Claim Missing Document
Check
Articles

Found 13 Documents
Search

Decision tree Penerapan Perhitungan Metode Decision Tree Menggunakan Algoritma Iterative Dichotomiser 3 (ID3) Berbasis Website Agus Susanto; Sasmitoh Rahmad Riady; Shita Dwi Ranti; Rila Mandala
Jurnal Sains Indonesia Vol 1 No 2 (2020): Volume 1, Nomor 2, 2020 (Juli)
Publisher : PUSAT SAINS INDONESIA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59897/jsi.v1i2.11

Abstract

This study discusses the implementation of the Decision Tree algorithm into a website. Many calculation tools for machine learning methods include Rapidminner, WEKA, etc. These tools can also calculate various other machine learning methods. However, in this tool, many processes are carried out in testing even though only using one method such as the decision tree. Therefore, this study creates a website-based system that focuses on the Decision Tree ID3 method only, The results of the system calculations will be compared with the results of manual calculations. The website-based calculation tool implementation system for the ID3 algorithm is simple and easy to use, only by uploading training and testing data. Then the results of the calculation will be displayed on the website page. Based on the comparison of manual calculations with the system, there is no significant difference from the manual calculations. The system can run online and can access it from anywhere. In addition to displaying the results of calculations, this system also provides a predictive output and rule base from the calculation results of the decision tree algorithm Dipenelitian ini membahas mengenai penererapan perhitungan Decision Tree ID3 kedalam sebuah website, dimana tool perhitungan metode-metode machine learning yang sudah banyak digunakan diantaranya adalah Rapidminner, WEKA dll yang dapat menghitung berbagai metode machine learning lainnya, namun pada tool ini banyak proses dalam melakukan testing walaupun hanya menggunakan satu metode seperti halnya decision tree ID3. Maka dari itu peneliti membuat sebuah sistem berbasis website yang berfocus pada metode Decision Tree ID3 saja, dari hasil perhitungan sistem akan dibandingkan dengan hasil perhitungan manual. Sistem penerpan perhitungan algoritma ID3 berbasis website ini sangat sederhana dan mudah digunakan dalam prosesnya, cukup dengan upload data training dan testing saja maka hasil dari perhitungan akan ditampilkan dihalaman website. Dari hasil perbandingan perhitungan manual dengan sistem, tidak ada perbedaan yang signifikan dengan perhitungan manualnya, dan sistem dapat berjalan dengan online dan di akses dari mana saja, selain menampilkan hasil perhitungan, sistem ini pun memberikan sebuah output hasil prediksi dan rule base dari hasil perhitungan algoritma pohon keputusan.
Pemanfaatan Sistem Distribusi dalam Berbagi Paket Pulsa untuk Short Message Service (SMS) Sasmitoh Rahmad Riady; Tjong Wan Sen
Jurnal Sains Indonesia Vol 1 No 2 (2020): Volume 1, Nomor 2, 2020 (Juli)
Publisher : PUSAT SAINS INDONESIA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59897/jsi.v1i2.9

Abstract

Penelitian ini membahas mengenai Short Message Service (SMS) yang telah menjadi fitur default yang terdapat pada ponsel pintar atau ponsel tradisional dan salah satu fitur yang mulai dieksploitasi sepenuhnya oleh pengguna ponsel dalam beberapa tahun terakhir. Dalam konsep Short Message Service kali ini berbeda seperti biasanya dimana proses pengiriman pesan dari aplikasi seluler Client ke aplikasi seluler Transceiver dan kemudian diteruskanke ponsel target, dalam mengirim pesan ini dilakukan secara berurutan. untuk memanfaatkan paket SMS di dalam aplikasi Transceiver yang akan digunakan oleh aplikasi Client dalam mengirim sebuah pesan. Konsep SMS menggunakan konsep Device-to-Device untuk komunikasi antara ponsel ke ponsel, kemudian untuk komunikasi antara ponsel menggunakan TCP / IP Socket sebagai jalur komunikasi dalam mengirim SMS lalu memanfaatkan paket pulsa yang terdapat dalam aplikasi Transceiver dalam meneruskan pesan SMS dari ponsel Client tersebut ke ponsel Target. Ini adalah sebuah teknik Sistem Terdistribusi untuk berbagi sumber daya paket pulsa SMS dalam pengiriman pesan.
Implementasi Cost Control System Berbasis Website pada Departemen PPIC PT XYZ Menggunakan Analisis SWOT Khalis Sofi; Aswan Supriyadi Sunge; RR Wening Ken Widodasih; Sasmitoh Rahmad Riady
Jurnal Sains Indonesia Vol 1 No 2 (2020): Volume 1, Nomor 2, 2020 (Juli)
Publisher : PUSAT SAINS INDONESIA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59897/jsi.v1i2.12

Abstract

Cost Control merupakan komponen penting penunjang berjalannya operasional perusahaan. Pada PT XYZ pekerjaan Cost Control dikendalikan oleh departemen PPIC (Production Planning and Inventory Control). Permasalahan yang dihadapi dalam menjalankan pekerjaan ini adalah kurang terkontrolnya dokumen-dokumen Purchase Requirement yang berupa lembaran kertas NCR (No Carbon Required) sehingga berakibat terhadap kurang teraturnya pembuatan Purchase Order yang dilakukan oleh departemen Purchasing. Hal-hal tersebut berimbas kepada budget yang tersedia, akibatnya PPIC sering menemukan kasus over budget. Maka dari itu, dibutuhkan sistem Cost Control untuk memonitoring aktivitas-aktivitas tersebut dan membantu untuk pengambilan keputusan bagi manajamen terkait budgeting. Sebelum dijalankannya sistem, maka diperlukan sebuah analisis mengenai kelemahan, kelebihan, peluang, dan ancaman yang terdapat dalam analisis SWOT. Sehingga sistem Cost Control mampu mengendalikan biaya untuk pemakaian budget berikutnya.
Implementasi Sistem Monitoring Suhu Pada Produk Makanan di Mesin Sterilisasi Menggunakan Fuzzy Logic Berbasis Internet of Things Sasmitoh Rahmad Riady; Donny Maulana; Agus Suwarno; Agung Nugroho
InComTech : Jurnal Telekomunikasi dan Komputer Vol 8, No 2 (2018)
Publisher : Department of Electrical Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22441/incomtech.v8i2.4089

Abstract

Penelitian ini bertujuan untuk mengimplementasikan teknologi internet of things dan metode logika fuzzy terhadap monitoring suhu pada proses sterilisasi agar data suhu update secara real time dan bisa di akses dari manapun oleh management atau customer dan memberikan keputusan yang akurat dari hasil proses sterilisasi menggunakan metode logika fuzzy. Pada study kasus yang dihadapi terdapatlah beberapa masalah diantaranya adalah record data masih dilakukan secara manual oleh pihak operator machine dan hasil perhitungan dari team quality control untuk mencari nilai  masih terdapat ambiguitas atau ketidakjelasan. Maka peneliti mengusulkan diterapkannya teknologi internet of things sebagai monitoring suhu pada proses sterilisasi beserta alat ukur suhu menggunakan Raspberry Pi3 serta sensor DS18B20 dan mengolah nilai  dan isi mesin sterilisasi kedalam himpunan fuzzy. Sistem yang di bangun dengan metode prototyping ini dapat menampilkan data suhu dan hasil dari proses sterilisasi pada produk makanan secara update dan dapat menentukan status sterilisasi dengan kedua variabel tersebut.
Penerapan Timbangan Ikan Pintar dalam Meningkatkan Ekonomi UKM Masyarakat Pesisir Berbasis IoT Ardi Gunawan; Sasmitoh Rahmad Riady; Ismasari Nawangsih
Jurnal Tekno Insentif Vol 16 No 1 (2022): Jurnal Tekno Insentif
Publisher : Lembaga Layanan Pendidikan Tinggi Wilayah IV

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36787/jti.v16i1.695

Abstract

Abstrak Penimbangan merupakan sumber data yang akan menjadi acuan hasil tangkapan nelayan guna menentukan harga penjualan ikan. Tempat Pelelangan Ikan (TPI) yang masih menggunakan timbangan secara manual untuk keperluan transaksi penjualan tidak terdapat data catatan penimbangan, mengakibatkan banyak nelayan yang mengalami kerugian dan kesulitan dalam pengembalian modal kepada pihak Koperasi Perikanan Laut (KPL). Timbangan Ikan Pintar adalah sebuah sistem timbangan online yang berfungsi untuk menimbang ikan hasil tangkapan nelayan yang akan dikalkulasi dengan harga yang dapat diakses melalui website dan aplikasi mobile yang berbasis IoT. Penerapan Timbangan Ikan Pintar mempermudah dalam mencatat data transaksi peminjaman, pengembalian pinjaman dan penjualan. Penggunaan IoT juga dapat meningkatkan efektifitas dan efisiensi pada saat penimbangan ikan. Abstract Weighing is a source of data that will become a reference for fishermen's catches to determine the selling price of fish. Fish auction places (TPI) which still use manual scales for the purposes of sales transactions have no weighing records, resulting in many fishermen experiencing losses and difficulties in returning capital to the marine fisheries cooperative (KPL). Smart Fish Scales is an online weighing system that functions to weigh fish caught by fishermen which will be calculated at prices that can be accessed through websites and mobile applications based on IoT. The application of Smart Fish Scale makes it easier to report loan transaction data, loan repayments and sales. The use of IoT can also increase effectiveness and efficiency when weighing fish.
PERBANDINGAN ALGORITMA LINEAR REGRESSION, LSTM, DAN GRU DALAM MEMPREDIKSI HARGA SAHAM DENGAN MODEL TIME SERIES Khalis Sofi; Aswan Supriyadi Sunge; Sasmitoh Rahmad Riady; Antika Zahrotul Kamalia
PROSIDING SEMINASTIKA Vol 3 No 1 (2021): 3rd SEMINASTIKA 2021
Publisher : Universitas Mulia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47002/seminastika.v3i1.275

Abstract

Penelitian ini bertujuan untuk memprediksi harga saham dengan membandingkan algoritma Linear Regression, Long Short-Term Memory (LSTM), dan Gated Recurrent Unit (GRU) dengan dataset publik kemudian menentukan performa terbaik dari ketiga algoritma tersebut. Dataset yang diuji bersumber dari Indonesia Stock Exchange (IDX), yaitu dataset harga saham KEJU berbentuk time series dari tanggal 15 November 2019 sampai dengan 08 Juni 2021. Parameter yang digunakan untuk pengukuran perbandingan adalah RMSE (Root Mean Square Error), MSE (Mean Square Error), dan MAE (Mean Absolute Error). Setelah dilakukan proses training dan testing, dihasilkan sebuah analisis bahwa dari hasil perbandingan algoritma yang digunakan, algoritma Gated Recurrent Unit (GRU) memiliki performance paling baik dibandingkan Linear Regression dan Long-Short Term Memory (LSTM) dalam hal memprediksi harga saham, dibuktikan dengan nilai RMSE, MSE, dan MAE dari uji coba GRU paling rendah, yaitu nilai RMSE 0.034, MSE 0.001, dan nilai MAE 0.024.
Prediction of Electrical Energy Consumption Using LSTM Algorithm with Teacher Forcing Technique Sasmitoh Rahmad Riady; Tjong Wan Sen
JISA(Jurnal Informatika dan Sains) Vol 4, No 1 (2021): JISA(Jurnal Informatika dan Sains)
Publisher : Universitas Trilogi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31326/jisa.v4i1.904

Abstract

Electrical energy is an important foundation in world economic growth, therefore it requires an accurate prediction in predicting energy consumption in the future. The methods that are often used in previous research are the Time Series and Machine Learning methods, but recently there has been a new method that can predict energy consumption using the Deep Learning Method which can process data quickly for training and testing. In this research, the researcher proposes a model and algorithm which contained in Deep Learning, that is Multivariate Time Series Model with LSTM Algorithm and using Teacher Forcing Technique for predicting electrical energy consumption in the future. Because Multivariate Time Series Model and LSTM Algorithm can receive input with various conditions or seasons of electrical energy consumption. Teacher Forcing Technique is able lighten up the computation so that it can training and testing data quickly. The method used in this study is to compare Teacher Forcing LSTM with Non-Teacher Forcing LSTM in Multivariate Time Series model using several activation functions that produce significant differences. TF value of RMSE 0.006, MAE 0.070 and Non-TF has RMSE and MAE values of 0.117 and 0.246. The value of the two models is obtained from Sigmoid Activation and the worst value of the two models is in the Softmax activation function, with TF values is RMSE 0.423, MAE 0.485 and Non-TF RMSE 0.520, MAE 0.519. 
Water Quality Monitoring System with Parameter of pH, Temperature, Turbidity, and Salinity Based on Internet of Things Yazi Adityas; Muchromi Ahmad; Moh Khamim; Khalis Sofi; Sasmitoh Rahmad Riady
JISA(Jurnal Informatika dan Sains) Vol 4, No 2 (2021): JISA(Jurnal Informatika dan Sains)
Publisher : Universitas Trilogi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31326/jisa.v4i2.965

Abstract

This research aims to monitor the quality of water used for aquariums. The physical parameters used are water pH, water temperature, water turbidity, and water salinity. Using a pH sensor, temperature sensor, turbidity sensor, and salinity conductivity sensor with Arduino as the controller. The prototype method used in this research, starting from the formulation, research, building stages to testing and evaluating the results of the research. The working process of the system is when the system is activated, the sensors will detect and capture the amount of value contained in the water, then the data from the sensor is sent to a database in the cloud using an ethernet shield that is connected to the media router as a liaison for the internet network then displayed on the website dashboard in the form of graphs and monitoring record tables in real time. The sensors function to detect water quality, where quality standards have been set in this system, namely temperature standards of 27-30°C, pH standards of 7.0-8.0, turbidity standards of 2.5-5 ntu, and salinity of 20-28 ppt. If the sensor detects non-compliance with water quality standards, the buzzer in this system will sound. From the results of system testing, sensors can detect water quality in real time within 5-10 seconds. Based on the research results, this water quality monitoring system is effective to help ensure the quality of the water in the aquarium so that it always meets the standards.
Selection of Feature Driven Development (FDD) Model in Agile Method for Developing Information System of Mosque Management Sasmitoh Rahmad Riady; Khalis Sofi; Jafar Shadiq; Rita Wahyuni Arifin
Journal of Computer Networks, Architecture and High Performance Computing Vol. 4 No. 2 (2022): Article Research Volume 4 Number 2, July 2022
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v4i2.1469

Abstract

This paper discusses about software development for building a project using the Feature Driven Development (FDD) model contained in the agile method. As in other agile methods, the Feature Driven Development model has additional properties to implement functions and needs in a short iteration. In terms of the characteristics contained in several other agile models almost all of them have similarities so that makes the stakeholder confusion in determining which model will be chosen as a method of software development. This study focuses on the search for a number of suitable project specifications in the selection of Feature Driven Development models for software development. From the research that has been done on several papers there are several aspects of design and construction in software development by emphasizing the quality and high level of features in developing a software using the Feature Driven Development model. In facilitating the results of the Feature Driven Development model, we will provide a case study of Mosque Management Information System which has several features, such as content management, information of prayer time, online reading Qur’an, and Petty Cash Mosque management. This case is implemented by Feature Driven Development model with short iteration, because this project had done in several months
Stock Price Prediction using Prophet Facebook Algorithm for BBCA and TLKM Sasmitoh Rahmad Riady
International Journal of Advances in Data and Information Systems Vol. 4 No. 1 (2023): April 2023 - International Journal of Advances in Data and Information Systems
Publisher : Indonesian Scientific Journal

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25008/ijadis.v4i2.1258

Abstract

Stocks are an investment instrument that is starting to be in great demand by the public today. However, stock prices are fluctuating, making people feel doubts about when they are going to invest. To overcome these doubts, we need a way to predict stock prices. This study aims to predict stock price fluctuations using Facebook's Prophet Algorithm to help people decide their investment in stock. The research object used is BBCA and TLKM stock price data in the form of a time series from 03 May 2021 to 28 April 2022 with stock price testing data for the next week, namely 01 May 2022 to 07 May 2022. From the training and testing process done, a prediction is produced that is very close to the original value. Using the RMSE, MSE and MAE measurements, we get RMSE 49.6, MSE 2462.1 and MAE 37.5 for BBCA and RMSE stocks, namely 21.3, MSE 456.5 and MAE 19.2 for TLKM shares. The conclusion is that Facebook's Prophet Algorithm is suitable for predicting stock prices.