Garuda - Garba Rujukan Digital

Pendekatan Machine Learning: Analisis Sentimen Masyarakat Terhadap Kendaraan Listrik Pada Sosial Media X Kusuma, Gathot Hanyokro; Permana, Inggih; Salisah, Febi Nur; Afdal, M.; Jazman, Muhammad; Marsal, Arif
JUSIFO : Jurnal Sistem Informasi Vol 9 No 2 (2023): JUSIFO (Jurnal Sistem Informasi) | December 2023
Publisher : Program Studi Sistem Informasi, Fakultas Sains dan Teknologi, Universitas Islam Negeri Raden Fatah Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19109/jusifo.v9i2.21354

Environmental issues and the depletion of fossil fuels continue to escalate as the number of fossil fuel-based vehicle users increases in Indonesia. Electric vehicles emerge as one of the potential alternative solutions to address current environmental challenges, given their eco-friendly nature and lack of pollution emissions. Sentiment analysis is conducted to understand public responses, both supportive and opposing, towards electric vehicles. This research aims to analyze the sentiment of X-social media users regarding electric vehicles using machine learning techniques. The research stages include data collection, data selection, preprocessing, and classification using Naïve Bayes Classifier (NBC), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) algorithms. The test results show that on a balanced dataset using ROS, SVM performs the best with accuracy = 68.7%, precision = 77.9%, and recall = 68.4%. Meanwhile, NBC yields an accuracy of 60.3%, precision of 61.3%, and recall of 60.3%, while KNN has an accuracy of 53.9%, precision of 54%, and recall of 53.9%.

Perbandingan Algoritma KNN, NBC, dan SVM: Analisis Sentimen Masyarakat Terhadap Perparkiran di Kota Pekanbaru Intan, Sofia Fulvi; Permana, Inggih; Salisah, Febi Nur; Afdal, M.; Muttakin, Fitriani
JUSIFO : Jurnal Sistem Informasi Vol 9 No 2 (2023): JUSIFO (Jurnal Sistem Informasi) | December 2023
Publisher : Program Studi Sistem Informasi, Fakultas Sains dan Teknologi, Universitas Islam Negeri Raden Fatah Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19109/jusifo.v9i2.21357

The public response in Pekanbaru to parking policies and regulations has given rise to various sentiments, both positive and negative. This discussion extends not only within the local community but also across various social media platforms. This research aims to analyze public sentiment towards the new parking policies and regulations in the Pekanbaru area. The study involves the KNN, NBC, and SVM algorithms to classify public sentiment into positive, neutral, and negative categories. Balancing techniques used in this research include Random Over Sampling (ROS) and Random Under Sampling (RUS). The data utilized in this study were obtained from posts on the social media platform X. The testing of the dataset using ROS resulted in high accuracy, precision, and recall values. The findings of this research indicate that overall, the SVM algorithm outperforms KNN and NBC in terms of accuracy, precision, and recall. Additionally, the most dominant sentiment is negative, with 422 tweets expressing dissatisfaction with the current parking policies.

Penggunaan Metode FMEA dalam Penilaian Manajemen Risiko Keamanan Sistem Informasi Rumah Sakit Saputri, Setia Ningsih; Salisah, Febi Nur; Maita, Idria; Rozanda, Nesdi Evrilyan
Jurnal Inovtek Polbeng Seri Informatika Vol 9, No 1 (2024)
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/isi.v9i1.3951

Untuk mengurangi risiko yang terkait dengan suatu organisasi, identifikasi dan pengendalian risiko disebut manajemen risiko. Tujuan manajemen risiko adalah untuk melindungi dan memperkecil setiap kegagalan keamanan sistem informasi perusahaan dari tingkat risiko yang paling tinggi sehingga tidak dapat mencapai tujuan perusahaan. Metode FMEA (Failure Mode and Effect Analysis) adalah salah satu metode yang digunakan dalam penilaian dan pengelolaan risiko. Penelitian ini menganalisa risiko pada sistem informasi suatu industri yang bergerak dibidang kesehatan, yaitu RSU Indah Bagan Batu. RSU Indah Bagan Batu menerapkan sistem informasi yang Bernama SIMRS Khanza. Menentukan tingkat risiko yang ada pada sistem informasi RSU Indah Bagan Batu adalah tujuan dalam penelitian ini, karena SIMRS Khanza harus melindungi data pasien, dokter, obat-obatan, dan staf lainnya dari potensi ancaman. Berdasarkan penilaian risiko menggunakan metode FMEA yang dihitung melalui perkalian severity, occurrence, dan detection maka mendapatkan hasil RPN dengan 34 risiko yang terdapat pada sistem informasi yang digunakan. Dari 34 RPN yang dihasilkan terdapat 5 kategori level RPN yaitu 3 kategori very high dengan rentang 500-504, 5 kategori high dengan rentang 120-180, 6 kategori moderate dengan rentang 84-108, 17 kategori low dengan rentang 20-72 dan 4 kategori very low dengan rentang 8-15.

Perbandingan Metode TAM Dan UTAUT Dalam Penerimaan Dan Kepuasan Sistem Informasi Administrasi Akademik Nurrahma, Intan; Salisah, Febi Nur; Ahsyar, Tengku Khairil; Rahmawita, Medyantiwi
Jurnal Inovtek Polbeng Seri Informatika Vol 8, No 2 (2023)
Publisher : P3M Politeknik Negeri Bengkalis

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35314/isi.v8i2.3768

Universitas XYZ merupakan lembaga yang bergerak dalam bidang pendidikan. Saat ini banyak universitas yang menggunakan sistem informasi akademik untuk memudahkan pekerjaan agar menjadi lebih efektif dan efisien. Hasil survei menunjukkan bahwa adanya beberapa kendala pada sistem informasi administrasi akademik yaitu terjadinya error ataupun server down kondisi error ini seperti tidak bisa login ataupun tidak bisa mengakses fitur yang tersedia kendala berikutnya yaitu ketika mahasiswa ingin mengganti password mereka tidak bisa langsung mengubah di sistem tersebut tetapi harus ke sistem yang lain untuk mengubah password. Penelitian ini menggunanakan metode TAM Dan UTAUT. Tujuan penelitian ini yaitu untuk mengetahui metode mana yang lebih baik dalam penerimaan dan kepuasan penerapan sistem informasi. Hasil dari penelitian menunjukkan Metode UTAUT lebih mampu menjelaskan dengan baik dari pada metode TAM dalam membandingkan penerapan sistem informasi administrasi akademik dimana dengan menggunakan metode TAMdiperoleh nilai R-square pada TAM variabel IT Acceptance adalah 78,2% yang mana nilai ini termasuk kategori model yang kuat dan variabel User Satisfaction nilai R-square nya yaitu 65,3% dimana nilai tersebut masih termasuk dalam kategori kuat. sedangkan pada metode UTAUT R-square untuk variabel IT Acceptance yaitu 83,6% yang mana nilai ini termasuk kategori model yang kuat pada variabel User Satisfaction nilai R-square nya yaitu 74,6% dimana nilai tersebut termasuk kategori kuat.

Peramalan Jumlah Kedatangan Wisatawan Menggunakan Support Vector Regression Berbasis Sliding Window Fitriah, Ma’idatul; Permana, Inggih; Salisah, Febi Nur; Munzir, Medyantiwi Rahmawita; Megawati, Megawati
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.v8i3.7408

As a developing city, Pekanbaru has the potential for attractive tourist attractions for tourists. The arrival of tourists has had a big positive impact on the economy of Pekanbaru City. The number of tourist arrivals can experience ups and downs every month, for this reason it is necessary to forecast the number of tourists in the future. This research aims to apply the Orange Data Mining application in predicting the number of tourist arrivals by comparing the kernels in the Support Vector Regression (SVR) method and applying Sliding Window size 3 to window size 13 to transform into time series data. As well as sharing data using the K-Fold Validation method with a value of K-10. Then the performance of the kernels used can be seen using the Test and Score widget which presents the results of Root Mean Absolute Error (MAE), Mean Square Error (MSE), Root Mean Square Error (RMSE), dan R-squared(R2). The results for forecasting the number of tourist arrivals to Pekanbaru City using the SVR method show that the RBF Kernel is the optimal choice compared to the Polinomial and Linear Kernels. The results of the Test and Score widget show that the RBF Kernel with window size 10 has lower MAE, MSE and RMSE values, namely 0.118, 0.022 and 0.147. Apart from that, the comparison of R2 in window size 10 for Kernel RBF shows better performance with a value of 0.519.

Prediksi Jumlah Bayi Penerima Imunisasi DPT 1 dan DPT 2 Menggunakan Support Vector Regression Idriani R, Nova; Permana, Inggih; Salisah, Febi Nur; Megawati, Megawati; Rahmawita M, Medyantiwi
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.v8i3.7694

Vaccination against diphtheria, pertussis (whooping cough), and tetanus is known as DPT immunization, which protects a person from three serious diseases. This vaccine is given in the form of an injection where there are 5 antigens in one injection of the vaccine. DPT immunization is a complete routine immunization that will be continued in grades 1 to 6 elementary school. DPT immunization is feared by mothers because of the side effects that occur in babies after the vaccine injection, namely that the baby will have a fever and be fussy. This has resulted in delays in collecting data on babies who have received this immunization, which has an impact on estimates of babies who will receive DPT immunization in the following month. Of course, this will disrupt the stock of vaccines provided, causing the potential for them to be out of stock. To overcome this problem, it is necessary to collect data on babies who have received DPT in the previous month. This data will be used to predict babies who will receive DPT immunization in the following month using the Support Vector Regression (SVR) method. So that the community health center can provide information regarding the prediction of the number of babies who will receive DPT immunization. This method uses three kernels and a Sliding Window to divide the data into smaller segments, moving alternately across the time series data, making it suitable for predicting babies who will receive DPT immunization in the next time interval. From the three kernels used on the two data that have been separated into DPT 1 and DPT 2, windowing size 3 linear kernels were obtained which were selected as an accurate evaluation of model work on DPT 1 with MAPE values of 3.35, RMSE 0.193, and R2 0.1. And windowing size 3 RBF kernels are more optimal in DPT 2 with MAPE values of 7.86, RMSE 0.163, and R2 0.288.

Analysis User Satisfaction of XYZ Application with End User Computing Satisfaction Method and Delone and Mclean Giansyah, Qhoiril Aldi; Maita, Idria; Megawati, Megawati; Salisah, Febi Nur
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 18, No 4 (2024): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.101004

Rapidly developing technology has made internet service providers such as XYZ provide an application which promises convenience for its consumers to make transactions to purchase credit, internet quota, and other services. this application is one of the fairly well-known providers and is currently competing with other providers. this provider targets young people because in terms of its products which are quite competitive with a wide reach and prices that are quite cheap for young people. Of course, customers will feel the convenience of transacting using the official service application. The convenience offered can make business processes more efficient and can be used to improve the quality of their services.This study uses the End User Computing Satisfaction model method and the Delone and Mclean model with 9 research variables (content, accuracy, format, ease of use, timeliness, system quality, information quality, and service quality, and security), The purpose of this study is to assess the effectiveness of the application..Out of the nine variables included in the research results, only two Service Quality and Accuracy have a statistically significant impact. The model provided in this study has a Rsquared score of 0.792, indicating a strong level of customer satisfaction.

Perbandingan Performa Algoritma NBC, C4.5, dan KNN dalam Analisis Sentimen Masyarakat terhadap Krisis Petani Muda pada Media Sosial Facebook Nurkholis, Nurkholis; Permana, Inggih; Salisah, Febi Nur; Mustakim, Mustakim; 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.v6i3.6082

In Indonesia, young farmers face various challenges and crises that hinder the growth and sustainability of the agricultural sector. They face obstacles such as lack of access to capital, limited technology, climate change, and low selling prices for their crops. In addition, they also often face problems in obtaining accurate and relevant information in an effort to facilitate better decision-making in agricultural businesses, so that the interest of young people today to become farmers is decreasing. The study aims to Compare the Performance of NBC, C4.5, and KNN Algorithms in the Analysis of Public Sentiment towards the Young Farmer Crisis on Facebook Social Media. The application of the K-Fold Cross Validation method is (K = 10). Sentiment analysis is carried out with 3 labels (positive, negative, and neutral). The data used in making the classification model (data from preprocessing the stemming column) using (Google Colab) amounted to 4,878 data with Positive sentiment of 43.13% (2,104), Neutral 39.59% (1,931), Negative 17.28% (843) from the initial data without nested comments, which is 4,981 and the total number of Facebook data is 2,900 likes, 6,700 comments, and 3.3 million viewers. The accuracy of the NBC algorithm is 57.32%, the C4.5 algorithm is 98.42%, and the KNN algorithm (K = 19) is 97.33%. It can be concluded that the results of the comparison of the performance of the three algorithms using (Rapidminer10.3), the C4.5 algorithm gets a higher accuracy of 98.42% and is superior because it produces a decision tree.

IMPLEMENTASI SISTEM PREDIKSI KETERSEDIAAN PRODUK DENGAN METODE SIMPLE MOVING AVERAGE PADA UD. MITRA ARTHA JAYA Salisah, Febi Nur; Gustinov, Mhd Dion; Mawaddah, Zuriatul
Seminar Nasional Teknologi Informasi Komunikasi dan Industri 2024: SNTIKI 16
Publisher : UIN Sultan Syarif Kasim Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar

UD. Mitra Artha Jaya menghadapi masalah dalam pengelolaan persediaan barang, yang menyebabkan beberapa produk sering kedaluwarsa atau kehabisan stok. Penelitian ini bertujuan untuk mengimplementasikan sistem peramalan menggunakan metode Simple Moving Average (SMA) untuk memperbaiki perencanaan kebutuhan pelanggan dan mengoptimalkan persediaan produk. Metode SMA dipilih karena kemampuannya dalam menggunakan data permintaan aktual untuk menghasilkan ramalan yang akurat. Penelitian ini menggunakan metode Waterfall untuk pengembangan sistem, yang mencakup identifikasi kebutuhan aplikasi, pengelolaan data master, dan perhitungan error menggunakan MAPE, MAD, dan MSE. Hasil penelitian ini didapatkan Teh Gelas dengan MAD dan MSE terendah, MAPE sebesar16,09% hal ini menunjukan bahwa produk Teh gelas Akurat dan Produk dengan peramalan paling tidak akurat terdapat pada produk Aqua dengan MAD dan MSE tertinggi, MAPE sebesar 20,92% dan juga produk Teh Pucuk dengan MAPE tertinggi sebesar 27,45%. Pengujian sistem yang dilakukan mendapatkan hasil yang bagus sehingga sitem dapat membanti dalam proses peramalan

Prediksi Jumlah Bayi Penerima Imunisasi DPT 1 dan DPT 2 Menggunakan Support Vector Regression Idriani R, Nova; Permana, Inggih; Salisah, Febi Nur; Megawati, Megawati; Rahmawita M, Medyantiwi
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.v8i3.7694

Vaccination against diphtheria, pertussis (whooping cough), and tetanus is known as DPT immunization, which protects a person from three serious diseases. This vaccine is given in the form of an injection where there are 5 antigens in one injection of the vaccine. DPT immunization is a complete routine immunization that will be continued in grades 1 to 6 elementary school. DPT immunization is feared by mothers because of the side effects that occur in babies after the vaccine injection, namely that the baby will have a fever and be fussy. This has resulted in delays in collecting data on babies who have received this immunization, which has an impact on estimates of babies who will receive DPT immunization in the following month. Of course, this will disrupt the stock of vaccines provided, causing the potential for them to be out of stock. To overcome this problem, it is necessary to collect data on babies who have received DPT in the previous month. This data will be used to predict babies who will receive DPT immunization in the following month using the Support Vector Regression (SVR) method. So that the community health center can provide information regarding the prediction of the number of babies who will receive DPT immunization. This method uses three kernels and a Sliding Window to divide the data into smaller segments, moving alternately across the time series data, making it suitable for predicting babies who will receive DPT immunization in the next time interval. From the three kernels used on the two data that have been separated into DPT 1 and DPT 2, windowing size 3 linear kernels were obtained which were selected as an accurate evaluation of model work on DPT 1 with MAPE values of 3.35, RMSE 0.193, and R2 0.1. And windowing size 3 RBF kernels are more optimal in DPT 2 with MAPE values of 7.86, RMSE 0.163, and R2 0.288.

Title

Found 30 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 30 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 30 Documents
Search