Article Per Year (5 Year)
p-Index From 2020 - 2025
4.781
P-Index
This Author published in this journals
All Journal JUSIFO : Jurnal Sistem Informasi Sistemasi: Jurnal Sistem Informasi JOURNAL OF INFORMATICS AND TELECOMMUNICATION ENGINEERING Jurnal CoreIT JURNAL MEDIA INFORMATIKA BUDIDARMA INOVTEK Polbeng - Seri Informatika Jurnal Sistem Cerdas Building of Informatics, Technology and Science Malcom: Indonesian Journal of Machine Learning and Computer Science The Indonesian Journal of Computer Science
Rice Novita
Universitas Islam Negeri Sultan Syarif Kasim Ria
Automotive Engineering Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 38 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 38 Documents
Search

 1   2   3   4 
Penerapan Machine Learning Pada Analisis Sentimen Twitter Sebelum dan Sesudah Debat Calon Presiden dan Wakil Presiden Tahun 2024 Dwinnie, Zairy Cindy; Novita, Rice
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 2 (2024): April 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i2.7504

Abstract

The 2024 Presidential Election has become the hottest topic in the past two years. The KPU has confirmed that there are 3 candidates for President and Vice President. For this reason, as a momentum for voters to assess the 2024 Presidential and Vice Presidential candidates, the KPU is holding the 2024 Presidential Choice Debate which is based on Law Number 7 of 2017 concerning General Elections. Based on the information presented on the kpu.go.id page, the debate will be held 5 times with 3 presidential candidate debates and 2 vice presidential candidate debates. For this reason, it is necessary to carry out an analysis to find out how public sentiment is positive, negative, and neutral on Twitter towards the three candidates for President and Vice President in 2024 before and after the debate was held. The aim is to estimate public support or disapproval of the three candidate pairs. This research uses three algorithms as a comparison of classification accuracy, namely the Support Vector Machine algorithm, Random Forest, and Logistic Regression. Where the data used is tweet data on Twitter related to before and after the debate as many as 30 datasets with a total of 9000 data. From the classification results, the average accuracy obtained for the three algorithms, namely SVM and Random Forest, was 78%, and Logistic Regression was 79%. The highest polarity obtained from the classification of the three algorithms is in the positive class. This indicates that the Logistic Regression algorithm provides better performance in classifying Twitter sentiment regarding the 2024 presidential and vice presidential candidate pairs.
Penerapan Algoritma K-Means Menggunakan Model LRFM Dalam Klasterisasi Nilai Hidup Pelanggan Afifah, Tiara Afrah; Novita, Rice; Ahsyar, Tengku Khairil; Zarnelly, Zarnelly
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 2 (2024): April 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i2.7605

Abstract

In implementing customer relationship management, there are still many companies that have not utilized CRM optimally as part of their business strategy. As is the case with UD Sandeni. UD Sandeni still has problems in managing its relationships with customers because UD Sandeni does not fully understand the difference between customer information that is profitable and unprofitable for the company's sustainability. UD Sandeni has used a system to manage customer transaction data. However, this system is only used to calculate profits and create bookkeeping for registered agents so that UD Sandeni does not have an in-depth understanding of the characteristics of its customers. To overcome this problem, the solution that can be applied is to use customer grouping techniques, such as clustering. Customer transaction data is processed using a clustering process with K-Means and LRFM. Test the validity of cluster results using DBI and calculate CLV values using AHP weights to produce cluster rankings. The results of this research obtained customer clustering which consists of 2 segments, namely cluster 1 which has the highest CLV value of 0.3171156 with a total of 298 customers and includes the High Value Loyal Customers segmentation, and cluster 2 with a CLV value of 0.1434054 with a total of 72 customers. which is included in the segmentation of uncertain new customers (uncertain lost customers).
Implementasi Algoritma Support Vector Machine Untuk Analisa Sentimen Data Ulasan Aplikasi Pinjaman Online di Google Play Store: Implementation of Support Vector Machine Algorithm for Sentiment Analysis of Online Loan Application Review Data on Google Play Store Iqbal, Muhammad; Afdal, M; Novita, Rice
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 4 No. 4 (2024): MALCOM October 2024
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v4i4.1435

Abstract

Pinjaman online (pinjol) banyak menuai pro dan kontra karena aksesnya yang mudah dan iklannya tersebar di media sosial. Penyelenggara pinjaman daring juga seringkali menggunakan metode penagihan yang mengganggu, memberlakukan bunga yang tinggi, dan menetapkan jangka waktu pembayaran yang pendek, terutama pada pinjaman daring ilegal. Karenanya, penelitian ini melakukan analisis sentimen pada lima aplikasi pinjol, yaitu Kredivo, Easycash, Rupiah Cepat, Kredit Pintar, dan Ada Pundi. Data ulasan aplikasi diambil dari Google Play Store menggunakan teknik scraping. Kemudian, pelabelan sentimen dilakukan secara otomatis menggunakan kamus sentimen Bahasa Indonesia (Inset). Hasil pelabelan menunjukkan bahwa semua aplikasi pinjol mayoritas memiliki sentimen negatif. Kredivo menjadi aplikasi dengan jumlah sentimen positif terbanyak (46%), sementara itu Easycash memiliki sentimen negatif terbanyak (65%). Data yang di labeli kemudian digunakan untuk pemodelan klasifikasi dengan algoritma Support Vector Machine (SVM). Hasil evaluasi menghasilkan algoritma SVM mempunyai kinerja yang cukup baik dengan rata-rata akurasi sebesar 72%, presisi 76%, dan recall 85%. Namun secara khusus, SVM sangat baik melakukan klasifikasi sentimen pada aplikasi Kredit Pintar dengan akurasi sebesar 83%. Analisis visualisasi menggunakan word cloud juga dilakukan untuk memahami konteks ulasan pengguna aplikasi pinjol. Hasil pengamatan menunjukkan bahwa pengguna hampir selalu membahas tentang limit pinjaman disetiap sentimen pada kelima aplikasi.
Analisis Sentimen Traveloka Berdasarkan Ulasan Google Play Store Menggunakan Algoritma Support Vector Machine dan Random Forest Rohimah, Siti; Afdal, M; Mustakim, Mustakim; Novita, Rice
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6300

Abstract

The internet has become a key element in supporting technological and information advances in various sectors of human activity. In the trade and tourism sector, the Traveloka application is the favorite choice of Indonesian people. Reviews or reviews from users play an important role for the Company to understand the level of customer satisfaction. However, currently there are several users who give high ratings but contain negative reviews. Based on these problems, this research aims to understand more deeply user opinions, so that they can be used to improve services and features as well as test and compare the accuracy of the two algorithms in classifying user sentiment. In this research, the Support Vector Machine and Random Forest classification methods were used. The research results show that Random Forest has superior and stable performance compared to SVM, with higher average accuracy for most features, such as Traveloka (71% & 67%) and Airplanes (75% & 74%). Evaluation with k-fold cross validation supports these results, with higher average Random Forest accuracy on features such as Traveloka (70% & 66%) and Airplanes (75% & 74%).
Implementasi Metode Holt-Winters dan FP-Growth dalam Melakukan Peramalan Stok Barang Pada Swalayan Berdasarkan Pola Asosiasi Loka, Septi Kenia Pita; Afdal, M; Novita, Rice; Mustakim, Mustakim
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6305

Abstract

At present, competition in the business world is extremely fierce, particularly in the convenience store sector. The development of retail trade is progressing rapidly, accompanied by the emergence of many small markets and online shops. This situation encourages store owners to make wiser decisions, such as managing stock replenishment. If overlooked, this matter way hinder employees from locating the necessary items, thereby increasing the potential risk of goods expiring or being damage before they are sold. Therefore, store owners need to understand consumer behavior and shopping habits to assist iin stock management. Based on this issue, the research aims to analyze consumer purchasing patterns and optimize inventory stock. The result of this experiment identified two best rules, namely biscuit and consumption/food, with a confidence of 53,61%, a support of 15,57%, and a lift ratio of 1,116 the error measurement MAPE shows a value of 6,79 using alpha, beta and gamma values of 0,1. The total predicted stock in 52,086 with an actual value of 72,275, which is close to actual value of data prior to the significant observed in the last three months.
Expert System For Identifying Diseases In Native Chickens Using The Certainty Factor Method Arifin, Abdullah; Novita, Rice; Permana, Inggih; M. Afdal
Jurnal Sistem Cerdas Vol. 7 No. 3 (2024)
Publisher : APIC

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37396/jsc.v7i3.465

Abstract

Farming is the business of breeding and raising animals, divided into two groups: large animals (cows, buffaloes, horses) and small animals (chickens, ducks, birds). The demand for livestock, especially poultry like free-range chickens, is on the rise. However, many small to medium-sized free-range chicken farms still rely on conventional methods for disease treatment, which depend on the experience of the farmers. An expert system is a piece of computer software that mimics the choices and behaviors of a person or group with in-depth knowledge and expertise in a certain field. The objective of this study is to enhance the effectiveness of disease treatment for free-range chickens and streamline the diagnosis procedure. Farmers can determine which diseases are harming their free-range hens by using the Certainty Factor approach. Experts were surveyed to provide the data used in this study. Accurate diagnosis of diseases in free-range chickens and suitable treatment recommendations are provided by the system's diagnostic results.
Prediksi Produksi Kelapa Sawit Menggunakan Algoritma Support Vector Regression dan Recurrent Neural Network Alfakhri, Rezky; Permana, Inggih; Novita, Rice; Afdal, M
Building of Informatics, Technology and Science (BITS) Vol 6 No 3 (2024): December 2024
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i3.6441

Abstract

Oil palm is one of the important plantation crops and a leading commodity in Indonesia. PT. XYZ is a company engaged in receiving Fresh Fruit Bunches (FFB) to be processed into Crude Palm Oil (CPO) and Palm Kernel (PK). So far, the company has conducted statistical analysis with a correction value of 5% - 12% on the production results each month in targeting production results. However, this method is still lacking, because it uses manual calculations and considers estimates from personal experience. Therefore, this research proposes a data mining technique with Support Vector Regression (SVR) and Recurrent Neural Network (RNN) algorithms to predict production output precisely. In this study, testing was carried out on SVR hyperparameters, namely Kernel, C, Gamma, and Epsilon. While in RNN, testing is carried out on the optimizer, and the learning rate. In addition, the window size is also determined through a series of experiments, namely 3, 5, and 7. The comparison results show that the RNN model outperforms SVR with an RMSE value of 0.0928, MAPE of 14.32%, and R2 of 0.6164. The RNN model was then implemented to predict the next 3-month period. The prediction results show that there will be a significant increase in production in the first month, then a slight decrease in the second month, and an increase again in the third month.
Analysis of User Satisfaction Levels for X Mobile Application in Pekanbaru using End-User Computing Satisfaction (EUCS) and Technology Acceptance Model (TAM) Methods Butar Butar, Febiola Siska; Zarnelly, Zarnelly; Jazman, Muhammad; Novita, Rice; Marsal, Arif
Sistemasi: Jurnal Sistem Informasi Vol 14, No 1 (2025): Sistemasi: Jurnal Sistem Informasi
Publisher : Program Studi Sistem Informasi Fakultas Teknik dan Ilmu Komputer

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v14i1.4851

Abstract

Mobile X is a digital innovation in banking transactions, combining mobile banking, internet banking, and e-money functions into a single platform. However, Mobile X faces several usage challenges that lead to user dissatisfaction. Therefore, an analysis of user satisfaction is essential to improve customer loyalty. The End-User Computing Satisfaction (EUCS) and Technology Acceptance Model (TAM) methods are evaluation tools used to measure user satisfaction with an application system. This study employed a quantitative approach by distributing questionnaires to 100 respondents, determined using the Lameshow equation. The research model was analyzed through demographic analysis and model analysis using PLS-SEM, resulting in both internal and external models. The findings revealed that three hypotheses were accepted: the content variable (p-value = 0.495), the perceived usefulness variable (p-value = 0.007), and the timeliness variable (p-value = 0.001). Meanwhile, three hypotheses were rejected: the accuracy variable (p-value = 0.734), the content variable (p-value = 0.495), and the format variable (p-value = 0.184). Additionally, three user satisfaction factors were found to be significant for the accepted variables, indicating that meeting user expectations, perceived usefulness, information quality, and timeliness positively contribute to user satisfaction. This demonstrates that these factors effectively address user needs and enhance overall satisfaction with the Mobile X application.
Implementasi Algoritma Fuzzy C-Means menggunakan Model LRFM untuk Mendukung Strategi Pengelolaan Pelanggan Aini, Delvi Nur; Afdal, M.; Novita, Rice; Mustakim, Mustakim
JURNAL MEDIA INFORMATIKA BUDIDARMA Vol 8, No 3 (2024): Juli 2024
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/mib.v8i3.7616

Abstract

The same treatment of all customers will cause customers who are not so valuable to become value destroyers in the concept of Customer Relationship Management. Providing discounts and promos to all customers without differentiating customer segments has not provided significant benefits for a company. These two things are being experienced by BC 4 HNI Pekanbaru, so changes are needed in evaluating the strategies taken to maintain relationships with customers and form segments according to customer characteristics. Customer segments can be analyzed from sales transaction data. The purpose of this study is to manage and group sales transaction data in determining customer segmentation so that the strategy is more targeted. The analysis of customer transaction data was carried out by grouping the data using the Fuzzy C-means algorithm and the length, recency, frequency, monetary (LRFM) model, and AHP weighting.  The formation of the number of validated clusters of the silhouette index and ranking is carried out by multiplying the weight of AHP to find the customer lifetime value (CLV) so that it can be known which customer groups provide high value to the company. The result of this study is that BC 4 HNI Pekanbaru customers are grouped into 2 segments, namely the potential customer group which has a fairly frequent transaction value with an average monetary value of Rp. 2,802,495.00 and a fairly high number of transactions contribute greatly to the Company and the new customer group which means a new customer segment with uncertain funds, an average monetary of Rp. 104,567.00. Based on the segment, BC 4 HNI Pekanbaru can carry out a strategy in managing its customers according to the type of segment generated from this research.
Klasifikasi Citra X-Ray Tuberkulosis Menggunakan Convolutional Neural Networks Mubarak, Haykal Alya; Novita, Rice
Building of Informatics, Technology and Science (BITS) Vol 6 No 4 (2025): March 2025
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v6i4.6515

Abstract

Tuberculosis (TB) is a serious infectious disease that is still one of the main causes of death in the world, especially in developing countries. X-ray image analysis is an important step in controlling this disease. This research aims to classify X-ray images of tuberculosis using a deep learning approach with three Convolutional Neural Networks (CNN) architectures: DenseNet201, Xception, and MobileNetV2. The dataset used consists of 3,000 X-ray images, divided into two categories: normal and TBC, obtained from Kaggle, which are then processed through normalization, augmentation, and data division using the hold-out method with a ratio of 70:30, 80:20 , and 90:10. The research results show that DenseNet201 with the Nadam optimizer at 90:10 data division produces the highest accuracy of 100%, making it the best combination for TBC X-ray image classification. Xception achieved the best accuracy of 96.66% with the Nadam optimizer at a data split of 80:20. MobileNetV2 shows an optimal accuracy of 98.69% using the Adam optimizer at a 90:10 data split. This research proves that DenseNet201 with the Nadam optimizer is very effective in handling medical image classification, especially for tuberculosis. These results provide an important contribution to the development of deep learning-based technology to improve the accuracy of tuberculosis diagnosis.
Co-Authors Abror, Naufal Afandi, Rival Afdal, M.Afdal Afifah, Tiara Afrah Ahsyar, Tengku Khairil Aini, Delvi Nur Akhiril Anwar Harahap Al-Qadr, Nola Ardelia Alfakhri, Rezky Alfian, Zhevin Andriawan, Ahmad Rizky Angraini Angraini Anisa Putri Arif Marsal Arifin, Abdullah Arrazak, Fadlan Auliani, Sephia Nazwa Butar Butar, Febiola Siska Dwinnie, Zairy Cindy F. Safiesza, Qhairani Frilla Hamzah, Muhammad Luthfi Inggih Permana Jazman, Muhammad Khairil Ahsyar, Tengku Kurniawan, Viki Loka, Septi Kenia Pita M Afdal M. Afdal M. Afdal Megawati - Mochammad Imron Awalludin Mubarak, Haykal Alya Mulya, Anggi Munzir, Medyantiwi Rahmawati Munzir, Medyantiwi Rahmawita Mustakim Mustakim Mustakim Mustofa, Asdar Nabiilah, Nabiilah Novansyah, Rizky nursalisah, febi Pangestu, Prayugo Pertiwi, Tata Ayunita Pulungan, Joliando Putri, Suci Maharani Rahayu, Dwi Sri Rahmawita M, Medyantiwi Rozanda, Nesdi Evrilayan Safiq Tama, Naufal Sasmita, Indah Siti Rohimah Vamilina, Vina Wibisono, Yudistira Arya Yuda, Afi Ghufran Zarnelly Zarnelly Zarnelly Zarqani, Zarqani Zulkarnain Zulkarnain