Articles
Akurasi Algoritma Fletcher-Reeves untuk Prediksi Ekspor Karet Remah Berdasarkan Negara Tujuan Utama
Rapianto Sinaga;
Mora Malemta Sitomorang;
Deri Setiawan;
Anjar Wanto;
Agus Perdana Windarto
Journal of Informatics Management and Information Technology Vol. 2 No. 3 (2022): July 2022
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.47065/jimat.v2i3.170
Crumb rubber is a natural rubber specially designed to ensure its technical quality. Rubber is produced mainly in Southeast Asia, where Indonesia is the second largest producer in the world after Thailand. This study aims to predict the export of powdered rubber in Indonesia. The prediction method used is FletcherReeves which is one of the artificial neural network methods commonly used to predict data. The research data used is crumb rubber export data by main destination country for the period 2012-2020 which was obtained from the website of the Indonesian Central Statistics Agency. Based on this data, network architecture models will be trained and defined, including 7-10-1, 7-15-1, 7-20-1, 7-25-1, 7-30-1 (trancgf). Of the five models, after training and testing, the best data architecture model is 7-15-1 (trancegf) 7 is the input layer, 15 is the number of neurons in the hidden layer and 1 is the exit layer. The level of accuracy of the architectural model with the MSE value is 0.00482054.
Analisis Kepuasan Konsumen terhadap Pelayanan Store Ms Glow Menggunakan Metode Naïve Bayes
Ade Dwi Amanda;
Agus Perdana Windarto;
Hendry Qurniawan
Hello World Jurnal Ilmu Komputer Vol. 1 No. 3 (2022): Edisi Oktober
Publisher : Ilmu Bersama Center
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.56211/helloworld.v1i3.139
Store Wanda MS Glow (Magic Skin For Glowing) merupakan salah satu usaha jasa yang bergerak dibidang perawatan kecantikan yang sudah memiliki banyak konsumen. Store ini memerlukan tindakan eksklusif dikarenakan semakin banyak nya jumlah pesaing dibidang yang sama, dan diperlukan adanya evaluasi kualitas pelayanan. Oleh karena itu diperlukan analisis data yang melibatkan pernyataan konsumen. Sehingga nantinya diperoleh pernyataan konsumen terkait dengan beberapa aspek kepuasan konsumen dengan menggunakan metode data mining Naïve Bayes. Sumber data diperoleh dari kuesioner yang diberikan kepada konsumen secara random sebanyak 100 konsumen. Variabel yang digunakan dalam kepuasan konsumen terhadap pelayanan Store MS Glow antara lain : pelayanan, respon, gift, promo, terpercaya. Kesimpulan yang dihasilkan oleh peneliti dan software Rapid Miner dengan data training sebanyak 75 data. Data pengujian sebanyak 25 data testing yang diolah didalam Rapid miner 5.3. mendapatkan hasil pengujian dengan akurasi sebesar 88.00% yaitu 13 konsumen Puas dan sebanyak 12 konsumen Tidak Puas.
Penerapan Metode VIKOR Dalam Menentukan Aplikasi Belanja Online Terbaik Berdasarkan Konsumen
Rika Setiana;
Widya Try Taradipa;
Agus Perdana Windarto
Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1 (2022): Edisi Desember
Publisher : LPPM STIKOM Tunas Bangsa
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1.105
People's lives are becoming more instantaneous as technology advances, and many people have a lot of freedom in their daily activities. This means that people need comfortable goods to meet their daily needs. A marketplace is a type of media that serves as a place for sellers and buyers to do business and trade. People buy things online a lot these days, but a lot of them don't know how to choose an online marketplace, so the authors decided to do some research on how to choose the best online shopping app so that people can easily figure out which app to use and won't be disappointed in the future. So, it's important to figure out which is the best online shopping app. This study uses the VIKOR method to make a decision-making system. In this study, the authors started by doing surveys. They chose a data collection method based on a questionnaire or a Google form questionnaire. Based on how the VIKOR method was used in this study, the best marketplace that can be used is Shoope. With this research, it is hoped that the public will be able to find out how to choose the best app for shopping online.
Application of Data Mining Classification to Store Customer Satisfaction Bombay Textiles
Siti Sundari;
Agus Perdana Windarto;
Yuegilion Pranayama Purba
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 4 (2022): December
Publisher : Yayasan Literasi Sains Indonesia
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (491.113 KB)
|
DOI: 10.55123/jomlai.v1i4.1672
This study aims to obtain a model of rules in classifying the level of customer satisfaction at Bombay Textile Stores. By knowing the level of customer satisfaction, shop owners can improve service if it is not good and further improve service if the level of satisfaction is good. This study measures the level of customer satisfaction at the Bombay Textile Store. The method used in this study is the C4.5 Algorithm, where the data source used is a questionnaire/questionnaire technique given to Bombay Textile Store customers. The variables used include Service, Quality of Goods, Price, Facilities, and Promotion. The results obtained 11 rules for the classification of customer satisfaction levels with 5 rules satisfied status and 6 rules dissatisfied status. The C4.5 algorithm can be used in the case of customer satisfaction levels with an accuracy rate of 96.67%. From the results of the analysis, it is hoped that it can be applied so that it can be used as a decision to improve service to customers.
Application of Multiple Regression in Estimating the Amount of Population Growth in Siantar District
Ayu Wulandari;
Agus Perdana Windarto;
Hendry Qurniawan
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 4 (2022): December
Publisher : Yayasan Literasi Sains Indonesia
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (511.466 KB)
|
DOI: 10.55123/jomlai.v1i4.1677
The purpose of this study is to solve a case or problem that makes decision makers experience various obstacles in estimating the amount of population growth each year by using multiple linear regression algorithms as a solution to solving cases. The data used in this study was obtained directly from the official website of the Central Statistics Agency (BPS) Simalungun in the form of a softcopy book file entitled "Siantar District in Figures 2020" via the url http://simalungun.bps.go.id. with population data from the Siantar District from 2016-2020, there are 17 villages. The data that has been obtained is then processed using Data Mining estimates of the Multiple Linear Regression algorithm and research testing is carried out using the help of Rapid Miner 9.10 Software. By doing this research, research results are obtained that can provide information or input to the government through related agencies to anticipate the number of population growth in Siantar District every year.
Analisis Pemilihan Produk Bedak Padat Terbaik Berdasarkan Pilihan Konsumen Menggunakan Metode SMART
Leza Khairani;
Zahra Nur Atthiyah;
Agus Perdana Windarto
Explorer Vol 3 No 1 (2023): January 2023
Publisher : Forum Kerjasama Pendidikan Tinggi
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.47065/explorer.v3i1.453
Facial beauty at this time can usually use cosmetics. Cosmetics have been known to people for centuries, and it was only in the 19th century that they received special attention as beauty. Cosmetics have become part of the business. Products used to enhance one's appearance are usually called cosmetics. Pressed powder is a type of cosmetic needed to complete facial makeup and is used to correct facial imperfections such as covering acne, dull skin, and facial blemishes. With so many counterfeit products out there, consumers are afraid to be careless in choosing pressed powder. So, in this article aimed at helping consumers decide which pressed powder is best for them, the authors used the SMART method to conduct research on pressed powder. In this article, data collection was carried out by distributing questionnaires to consumers via social media. In this article, 371 respondents were obtained using 5 criteria and 16 alternatives, and 132 respondents were randomly selected to be used as material to be processed using the SMART method. In this article, we have identified the best-pressed powder with the highest index value, namely the madam gie alternative with a value of 1. This research aims to assist consumers in choosing the best-pressed powder.
Penerapan Machine Learning Dalam Memprediksi Produksi Rute Pergerakan Pesawat Domestik di Indonesia
Rika Setiana;
Widya Try Taradipa;
Agus Perdana Windarto
Brahmana : Jurnal Penerapan Kecerdasan Buatan Vol 4, No 1A (2022): Edisi Desember (Spesial Issue)
Publisher : LPPM STIKOM Tunas Bangsa
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.30645/brahmana.v4i1A.141
Seiring terjadinya pandemi Covid-9 di Indonesia, jumlah penumpang pesawat di Indonesia mengalami penurunan. Penurunan jumlah penumpang mengakibatkan jumlah rute pergerakan domestik menurun. Perlu ada kajian yang mendalam mengenai prediksi jumlah rute penerbangan domestik ke depan agar pihak maskapai dapat melakukan pengaturan jadwal penerbangan domestik kembali agar tidak mengalami kerugian akibat operasional yang tidak sesuai dengan pemasukan. Solusi yang digunakan pada permasalahan ini adalah menggunakan metode machine learning untuk memprediksi produksi rute pergerakan pesawat domestik. Algoritma yang digunakan adalah algoritma backpropagation dengan dua metode yaitu conjugate gradient fletcher reeves dan powell-beale. Hasil pelatihan dan pengujian menggunakan algoritma backpropagation dengan kedua metode menujukkan bahwa metode powell-beale adalah metode yang terbaik dengan nilai performance pengujian terkecil adalah = 0,0010 dengan epoch 34.
Classification Analysis of Back propagation-Optimized CNN Performance in Image Processing
Putrama Alkhairi;
Agus Perdana Windarto
Journal of Systems Engineering and Information Technology (JOSEIT) Vol 2 No 1 (2023): March 2023
Publisher : Ikatan Ahli Informatika Indonesia
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29207/joseit.v2i1.5015
This study aims to optimize the performance of the Convolutional Neural Network (CNN) in the image classification task by applying data augmentation and fine-tuning techniques to a case study of mammal classification. In this study, we took a fairly complex image classification dataset and used the CNN model as a basis for training and evaluating the performance of the model compared to Back propagation. From this study, the CNN VGG16 architecture optimized with ADAM optimization has been compared with the Back propagation optimization of SGD. We also conducted a literature review on several related studies and basic concepts in CNN, such as convolution, pooling, and fully connected layers. The research methodology involves creating datasets using data augmentation techniques, model training using fine-tuning techniques, and testing model performance using a number of evaluation metrics, including accuracy, precision, and recall. The results of this study indicate that the techniques used have succeeded in improving the performance of the CNN model in complex image classification tasks with accuracy in identifying and monitoring animal species more accurately, with an accuracy of 91.18% for the best model. Model accuracy increased by 2% after applying data augmentation and fine-tuning techniques to the CNN model. These results indicate that the techniques applied in this study can be a good alternative in improving the performance of the CNN model in the image classification task.
Penerapan Data Mining Klasifikasi Pada Calon Pelanggan Baru Indihome dengan C.45
Arfandi Arfandi;
Agus Perdana Windarto;
Ilham Syahputra Saragih
Journal of Informatics, Electrical and Electronics Engineering Vol. 1 No. 1 (2021): September 2021
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Indihome customers are people who are buyers of products that have been made and marketed by a company, where this person not only buys the product once but repeatedly. Meanwhile, Prospective Indihome customers are those who have not become your customers (buyers/service users) but are considered to have the opportunity to become customers in the near future or in the future. An algorithm is needed to classify indihome's new prospective customers so that there is no loss between the two parties. The C4.5 algorithm was chosen because it is able to classify new prospective customers of Indihome by using the rapid miner application and the calculation using Microsoft Excel. From the calculations using the two applications, the results obtained include: If Income <2000000 and Type of House = Contract, then the result is Not Eligible {Eligible = 0 and Not Eligible = 7}and the result is feasible, namely the rule If Employment = Self Employed and Income> 2000000 then the result is Eligible (Eligible = 54 and Not Eligible = 0)
Algoritma Machine Learning untuk penentuan Model Prediksi Produksi Telur Ayam Petelur di Sumatera
Ihsan Maulana Muhamad;
Sigit Anugerah Wardana;
Anjar Wanto;
Agus Perdana Windarto
Journal of Informatics, Electrical and Electronics Engineering Vol. 1 No. 4 (2022): Juni 2022
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
Laying hens eggs are one of the livestock commodities that make a very large contribution to the supply of eggs as a community need. Therefore, it is necessary to predict the egg production of laying hens in the future so that in the future the need for eggs in Indonesia is stable and can meet the demands of the Indonesian people. The method used in this research is a machine learning algorithm, namely Polak-Ribiere which is one of the artificial neural network methods that is often used to predict data. This study does not discuss the prediction results, but will discuss the ability of the Machine Learning algorithm to make predictions based on the egg production dataset of laying hens obtained from the Central Statistics Agency. The research data used is data on the production of laying hens in Sumatra from 2015-2020. Based on this data, network architecture models will be determined, including 4-5-1, 4-10-1, 4-15-1, 4-20-1, and 4-25-1. Of the five models, training and testing were carried out first and then obtained the results that the best architectural model was 4-25-1 with 0.03144841, the lowest among the other 4 models. So it can be concluded that the model can be used to predict the egg production of laying hens.