Claim Missing Document
Check
Articles

Found 6 Documents
Search

THE ARTIFICIAL NEURAL NETWORK (ANN) ALGORITHM IMPLEMENTATION FOR PREDICTING THE AMOUNT OF BOOK SALES AT ERLANGGA PUBLISHER PEMATANGSIANTAR HOTMALINA SILITONGA; INDRA GUNAWAN; BAHRUDI EFENDI DAMANIK
INTERNATIONAL JOURNAL OF MULTI SCIENCE Vol. 1 No. 12 (2021): INTERNATIONAL JOURNAL OF MULTISCIENCE - MARCH 2021 EDITION
Publisher : CV KULTURA DIGITAL MEDIA

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

Abstract

Selling is one of the main goals of a company after producing its goods. The more goods sold, the more economic value the company is selling. Therefore, the purpose of this study is to determine how much the rate of increase or decrease in the number of book sales at the publisher of Erlangga Pematangsiantar is in the form of prediction. This study uses an Artificial Neural Network (ANN) with the Backpropagation method. Backpropagation is a method that is often used for prediction. The research data is secondary data (sales data) sourced from PT. Publisher Erlangga Pematangsiantar from 2013 to 2017. Data is divided into 2 parts, namely training data and testing data. There are 5 architectural models used in this study, 3-9-1, 3-11-1, 3-15-1, 3-30-1, and 3-31-1. Of the 5 (five) architectural models used, the best architecture is 3-11-1 with an accuracy rate of 80% and MSE 0.13001601. So this model is good for predicting the number of book sales at PT. Publisher Erlangga Pematangsiantar.
Utilization of K-Medoids Algorithm for Klustering of Oil Palm Sprouts Sri Nuraini; Indra Gunawan; Widodo Saputra
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 1 (2022): March
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (807.895 KB) | DOI: 10.55123/jomlai.v1i1.160

Abstract

Palm oil is still a prima donna commodity in the plantation sector and as a major foreign exchange earner to date. Research and development of this commodity is very important to maintain Indonesia's position as the largest palm oil producing country in the world. The purpose of this study was to analyze what internal and external factors are the strengths, weaknesses, opportunities and threats for marketing oil palm sprouts in PPKS Marihat. To analyze what are the priority strategies to be implemented for the marketing of sprouts at PPKS Marihat. The research method used is the K-Medoids clustering algorithm by selecting the sprout data in order to determine the best quality of sprouts. Based on the results of research using the K-Medoids algorithm with manual calculations and testing, the same results were obtained, namely cluster 1 with very good sprouts category had 7 members, cluster 2 with good sprouts category had 12 members and cluster 3 with poor sprouts category had 7 members. . Testing data on Rapid Miner using the K-Medoids algorithm can display 3 classes with an accuracy percentage of 100%. So it can be concluded that the K-Medoids algorithm can be used for clustering oil palm sprouts at PPKS Marihat.
Implementation of Data Mining Algorithm for Clustering of Palm Oil Harvested Data Widya Juli Mawaddah; Indra Gunawan; Ika Purnama Sari
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 1 (2022): March
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (875.644 KB) | DOI: 10.55123/jomlai.v1i1.163

Abstract

Palm oil is one of the plantation commodities that has a strategic role in Indonesia's economic development. In this study, we will discuss oil palm yields at PPKS Marihat, one of the Oil Palm Research Center branches located in Simalungun Regency, Medan, North Sumatra. Know how it grows. The Clustering algorithm is used in K-Means. Using this method, the data will be grouped into 3 (three) Clusters, where the application of the K-Means Clustering process uses the Rapid Miner tools. The data used is data on oil palm harvests at PPKS Marihat in 2020, consisting of 100 data items. The results obtained are crop yields with an excellent value of 66 items, harvest data with a good deal of 32 items, and harvest data with a reasonably good value of 2 items, based on net total and gross amount for each region. Based on this, it can be concluded that the K-Means Algorithm can be used to Cluster oil palm yields at PPKS Marihat
Material Sales Clustering Using the K-Means Method Sri Rahayuni; Indra Gunawan; Ika Okta Kirana
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 1 (2022): March
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1469.106 KB) | DOI: 10.55123/jomlai.v1i1.177

Abstract

Along with the increasing growth of technology and the development of science, business competition is also getting faster and therefore we are required to always develop the business in order to always survive in the competition. Family Gypsum is a store whose sales system is the same as a supermarket, namely the buyer will take the goods to be purchased himself. From this, data on sales, purchases of goods, and unexpected expenses are not structured properly so that the data only functions as an archive. In this research, data mining is applied using the K-Means calculation process which provides a standard process for using data mining in various fields to be used in clustering because the results of this method are easy to understand and interpret. The results obtained from the K-Means method that has been implemented into Rapid Miner have the same value, which produces 3 clusters, namely clusters that do not sell, clusters that sell, and clusters that sell very well. With red clusters with 2 items, the clusters selling green with 28 items, the clusters selling with blue with 30 items. The results of this study can be entered into the Family Gypsum store Jl. H. Ulakma Sinaga, Red Rambung who is getting more attention on each sale based on the cluster that has been done
Optimization of Computer Network Security System Against Malware Attacks Using Firewall Filtering with Port Blocking Method Andri Andri; Indra Gunawan; Ika Okta Kirana
JOMLAI: Journal of Machine Learning and Artificial Intelligence Vol. 1 No. 2 (2022): June
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (860.506 KB) | DOI: 10.55123/jomlai.v1i2.816

Abstract

Computer networks have an important role in teaching and learning activities in schools, but there are also negative impacts. One of them is prone to attack by malware such as viruses and so on, as is the case at the Satria Mandiri Private Vocational School in Bandar Huluan. So far, the computer network at the school is very easy to attack by malware. The negative impact of malware on the network is bandwidth traffic overload, causing bandwidth constraints to run out quickly or data transfer traffic to be slower than usual. The reliability of a network can be determined from the security factor of the network itself. Some routers have firewall settings that are quite capable but need to be managed more specifically based on the needs of the 1500 Kbps network scale and available bandwidth. Creating good rules in the firewall will make it easier to filter network traffic and bandwidth so that it can create security and convenience for network and bandwidth users. Port blocking allows users or users to interact with the proxy server on the local network, where the connected user has gone through verification that can filter malware activity with embedded rules
Clustering Production of Plantation Crops by Province Using the K-Means Method Azhari Abdillah Simangunsong; Indra Gunawan; Zulaini Masruro Nasution
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 (683.427 KB) | DOI: 10.55123/jomlai.v1i4.1661

Abstract

The purpose of this research is to classify the results of plantation crop production each year based on provinces in Indonesia, so that it can be known which provinces produce the most plantation crop production and which produce less. In this study using the K-Means Algorithm Data Mining technique. The data source for this research was collected based on plantation data obtained from the Indonesian Central Bureau of Statistics (BPS). The data used is data from 2018-2020 which consists of 34 provinces. The results of this study are groupings which are divided into 3 Clusters, namely low Clusters, medium Clusters, and high Clusters. Based on the results of calculations using the K-Means Algorithm, 6 items (Provinces) were obtained for high Clusters, 2 Provinces for medium Clusters and 27 Provinces for low Clusters. The conclusion that can be obtained is that the grouping of plantation crop production in Indonesia can be solved by applying the K-Means algorithm.