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 1,182 Documents
Hate Speech Detection in Indonesia Twitter Comments Using Convolutional Neural Network (CNN) and FastText Word Embedding Fadhilah Nadia Puteri; Yuliant Sibaroni; Fitriyani F.
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6401

Abstract

Hate speech is a problem that is often present in Indonesia, including on social media platforms such as Twitter. Refers to any form of communication, whether oral, written, or symbolic, that may offend, threaten or insult an individual or group based on attributes such as religion, race, ethnicity, sexual orientation, or other characteristics. The existence of freedom of expression and communication on social media triggers the spread of hate speech quickly and widely. To avoid this, a system is needed that can detect hate speech on social media. Deep learning is potentially better at recognizing and analyzing language patterns that reflect hate speech in text. In the previous study, the accuracy obtained was 73.2% using the Convolutional Neural Network method. This study proposed a hate speech detection system using Convolutional Neural Network model and FastText word embedding. The performance of Convolutional Neural Network classification model and FastText as word embedding provide excellent performance results in detecting hate speech, by involving the K-Fold Cross Validation process to the appropriate dropout value is able to achieve an accuracy value of 80%. The resulting accuracy value can be a benchmark that the model that has been built is able to avoid the spread of hate speech on social media.
Analisis Sentimen Terhadap Pemindahan Ibu Kota Negara Menggunakan Algoritma Naive Bayes Classifier dan K-Nearest Neightbors Dedi Pramana; M Afdal; Mustakim Mustakim; Inggih Permana
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6523

Abstract

The relocation of Indonesia's capital city is a hot topic of discussion at the moment. So that this government policy reaps a lot of reactions from various parties, especially the general public in Indonesia. Various reactions were shown with various expressions on various social media. One of the social media that has become a place for people to express themselves in responding to this government policy is Instagram. The comments poured by the community on posts on Instagram are very diverse ranging from positive, negative, and neutral comments. If these comments are processed properly, they can be used as evaluation material for the relocation of the State capital. Seeing this, a sentiment analysis is needed which is intended to classify the various comments so that they can be presented into information which will be intended to help the government make considerations in carrying out policies towards moving the national capital. In this study, data processing was carried out with the Naive Bayes Classifier and K-Nearest Neightbors algorithms with Instagram comment data on posts related to moving the national capital. Where the amount of data used is 2,404 comments. It was found that the accuracy of the NBC algorithm was 63.09% and K-Nearest Neightbors was 69.23% so it can be concluded that KNN is better than NBC. In addition, the popularity of public sentiment towards the relocation of the National Capital was also obtained with a positive sentiment of 28% totaling 643 comments, a neutral sentiment of 42% totaling 1025 comments, and a negative sentiment of 30% totaling 730 comments.
Penerapan Algoritma Association Rules Dalam Penentuan Pola Pembelian Berdasarkan Hasil Clustering Sania Fitri Octavia; Mustakim Mustakim; Inggih Permana; Siti Monalisa
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6129

Abstract

Zanafa Bookstore is one of the bookstores in Pekanbaru city that is required to meet customer needs and has the right focus in developing sales strategies every day. During the new school year there is an increase in sales, it is known that in July there are the most purchase transactions which are the beginning of the new school year for students and students. In addition, the placement of the book layout is only based on the employee's estimated shelf so that it will affect the convenience of consumers in choosing and finding books if the books are arranged far apart. By placing the layout in accordance with consumer purchasing patterns, it can improve the quality of customer service in bookstores. The book layout can also be used as a reference when adding book stock, information is needed by utilizing transaction data using data mining, namely by using Association rules commonly called Market Basket Analysis. This research uses K-Medoid for clustering on Apriori and FP-Growth in generating rule patterns on large-scale data. Several experiments were conducted on K-Medoid starting from cluster 2 to cluster 7, each of which will be applied to Apriori and FP-Growth with 30% support and 70% confidence. By comparing the evaluation results of each algorithm with each other, it is known that FP-Growth has superior results to Apriori with a total strength of rules of 1.2012. So that the results of the association rules obtained can be used as a reference in the placement of book layouts in the Zanafa bookstore.
Sentiment Analysis using Random Forest and Word2Vec for Indonesian Language Movie Reviews Fahriza Ichsani Rafif; Mahendra Dwifebri Purbolaksono; Widi Astuti
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6299

Abstract

The film industry in recent years has become one of the industries that people are most interested in. The convenience of watching movies through streaming services is one of the reasons why watching movies is so popular. This ease of access resulted in a large selection of available movies and encouraged the public to look for movie reviews to find out whether the movies was good or bad. Freedom of expression on the internet has resulted in many movie reviews being spread. Therefore, sentiment analysis was conducted to see the positive or negative of these reviews. The method used in this research is Random Forest and Word2Vec skip-gram as feature extraction. The Random Forest classification was chosen because Randomforest is a highly flexible and highly accurate method, while Word2Vec Skip-Gram is used as a feature extraction because it is an efficient model that studies a large number of word vectors in an irregular text. The best model obtained from this experiment is a model built with stemming, Word2Vec with 300 dimensions, and a max_depth value of 23, achieving an f1-score of 83.59%.
Sentiment Classification of Fuel Price Increase With Gated Recurrent Unit (GRU) and FastText Aditya Andar Rahim; Yuliant Sibaroni; Sri Suryani Prasetiyowati
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6391

Abstract

The government usually implements a policy of increasing fuel prices and reducing subsidized fuel every year. Rising fuel prices have had a mixed impact on society. The rapid development of information technology has led to easy access and an increase in the number of internet users. Social media platforms, such as Twitter, are widely used by people to express themselves in everyday life. Through this social media, the public can submit reviews regarding public policies implemented by the government regarding fuel prices. The reviews submitted varied, ranging from positive, neutral to negative. Sentiment analysis can analyze the types of reviews submitted by people, including positive, negative, or neutral. This research uses Gated Recurrent Unit and FastText feature expansion to classify sentiments related to rising fuel prices on Twitter. This system was developed through several stages, namely data crawling, data labeling, data initial processing, feature expansion, classification, and evaluation. This study aims to determine the classification performance using Gated Recurrent Unit and FastText. The data used was 8,635, and the highest accuracy reached 90.15% with an F1 score of 90.06%. The research results may help the government in determining how individuals feel about fuel price increases. By understanding public sentiment, the government can reevaluate its policies or even establish new ones that serve the public interest.
Handling Unbalanced Data Sets Using DBMUTE and NearMiss Methods to Improve Classification Performance of Yeast Data Sets Bima Mahardika Wirawan; Mahendra Dwifebri Purbolaksono; Fhira Nhita
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6306

Abstract

Yeast vacuole biogenesis was chosen as a model system for organelle assembly because most vacuole functions can be used for vegetative cell growth. Therefore it is possible to generate an extensive collection of mutants with defects in unbalanced vacuole assembly. With this in mind, we must find the structural balance of data in yeast. Imbalanced data is when there is an unbalanced distribution of data classes and the number of data classes is either more or lower than the number of other data classes. Our method uses the f1score performance matrix method and the balanced accuracy on DBMUTE and NearMiss undersampling. Previously, only a few studies explained the results of using a performance matrix and balanced accuracy. Then, find out the performance results of the f1 score and balanced accuracy and get the best score from the yeast datasets. In the study, a comparison between the imbalanced datasets using the undersampling method. Furthermore, to obtain the performance matrix results, use the f1 score and balance accuracy. After testing five yeast datasets, we performed an average f1 score and balance accuracy with the highest average NearMiss f1 score of 62.23% and the highest average balanced accuracy of 78.59%.
Music Recommender System Based on Play Count Using Singular Value Decomposition++ Muhamad Elang Ramadhan; Agung Toto Wibowo
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6424

Abstract

The availability of digital music content on various music streaming services, which is constantly growing, has increased the need for recommender systems (RS) to assist users in finding music that suits their taste without the need of searching manually. One of the commonly used paradigms is Collaborative Filtering (CF). In CF, the input used to predict ratings can take the form of explicit or implicit input from user feedback. In the music domain, implicit feedback such as the number of music plays can be utilized to predict a user's music preferences. Singular Value Decomposition++ is one of the Matrix Factorization (MF) algorithms that can leverage implicit feedback and address the sparsity issue. In this research, a music recommender system is built using the Million Song Dataset (MSD) Subset from The Echo Nest, utilizing SVD++ algorithm. Additionally, the performance of the built system is measured through k-fold cross-validation using the evaluation metrics RMSE and NDCG. The performance measurement results using RMSE and NDCG in 5-fold cross-validation yield an RMSE of 0.4423, NDCG@5 of 0.8232, and NDCG@10 of 0.8231 for the top 10 items.
Rancang Bangun Keamanan Kotak Amal dengan Akses Fingerprint Menggunakan ESP32-Cam dan Telegram Berbasis IOT Diky Hermawan; Jufrizel Jufrizel; Aulia Ullah; Ahmad Faizal
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6252

Abstract

The charity box is one of the objects that is often the center of attention by thieves, the security of the charity box in current technological developments can be further developed, the creation of a charity box security tool with Fingerprint access using an IOT-based ESP32-cam is one of the developments that aims to improve security in charity box and reduce the rate of theft that occurs. The tool is made using Arduino UNO as the main microcontroller that processes data in this design and ESP32-cam as the recipient of commands from Arduino Uno and Telegram after which it sends data in the form of notifications in the form of photos and sentences describing the images that have been sent. Using 2 sensor inputs, namely the Fingerprint sensor to open the lid of the charity box door, if the fingerprint is registered, it sends a command to the Solenoid to open the charity box door, whereas when the fingerprint is not registered, the Solenoid will not open after that, ESP32-cam gets an order to capture the photo. sent to the telegram application and the Limit Switch sensor functions to detect the state of the charity box door when there is a forced opening that occurs at the door it will send data to Arduino UNO that there is a threat then the buzzer will sound. After testing the results obtained have a success rate of 100%, the ESP32-cam can receive orders to take pictures properly if one of the sensors used is active and can send data in the form of photos to the telegram application properly then the sensor used also works well well, the buzzer will sound when a forced opening is detected by the Limit Switch sensor. In this research it is hoped that it can improve the security of the charity box by capturing photos through the applied ESP32-Cam.
Implementation of Collaborative Filtering Algorithms in Mobile-Based Food Menu Ordering and Recommendation Systems Nurini Siregar; Samsudin Samsudin
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6387

Abstract

In the business world, the application of technology is becoming common, including in the process of buying or ordering food products which can now be done through a mobile application. Makecents Coffee is a startup in the city of Medan that provides solutions for ordering food and drinks at Android-based restaurants using the QR Code ordering system. To make it easier for buyers to place orders, an automatic recommendation system is needed. One method that can be used to develop an ordering application with a recommendation system is a collaborative filtering algorithm. In this study, a collaborative filtering algorithm was used to work by storing and processing data provided by buyers, such as ratings or comments on the food menu ordered. Using buyer data provides results for users in placing orders because they use an application that has them, as well as making it easier to choose a menu to order because of a recommendation system. The level of accuracy of the prediction of the collaborative filtering algorithm itself has been tested using the MAE and RMSE tests. Where the MAE test obtained a value of 0.67 points, while the RMSE test obtained a value of 0.58 points. The two test results were fairly good when compared to the range of points which only ranged from 1 to 5 points. The results of the recommendations can be implemented in applications designed to increase sales and make it easier to place orders that have been recommended to users.
Analisa Market Basket Analysis untuk Melihat Pola Transaksi Customer Menggunakan Algoritma Apriori dan FP-Growth Griya Jitri Pabutungan; Hindriyanto Dwi Purnomo
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 7, No 3 (2023): Juli 2023
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v7i3.6152

Abstract

Online sales are considered an alternative approach that can positively impact product marketing. But the more online shops, the greater the competition in the business world. From the problems above, so that the store can survive among busy competitors, a strategy that is qualified as an effort to attract customer attention is needed. One effort that can be done is to look at customer patterns or tendencies during transactions. Knowing this is expected to provide additional information to stores to increase loyalty and meet customer needs so that online business stores that are built can last a long time. Based on the description that has been explained, the purpose of writing this research is to find customer spending patterns so that it can support XYZ company as one of the retail application managers in creating business strategies that can be used to attract customer interest using the market basket analysis method and the help of the a priori algorithm and fp-growth as a comparison. The market basket looks at product categories often purchased together in one receipt by customers. This study uses transaction data of 34,159,477, a minimum confidence value of 0.2, and a minimum support of 0.01 for both algorithms. The results of the two algorithms using the same minimum value give the same result in the form of 10 association rules itemset.

Page 88 of 119 | Total Record : 1182