Claim Missing Document
Check
Articles

Found 23 Documents
Search

MACHINE LEARNING FOR DATA CLASSIFICATION IN INDONESIA REGIONAL ELECTIONS BASED ON POLITICAL PARTIES SUPPORT Muhammad Fachrie
Jurnal Ilmu Komputer dan Informasi Vol 13, No 2 (2020): Jurnal Ilmu Komputer dan Informasi (Journal of Computer Science and Information
Publisher : Faculty of Computer Science - Universitas Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21609/jiki.v13i2.860

Abstract

In this paper, we discuss the implementation of Machine Learning (ML) to predict the victory of candidates in Regional Elections in Indonesia based on data taken from General Election Commission (KPU). The data consist of composition of political parties that support each candidate. The purpose of this research is to develop a Machine Learning model based on verified data provided by official institution to predict the victory of each candidate in a Regional Election instead of using social media data as in previous studies. The prediction itself simply a classification task between two classes, i.e. ‘win’ and ‘lose’. Several Machine Learning algorithms were applied to find the best model, i.e. k-Nearest Neighbors, Naïve Bayes Classifier, Decision Tree (C4.5), and Neural Networks (Multilayer Perceptron) where each of them was validated using 10-fold Cross Validation techniques. The selection of these algorithms aims to observe how the data works on different Machine Learning approaches. Besides, this research also aims to find the best combination of features that can lead to gain the highest performance. We found in this research that Neural Networks with Multilayer Perceptron is the best model with 74.20% of accuracy.
A Simple Vehicle Counting System Using Deep Learning with YOLOv3 Model Muhammad Fachrie
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 4 No 3 (2020): Juni 2020
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (834.469 KB) | DOI: 10.29207/resti.v4i3.1871

Abstract

Deep Learning is a popular Machine Learning algorithm that is widely used in many areas in current daily life. Its robust performance and ready-to-use frameworks and architectures enables many people to develop various Deep Learning-based software or systems to support human tasks and activities. Traffic monitoring is one area that utilizes Deep Learning for several purposes. By using cameras installed in some spots on the roads, many tasks such as vehicle counting, vehicle identification, traffic violation monitoring, vehicle speed monitoring, etc. can be realized. In this paper, we discuss a Deep Learning implementation to create a vehicle counting system without having to track the vehicles movements. To enhance the system performance and to reduce time in deploying Deep Learning architecture, hence pretrained model of YOLOv3 is used in this research due to its good performance and moderate computational time in object detection. This research aims to create a simple vehicle counting system to help human in classify and counting the vehicles that cross the street. The counting is based on four types of vehicle, i.e. car, motorcycle, bus, and truck, while previous research counts the car only. As the result, our proposed system capable to count the vehicles crossing the road based on video captured by camera with the highest accuracy of 97.72%.
Model Paralelisasi Algoritma Genetika Terpandu pada Sistem Penjadwalan Kuliah Universitas dengan Alokasi Waktu Dinamis Muhammad Fachrie; Anita Fira Waluyo
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 5 No 3 (2021): Juni 2021
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (381.209 KB) | DOI: 10.29207/resti.v5i3.2988

Abstract

One of the many techniques used to solve the University Course Timetable Problem (UCTP) is Genetic Algorithm (GA) which is a technique in the field of Evolutionary Computation. However, GA has high computational complexity due to the large number of evolutionary operators that must be performed during the evolutionary process, so it takes a long time to produce an optimal timetable. The computation time will also increase when the number of optimized variables is very large, such as in UCTP. Of course, this makes the application less reliable by users. Therefore, this article proposes a parallelization model for GA to reduce computation time in solving UCTP problems. The proposed AG is designed with a multithreading CPU scheme and implements a guided creep mutation mechanism and eliminates the recombination mechanism to reduce more computation time. The proposed system was tested and evaluated using two different UCTP datasets from the University of Technology Yogyakarta which contained 878 and 1140 lecture meetings in even and odd semesters. Unlike the previous ones, this study discusses UCTP with dynamic time slots where the duration of the lecture depends on the course credits. From the tests that have been done, it is found that the GA that was built is able to generate optimal course timetable without any clashes in a relatively fast time, that is less than 60 minutes for 1140 lecture meetings and less than 20 minutes for 878 lecture meetings. The use of the multithreading CPU model has succeeded in reducing computation time by 62% when compared to the conventional model which only uses one thread.
Sistem Pendukung Keputusan Untuk Mengukur Permintaan Produk Pada e-Commerce dengan Fuzzy Inference System: (Studi Kasus Orebae.com) Fadil Indra Sanjaya; Dadang Heksaputra; Muhammad Fachrie; Sulistyo Dwi Sancoko; Nuzula Afini; Zahra Septa Hati
J I M P - Jurnal Informatika Merdeka Pasuruan Vol 7, No 1 (2022): MARET
Publisher : Fakultas Teknologi Informasi Universitas Merdeka Pasuruan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37438/jimp.v7i1.404

Abstract

Measuring product demand is an important process for e-commerce companies to assess product viability in the future production. Measuring product demand can assist e-commerce companies to produce and developing new products based on market potential. Decision maker usually only using their best seller product as indicator to estimate future market trend. But in the fact future market trend will not only based on best seller product, but also there several criteria which is needs attention too. In order to use several criteria to estimate market trend, need some analysis so it will take a long time. With Decision Support System (DSS), decision making will be easier and faster. In this research the DSS takes into consideration the following input variables:  Total Sales (TS), Rating (R), Viewed (V), Total Comments (TC) and output Product Demand (PD). Once the Fuzzy Inference System model has been developed, an assessment of the variables is made through testing 1-years data, which allows verifying how the variables behave in the system under study, and their impact on the output variables. Through the application of Fuzzy Inference System in DSS regarding the modeling several criteria that impact product demand, it is possible to increased efficiency and maximizing profitKeyword— DSS, Fuzzy Inference System, Tsukamoto, e-Commerce, Product Demand
PEMANFAATAN JARINGAN SYARAF TIRUAN UNTUK MEMPREDIKSI KINERJA SATPAM Muhammad Fachrie; Adityo Permana Wibowo
JURNAL INFORMATIKA DAN KOMPUTER Vol 3, No 1 (2018): FEBRUARI - AGUSTUS 2018
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat - Universitas Teknologi Digital Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (631.376 KB) | DOI: 10.26798/jiko.v3i1.80

Abstract

Employee performance assessment is a very important task that has to be performed by Human Resource Department in a certain institution, because it would be useful for the policy holders to help them making a decision in promoting or not their employees. This paper delivered the analysis of using Multi Layer Perceptron (MLP) to predict the performance of security unit personnels which has been trained by a formal institution. The data that was used in this research were collected from PT. Garuda Merah Indonesia, that is a company that has role to train and to educate people who wants to be a security personnel. The data consist of 175 record with 10 attributes which include the assessment from aspects of cognitive, personality, and skill. MLP predicted the security personnel performances into three categories, i.e. “Good”, “Enough”, and “Fail”. The 10 folds Cross Validation technique was also used in testing phase to measure its performance comprehensively with the output of the best accuracy was 97,75%.
Estimation of Time Voting in Elections Using Artificial Neural Network Nur Hidayati; Muhammad Fachrie; Adityo Permana Wibowo
Compiler Vol 8, No 2 (2019): November
Publisher : Institut Teknologi Dirgantara Adisutjipto

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (453.369 KB) | DOI: 10.28989/compiler.v8i2.499

Abstract

Since the first election policy was enacted simultaneously, it does not mean that it does not have potential problems, instead it causes other problems, which require extra time and energy in doing recapitulation. Simultaneous elections consist of presidential elections, DPR elections, Provincial DPRDs, City / Regency DPRDs, DPD, the more they are elected, the more influential is the time of voting and the time of vote recapitulation. The longer the voting time is done by the voters, the longer the recapitulation time. The longer time of recapitulation results in the fatigue of KPPS members which triggers inaccurate work and prone to manipulation and fraud so that it can damage the quality of elections. This study aims to determine the estimated time needed for voting for ballots in elections using the Multilayer Perceptron Artificial Neural Network (ANN) approach. The resulting time estimate is based on the time of the voter in the voting booth. The results of this study indicate that ANN with the Multilayer Perceptron Algorithm can calculate the estimated time required for ballot balloting by producing the best combination of learning parameters with 4 hidden neurons, learning rate 0.001, and 2000 epoch iterations resulting in an RMSE value of 108,015 seconds.
Perancangan Aplikasi Pemesanan Tiket Wisata di Kecamatan Donorojo Menggunakan Metode Payment Gateway Hadi Wiyono; Muhammad Fachrie
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 5 No. 1 (2024): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v5i1.522

Abstract

Currently, the process of ordering tourist tickets is still done manually, especially in the Donorojo area, namely by visiting a travel agent or contacting the relevant party by telephone to order tickets. This process tends to take a long time, especially if users must queue to place an order. In addition, information regarding ticket prices and available schedules is not always easily accessible to users. The impact of this problem is confusion and errors in ordering tickets, as well as the buildup of queues at counters which causes congestion. To overcome this problem, we developed a tourist ticket booking application in Donorojo sub-district by implementing payment gateway methods including Flutter as a mobile application development framework and Firebase as a database management system that allows users to order tickets online, find out available tourist destination information and prices. and the schedule of tickets offered. The solution step taken is to collect and analyze data regarding popular tourist destinations, as well as implementing an online payment feature in the application. The interim results we have obtained are applications that are able to provide information about available tourist destinations, as well as provide convenience in the payment process and increase transaction security by using this payment gateway. Users can choose the desired destination, select the date and number of tickets needed, and make payment via the application. After payment is complete, the application will issue an electronic ticket which can be shown upon arrival at the tourist destination.
Pengembangan Aplikasi Android untuk Monitoring Stok Beras berbasis Internet of Things Bayu Ajiwicaksana; Muhammad Fachrie
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 5 No. 1 (2024): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v5i1.529

Abstract

Indonesia plays a significant role, particularly in the agricultural sector. The agricultural sector also contributes to the development of social, economic, and trade sectors. One of the products of the agricultural sector is rice. Rice is considered a staple in Indonesian households, serving as a crucial dietary component. Given its essential nature, rice owners must monitor their rice stock to ensure its consistent availability. However, in order to ascertain the rice stock, the rice owner must physically inspect it. The rice owner must, of course, rely on their memory to recall the remaining rice. This approach is less effective due to the limitations of human memory. To address this issue, the development of an Android-based system utilizing IoT (Internet of Things) technology becomes imperative. This system can provide real-time information on rice stock, accessible anytime and anywhere through a smartphone application. Therefore, rice owners will no longer face difficulties in monitoring their rice stock, as the information can be conveniently accessed via a mobile application.
Perancangan Sistem Pelayanan Administrasi Desa Berbasis Mobile (Studi Kasus: Desa Karangrena, Maos, Cilacap) Alfin Andrias Wardoyo; Muhammad Fachrie
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 5 No. 1 (2024): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v5i1.530

Abstract

Currently, village governments are still facing inadequate and inefficient village administration systems in providing services to the community. The incompetence and inefficiencies of village administration systems can be seen from the numerous complaints from residents. People have complained about complicated, time-consuming village management procedures with unclear requirements. Recent technological developments in practice can be utilized to solve existing problems in public services, such as the use of mobile applications. In other words, technological advances, especially in the field of mobile applications, can be a solution to improve the quality of public services, including village level administration services. The purpose of this research is to design a mobile-based village administration service system to facilitate administration services at the village level. The waterfall model is the system development method used in this research. The Dart programming language is utilized within the Flutter framework to develop the system. Black box testing is conducted to test the functionality of the system being built. The test results demonstrate that the system runs well as expected, thus successfully implementing a Village Administration Service System that leverages mobile technology.
Implementasi API Payment Gateway Midtrans pada Sistem Reservasi Dokter Gigi Berbasis Mobile Bilal Nurul Fauzi; Muhammad Fachrie
Jurnal Indonesia : Manajemen Informatika dan Komunikasi Vol. 5 No. 1 (2024): Jurnal Indonesia : Manajemen Informatika dan Komunikasi (JIMIK)
Publisher : Lembaga Penelitian dan Pengabdian Kepada Masyarakat (LPPM) AMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jimik.v5i1.535

Abstract

Common problems with dental clinics include large queues of patients, lack of optimization of patient administration, and manual payment systems that make it difficult for patients. A mobile-based dental clinic reservation application that is integrated with the Midtrans payment system is a solution to the difficulty of administering reservations at dental clinics, reducing patient queue problems, and optimizing payments. This research method consists of problem identification, system analysis, system design, implementation, and testing. System implementation uses Android Studio software with the Kotlin programming language and the Jetpack Compose framework. Payment implementation using midtrans API services. For data storage and management using Firebase Firestore. There are two users, namely patient and admin, each of whom has a different role. Patients can make new reservations and make payments, while admins can accept or reject reservations and payments. Application testing is carried out using the black box testing method. The test results show that the application successfully passed the test for each parameter tested.