Claim Missing Document
Check
Articles

Found 11 Documents
Search

Estimasi Pemberantasan Hama di Kebun Bah Jambi Menggunakan Algoritma Backpropagation Agung Bimantoro; Sumarno Sumarno; Heru Satria Tambunan
Journal of Computer System and Informatics (JoSYC) Vol 2 No 2 (2021): February 2021
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

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

Abstract

Oil palm, which is the largest palm oil producer, plays an important role in the welfare of the people in Indonesia because it creates many jobs. Oil palm plants cannot be separated from pests. The number of pest attacks on oil palm plants can cause a decrease in fruit production and can even cause the plant to die. In this study, the authors estimated the number of pest attacks in the oil palm plantation Unit Bah Jambi, North Sumatra using the backpropagation algorithm. The data used in this study were obtained directly from the Plantation Unit Bah Jambi plantation. In this study the authors used 5 architectural patterns; 2-10-1, 2-12-1, 2-14-1, 2-16-1, 2-16-1. Of the five architectural patterns used, the best architecture is obtained with an accuracy rate of 75%, 187 epoch iterations in 4 seconds, namely the 2-10-1 architectural pattern
Penerapan JST Backpropagation Untuk Memprediksi Data Penerimaan Mahasiswa Baru Pada Universitas Simalungun Choiril Ichsan Damanik; Dedy Haratama; Heru Satria Tambunan
Bulletin of Computer Science Research Vol. 2 No. 1 (2021): Desember 2021
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

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

Abstract

The growing era of globalization makes people continue their education to a higher level. The greater the interest of prospective students, each year the number of prospective students will increase. The increase has made the university can manage and know the estimated number of prospective students each year. So we need a method to help the university so that the number of prospective students can be predicted quickly and accurately. In predicting the data of prospective new students at Simalungun University the backpropagation method is used, this study is expected to predict the number of new students with smaller error results. The data used was obtained from the Simalungun university administration from 2015 to 2018. Backpropagation is one method that is often used in solving complex problems. Its application researchers conducted the test using the Matlab application. This study uses 5 architectural models: 2-5-1, 2-10-1.2-15-1, 2-25-1, 2-3-5-1, the best accuracy is obtained from architectural models 2-3-5-1 with values accuracy of 86%, epoch 8406 iteration, and MSE namely 0, 0778011336
Analisa Klasifikasi C4.5 Terhadap Faktor Penyebab Menurunnya Prestasi Belajar Mahasiswa Pada Masa Pandemi Dedy Hartama; Agus Perdana Windarto; Heru Satria Tambunan; Irfan Sudahri Damanik
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 4, No 1 (2020): The Liberty of Thinking and Innovation
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v4i1.2695

Abstract

The purpose of the study was to classify the factors causing the decline in student achievement during the pandemic using the C4.5 datamining method. Sources of research data were obtained by conducting interviews and distributing questionnaires to 7th semester students of the 2020-2021 school year information system study program. Attributes that used in the classification of the factors causing the decline in student achievement include: Learning Method (C1), Study Time (C2), Material Understanding (C3), Giving Assignments (C4) and Environment (C5). The results of the calculation show that the Material Understanding (C3) attribute is the attribute that most influences the decline in student learning achievement. Testing was also carried out using the help of Rapidminer software and obtained an accuracy of 97.5%.Keywords: Classification, Datamining, C4.5, learning achievement, Pandemic
Implementasi Inferensi Fuzzy Tsukamoto pada Prediksi Penjualan Telur Ayam Eropa pada Bisnis Raffa Telur Cici Astria; Harly Okprana; Anjar Wanto; Dedy Hartama; Heru Satria Tambunan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 4, No 1 (2020): The Liberty of Thinking and Innovation
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v4i1.2587

Abstract

Eggs are an animal product that comes from poultry. Eggs are known as a food that contains nutrients that are very good for the body because they contain a high protein source. Apart from being nutritious, people consume a lot of eggs because the price is relatively cheaper than other protein foods. This research was conducted on the Raffa Egg business located in Pematangsiantar City. The data collection process was carried out by means of interviews and observations with Raffa Eggs. This study aims to predict the number of purchases of European chicken eggs from suppliers. The research was conducted using Tsukamoto fuzzy logic with 3 variables, namely sales (x), inventory (y) and purchases (z). where the sales variable (x) consists of 2 fuzzy sets, including increasing and decreasing, inventory (y) consisting of many and few fuzzy sets and purchasing (z) consisting of many and few fuzzy sets. The results of the calculation of the prediction of the number of purchases of European chicken eggs with sales of 6500 and inventory of 25 000 are 29583 items.Keywords: Eggs, European chicken eggs, Fuzzy, Tsukamoto, Pematangsiantar
Penerapan Data Mining Dalam Mengelompokkan Calon Penerima Beasiswa Dengan Menggunakan Algoritma K-Means Nur Afriani Manihuruk; Muhammad Zarlis; Eka Irawan; Heru Satria Tambunan; Irawan Irawan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 4, No 1 (2020): The Liberty of Thinking and Innovation
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v4i1.2575

Abstract

Penelitian ini bertujuan untuk mencari pengelompokkan data yang ada pada siswa yang berhak menerima beasiswa. Penyaluran beasiswa yang berasal dari keluarga yang kurang mampu harus dapat melalui seleksi yang melibatkan kriteria-kriteria tertentu. Kriteria tersebut seperti kondisi rumah, nilai raport, status rumah. Algoritma K-Means dapat membantu untuk mengklasifikasi siswa-siswi yang sangat layak untuk berupa mendapatkan bantuan berupa beasiswa. Adapun tujuan yang ada pada penelitian ini adalah menentukan clustering penerima beasiswa sehingga dapat memberikan rekomendasi yang layak, layak dengan pertimbangan dan kurang layak untuk menerima beasiswa dengan 4 kriteria. Data set yang digunakan sebanyak 128 siswa yang berasal dari sekolah SMP Muhammadiyah 54 Kerasaan. Data-data tersebut dapat dihitung dengan menggunakan algoritma K-Means dan pengujian dapat dilakukan melalui aplikasi RapidMiner 5.3. Metode K-Means berusaha mengelompokkan data yang ada kedalam beberapa kelompok, dimana data dalam satu kelompok mempunyai karakteristik yang sama satu sama lainnya dan mempunyai karakteristik yang berbeda dengan data yang ada didalam kelompok yang lain. Hasil penelitian diperoleh C1:73Item, C2:30Item, C3:25Item. Dari hasil analisis diharapkan dapat membantu pemahaman siswa yang berhak menerima beasiswa. Kata kunci: Data Mining, Metode K-Means, Pengelompokkan Penerima Beasiswa
Penerapan Metode Fuzzy Mamdani Dalam Penentuan Penerima BLT-DD Di Mekar Sari Raya Muhammad Aliyul Amri; Dedy Hartama; Anjar Wanto; Sumarno Sumarno; Heru Satria Tambunan
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 4, No 1 (2020): The Liberty of Thinking and Innovation
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v4i1.2698

Abstract

Abstract−BLT-DD, which stands for Village Fund Direct Cash Assistance, is a village assistance program with the type of cash or assistance or various other types of assistance, whether conditional or unconditional for underprivileged people. The selection of BLT-DD recipients needs to be done accurately so that recipients who really need will receive BLT-DD. To be able to determine the BLT-DD recipient, one of artificial intelligence is needed, namely fuzzy mamdani logic which has a simple structure using min-max or max-product operations with a set of predetermined rules, namely IF-AND-THEN. Then supported by the Matlab application (R2007b) as a decision maker through the program provided. The results of the calculation of BLT-DD recipients can later assist the Village Head and staff staff in determining more accurate BLT-DD recipients.Keywords: Fuzzy, Mamdani, Receiver, BLT-DD, Matlab[1]        D. Rahakbauw, F. Rianekuay dan Y. Lesnussa, “Penerapan Metode Fuzzy Mamdani Untuk Memprediksi Jumlah Produksi (Studi Kasus : Data Persediaan Dan Permintaan Produksi Karet Pada PTP Nusantara XIV (Persero) Kebun Awaya, Teluk Elpaputih, Maluku-Indonesia,” Jurnal Ilmiah Matematika dan Terapan (JIMT), pp. 119-127, 2019. [2]        Z. Julisman dan Erlin, “Prediksi Tingkat Curah Hujan di Kota Pekanbaru menggunakan Logika Fuzzy Mamdani,” Jurnal SATIN (Sains dan Teknologi Informasi), vol. 3, pp. 65-72, 2014. [3]        N. Ningsih, N. T. Pambudi dan A. M. Abadi, “Penerapan Metode Fuzzy Mamdani Untuk Memprediksi Penjualan Gula,” Seminar Matematika Dan Pendidikan Matematika UNY, pp. 153-160, 2017. [4]        B. Prasetya, A. B. Setiawan dan B. F. Hidayatulail, “Fuzzy Mamdani Pada Tanaman Tomat Hidroponik (Mamdani Fuzzy on Hydroponics Tomato Plants),” Journal of Electrical and Electonic Engineering (UMSIDA), vol. 3, pp. 228-263, 2019. [5]        M. S. Asih, “Sistem Pendukung Keputusan Fuzzy Mamdani pada Alat Penyiraman Tanaman Otomatis,” Jurnal Sistem Informasi (QUERY), vol. 2, pp. 41-52, 2018. [6]        M. Y. T. Irsan, M. I. Kasau dan I. P. Simbolon, “Penggunaan Fuzzy Logic & Metode Mamdani untuk Menghitung Pembelian, Penjualan dan Persediaan,” Journal of Applied Accounting and Finance (JAAF), vol. 3, pp. 37-48, 2019.
Penerapan Data Mining Dalam Mengelompokkan Provinsi Rawan Kejahatan Menggunakan Algoritma K-Means Eka Desriani Aritonang; Heru Satria Tambunan; Jaya Tata Hardinata; Eka Irawan; Dedi Suhendro
KOMIK (Konferensi Nasional Teknologi Informasi dan Komputer) Vol 4, No 1 (2020): The Liberty of Thinking and Innovation
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/komik.v4i1.2576

Abstract

Kemajuan Teknologi informasi saat ini berkembang sangat cepat yang mengakibatkan peningkatan pada data dalam jumlah besar. Meningkatnya jumlah kejahatan pada setiap provinsi di Indonesia menyebabkan penumpukan pada data. Beragam jenis kejahatan terjadi di lingkungan masyarakat, seperti pembunuhan, penganiayaan, pemerkosaan, pencurian, penipuan, penyalahgunaan narkoba, dan perjudian. Dengan melihat banyaknya jumlah kejahatan tersebut, masyarakat menjadi khawatir dan merasa tidak nyaman sehingga perlu dilakukan penelitian agar dapat mengetahui wilayah/provinsi yang rawan akan kejahatan. Tujuan dari penelitian yaitu sebagai referensi bagi pemerintah untuk meningkatkan keamanan untuk setiap wilayah pada tahun-tahun berikutnya. Penelitian ini menggunakan metode data mining dengan algoritma k-means clustering dan dibantu dengan aplikasi Rapidminer. Penelitian ini mengelompokkan provinsi dengan dua cluster yaitu cluster tinggi dan cluster rendah. Hasil dari penelitian ini diperoleh 4 provinsi dengan jumlah kejahatan tertinggi (C1), 30 provinsi dengan jumlah kejahatan rendah (C2) dan pengujian menggunakan Rapidminer mendapatkan hasil yang sama. Algoritma K-means dapat diterapkan dan memberikan informasi tentang provinsi yang rawan terjadinya kejahatan.Kata kunci: k-means, clustering, kejahatan
Penerapan Algoritma K-Means Clustering Pada Penyebaran Penyakit Infeksi Saluran Pernapasan Akut (ISPA) di Provinsi Riau Ninaria Purba; Poningsih Poningsih; Heru Satria Tambunan
Journal of Information System Research (JOSH) Vol 2 No 3 (2021): April 2021
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (421.773 KB)

Abstract

Health is a valuable thing for humans because anyone can experience health problems, as well as humans who are very susceptible to various kinds of diseases but we don't realize the cause. The K-means algorithm is not affected by the order of objects used, this is proven when the author tries to randomly determine the starting point of the cluster center of one of the objects at the beginning of the calculation. The resulting number of cluster membership is the same when using another object as the starting point of the center of the cluster. However, this only affects the number of iterations performed. Object clustering (object clustering) is a process of object mining which aims to partition existing objects into one or more object clusters based on their characteristics. This study examines how to use the K-means Cluster Analysis Algorithm in case studies of human infectious diseases, namely Acute Respiratory Infection. This study examines the K-means Cluster Analysis method in Acute Respiratory Infection based on a set of variables established per municipality in Riau Province
Data Mining Menggunakan Metode Asosiasi Apriori untuk Merekomendasi Pola Obat Pada Puskesmas Dewinta Marthadinata Sinaga; Agus Perdana Windarto; Heru Satria Tambunan; Irfan Sudahri Damanik
Journal of Information System Research (JOSH) Vol 3 No 2 (2022): Januari 2022
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (540.994 KB) | DOI: 10.47065/josh.v3i2.1237

Abstract

Drugs are one of the most important components in terms of health, both to cure and reduce pain due to illness suffered by everyone, besides that the use of drugs also gives us information about what diseases everyone suffers so that the information is very helpful for health workers. For this reason, drugs need to be managed properly, effectively and efficiently. This study aims to analyze the a priori algorithm on drug output data at the Parsoburan Health Center Pematangsiantar to find out what types of drugs are most needed by patients at the same time. The data used is in the form of drug output data in April 2021. Based on the a priori algorithm calculations, 70 association rules were formed with a number minimum of support 90% and a minimum confidence of 90%. It is hoped that the results of the research can help the Parsoburan Health Center Pematangsiantar optimize quality health services for planning future drug needs and produce useful information for decision making.
Analisis Metode K-Medoids Cluster Dalam Mengelompokkan Siswa Yang Berprestasi Indri Fatma; Heru Satria Tambunan; Fitri Rizki
Bulletin of Informatics and Data Science Vol 1, No 1 (2022): May 2022
Publisher : PDSI

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

Abstract

Cluster analysis of outstanding students using data mining, namely the K-Medoid Cluster Algorithm. Previously, the school still used the manual method in determining students who excel at the school, so it took a long time and the results were not accurate. K-Medoid Cluster is one of the algorithms used for data classification or grouping, the authors apply the K-Medoid Cluster algorithm in grouping students with high achievement in order to get more accurate, fast, and effective results.