.6% after 200 epochs. Therefore, it can be concluded that using the LSTM algorithm for the chatbot model in this development is quite effective.
Decision Support System for Determining New Branch Locations Applying the Multi Attribute Utility Theory (MAUT) Method Achmad Fikri Sallaby; Agung Triayudi; Kelik Sussolaikah; Muhammad Zakaria Lubis
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7049

Abstract

The location of new branches that are close to community activities and have adequate facilities makes it easier for consumers to get the services and products they need. Determining the feasibility of new branch locations from several product or service producers still uses a system that is not accurate, which can cause problems in determining the location of new strategic and targeted branches. However, there are several obstacles in the selection of new branch locations, so technological assistance is needed in determining the location, product analysis, marketing management, and other matters concerning the development of the business being carried out. Technology that is considered efficient, easy, and flexible and is used by entrepreneurs, especially in determining the location of new branches using a decision support system using the MAUT method, is expected to help the location of new branches that are efficient and strategic. The decision support system is a conclusion and determination of the best using some data and computerized testing in each criterion so as to get valid results. After calculating each criterion and alternative, the best ranking is obtained in alternative A1 with a value of 0.7925 on Pertahanan Street.
Pengaruh Oversampling dan Cross Validation Pada Model Machine Learning Untuk Sentimen Analisis Kebijakan Luaran Kelulusan Mahasiswa Mufida Rahayu; Ardytha Luthfiarta; Lailatul Cahyaningrum; Alya Nurfaiza Azzahra
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7012

Abstract

The Minister of Education, Culture, Research and Technology issued a new policy on graduation standards for undergraduate and postgraduate students. This policy was delivered on August 29, 2023, on live streaming YouTube Kemendikbudristek at the Merdeka Belajar seminar episode 26: Transformation of National Standards and Higher Education Accreditation. The policy has caused various kinds of positive and negative responses in the community. Based on this problem, this research analyzes the sentiment of how the attitude and response of the community regarding this matter, so that it can be useful for the community in the future. This research uses two algorithms Nave Bayes Classifier (NBC) and K-Nearest Neighbor (KNN) with data collection done through YouTube video comments getting a total dataset of 1085 data. After that, enter the data pre-processing which is then labeled using the Lexicon-based method with the stemming Sastrawi method. Datasets are grouped into positive sentiment and negative sentiment where the labeling results show unbalanced label data. Then the oversampling method Synthetic Minority Over-sampling Technique (SMOTE) is performed so that the data can be balanced and produce good accuracy. The test results after the SMOTE technique show that the NBC algorithm has the highest accuracy compared to KNN. The accuracy results are 74%, precision 74.6%, recall 74% and f1-score 73.9%. While KNN produces an accuracy of 50.2%, precision of 75.2%, recall of 50.2%, and f1-score of 34.5%.
Algoritma Jaccard Similarity untuk Deteksi Kemiripan Judul Disertasi dengan Pendekatan Variasi Stop Word Removal Liga Mayola; M. Hafizh; Deri Marse Putra
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7109

Abstract

Choosing an unique dissertation title is a challenge. The number of dissertation titles rises as the number of students increases. The title of the dissertation must differ between students. Anticipation that can be done is to adopt a similarity algorithm to detect similarities in dissertation titles. The similarity algorithm chosen is the Jaccard Similarity Algorithm. Jaccard algorithm can be used to detect document similarities. Analysis process begins with preprocessing text. The stages of preprocessing text are case folding, tokenizing, stop word removal and stemming. In this study, variations of stop word removal were tested and the accuracy results obtained were tested after being analyzed using Jaccard Similarity. Researchers call it Stop Word Removal Version One (SWR1) and Stop Word Removal Version Two (SWR2). In SWR1 only prepositions and conjunctions are deleted. Meanwhile SWR2; what was done was the deletion of words in SWR1 plus the deletion of words that were often used in the title but did not make a significant contribution to the meaning of the title. The aim of this approach is to test the accuracy produced by Jaccard against these two stop word removal approaches. The research results show that Jaccard accuracy with SWR2 has an accuracy of 97.8% and SWR1 accuracy is 57.7%. stop word removal , is a critical stage in determining similarity and has a significant influence on the results of the Jaccard Algorithm.
Seleksi Fitur Information Gain untuk Optimasi Klasifikasi Penyakit Tuberkulosis Ardi Caesar Kurniawan; Abu Salam
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7122

Abstract

Tuberculosis (TB), caused by Mycobacterium tuberculosis, is a global health threat that spreads through the air. Factors such as gender, age, and geographical location influence its spread. Indonesia, the country with the second-highest number of TB cases globally, recorded a significant increase in TB cases from 2020 to 2022, especially in Semarang City. To minimize TBs impact, its crucial to identify the factors influencing its progression. Machine Learning techniques like feature selection (Information Gain) and classification algorithms (Random Forest) can be utilized. Feature selection helps determine which factors most influence TB by ranking attribute weights, while Random Forest is used for classification. Oversampling techniques like Synthetic Minority Oversampling Technique (SMOTE) are used to handle data imbalance and improve classification performance. The study concluded that the Random Forest classification model showed the best performance using all features or attributes from the highest to the lowest weight namely; tipe_diagnosis, jenis_fasyankes, usia, kelurahan_kecamatan, riwayat_dm, riwayat_HIV, tahun, paduan_OAT, status_pekerjaan, jenis_kelamin, tipe_TBC, riwayat_TBC, bulan and sumber_obat on the original TB disease dataset in Semarang City. The recall and accuracy rate reached 75%. This result is better than the TB classification model in Semarang City that uses the oversampling dataset with SMOTE and only uses the top 10-12 attributes, with a recall and accuracy rate of 74%. This research shows that certain techniques in Machine Learning can help understand the factors influencing TB treatment outcomes.
Prototipe Alat Pendeteksi Detak Jantung, Saturasi Oksigen, dan Suhu Tubuh Berbasis Arduino Mega Menggunakan Fuzzy Sugeno Annisa Dwita Aurum; Endah Fitriani
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7249

Abstract

Health is a highly sought-after aspect for all living beings on Earth. By health checkups can prevent health disturbances that may develop into diseases. Various health checks include heart rate, oxygen saturation, and body temperature checks. These checks can be conducted independently at home using various health monitoring devices available in the market. However, these health devices only display numerical values without providing detailed information about the results, requiring users to seek further information. With the development technological and sensors, three health parameters can be combined into a single detection device capable of storing results. This detection device is designed using an Arduino Mega microcontroller with the Fuzzy Logic Sugeno classification method. Fuzzy input is obtained from the MAX30102 sensor for heart rate and oxygen saturation, as well as the MLX90614 sensor for body temperature. The detection results are displayed based on their classifications along with recommendations and can be stored on the device using a MicroSD card and sent to a Telegram application. The device operates on batteries for increased practicality.This research an accuracy of 96.11% for the MAX30102 sensor and 98.83% for the MLX90614 sensor. The Fuzzy Sugeno method was successfully implemented in the device and met expectations, producing outputs consistent with calculations. The data storage process using the SDCard Module was successful with a delay of 6.17 seconds, allowing review previous results. The battery testing indicated that the device can operate for up to 4 hours.
Prediksi Retensi Pengguna Baru Shopee Menggunakan Machine Learning Wahyu Fajrin Mustafa; Syarif Hidayat; Dhomas Hatta Fudholi
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7074

Abstract

Shopee has evolved into one of the leading e-commerce platforms connecting sellers with consumers. However, the challenge of keeping users active and engaged on the platform has become increasingly complex. User retention, the ability of a platform to sustain and enhance user presence, is a key factor in the long-term success of an e-commerce platform. Understanding the factors influencing users' decisions to remain active or cease interactions with the platform involves analyzing various variables, including user behavior, preferences, shopping experiences, and interactions with the platform. This research is designed to develop an effective user retention prediction model using data from new Shopee users. By analyzing the data and applying machine learning techniques using Logistic Regression, Decision Tree, Gaussian Naive Bayes, Random Forest, KNN (K-Nearest Neighbors), MLP (Multi-Layer Perceptron), AdaBoost, and XGBoost methods, this study predicts user retention within a 14-day period after registration on Shopee. The results of this research indicate that the Random Forest model performs the best with an Accuracy value of 0.733677, Precision of 0.702161, Recall of 0.811626, and F1-Score of 0.752936. Cross-validation values demonstrate the model's consistency with an Accuracy of 0.727626, Precision of 0.698143, Recall of 0.801884, and F1-Score of 0.746328. The Random Forest model becomes a model with a high recall value, indicating good sensitivity in identifying users who retain. Consequently, the results of this research provide valuable insights for Shopee in developing retention strategies for new users, which is an important aspect in the growth and sustainability of the e-commerce business.
Sentimen Analisis Terhadap Fitur Tiktok Shop Menggunakan Nave Bayes dan K-Nearest Neighbor Sandi Saputra Hasibuan; Angraini Angraini; Eki Saputra; Megawati Megawati
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7238

Abstract

One of the features introduced by the TikTok platform is TikTok Shop, where sellers and buyers can interact easily through the live broadcast feature and promote products within the TikTok application without having to switch to other applications, as well as offering products at very cheap prices. However, this causes dissatisfaction from small traders who find it difficult to compete. Opinions about the TikTok Shop feature have created various responses from the public. This research aims to analyze public sentiment towards the TikTok Shop feature using Twitter as a data source. Applying the Naive Bayes Classifier (NBC) and K-Nearest Neighbor (KNN) algorithms and dividing data using 10-Fold Cross Validation. Labeling was carried out using the clustering method using the K-Means algorithm, divided into three categories positive, negative and neutral. The addition of qualitative data analysis techniques with thematic analysis method in this research aims to find patterns or themes in the data. The labeling results show that 75% of the total data expresses negative sentiment towards the TikTok Shop feature. And the results of the thematic analysis found that the main theme, namely "Inappropriate regulations", covers 32% of the data, with a total of 754. This research concludes that the KNN method is superior to NBC, with better accuracy, precision and recall. This is in line with previous research that KNN is superior to nave Bayes, but other research shows the opposite where nave Bayes is superior to KNN. Further research can be carried out to improve the performance of these two algorithms in sentiment analysis, for example by using more sophisticated preprocessing methods, more representative feature extraction, or more efficient optimization techniques.
Analisa Algoritma K-Means Untuk Menentukan Strategi Marketing Bias Yulisa Geni; Okto Kurnia; Nova Hayati; Muhammad Thoriq; Kiki Hariani Manurung
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7085

Abstract

This research explored the application of the K-Means algorithm in the marketing field to increase the effectiveness of ABC Cosmetic Store marketing strategies. Sales data can be processed using data mining to be used as decision making at ABC Cosmetic Stores. One of the techniques in data mining is Clustering, which is used to categorize data. The K Means algorithm is used to identify hidden patterns in the data. By using bodylotion sales data, this research aims to classify consumers into several groups. The data group in question is sales data that is of great interest to consumers and data that is of little interest to consumers. The results of clustering using k=2 show that cluster 1 consists of 5 products with product transactions sold being 1295 products. In this case, it shows that cluster 1 is a group of product data whose quantity sold has increased and can provide profits at the ABC Cosmetic Store. Meanwhile, cluster 2 has 1 data with 214 product transactions sold and is grouped as data with products that are less popular with consumers so there is no need to increase the stock available in the warehouse by ABC Cosmetic Shop. The results of this research show that K Means-based customer segmentation can increase personalization in marketing communications and increase the efficiency of marketing resource allocation. This study provides new insights into how data mining techniques can be involved in marketing strategies to determine product availability for the future.
Comparative Analysis of ARIMA and LSTM Models for Predicting Physical Fatigue in Bandung Workers Kiki Dwi Prasetyo; Rifki Wijaya; Gia Septiana Wulandari
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 1 (2024): Januari 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i1.7282

Abstract

In today's era of rapid economic growth, there is an increasing demand for workers to increase productivity by working longer and harder. However, these demands often lead to irregular and excessive working hours, which can potentially lead to negative consequences, such as physical fatigue-a state in which the body feels tired after physical activity. Factors that influence this fatigue include age, gender, health conditions, workload and work environment. Physical fatigue poses a significant challenge in ensuring workplace safety, especially in the transportation and industrial sectors, as it can reduce overall performance, productivity and quality of work. In addition, physical fatigue also increases the likelihood of decision-making errors and workplace accidents. Predicting physical fatigue is crucial to addressing these challenges. Heart rate serves as a parameter to measure fatigue, given its proven efficacy as a marker to predict physical fatigue, which is derived from the electrocardiogram and regulated by the autonomic nervous system. This research utilizes two machine learning algorithms - ARIMA and LSTM - with heart rate (bpm) and number of steps as variables. Performance evaluation, using Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), showed that the LSTM model outperformed the ARIMA model. The LSTM model showed better performance, with MSE of 0.1108 and RMSE of 0.3329, compared to the ARIMA model with MSE of 0.2397 and RMSE of 0.4895.