Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal JURNAL INSTEK (Informatika Sains dan Teknologi) Jurnal Varian JES-MAT (Jurnal Edukasi dan Sains Matematika) Pangulu Abdi: Jurnal Pengabdian Kepada Masyarakat
Muh. Irwan
Jurusan Matematika UIN-Alauddin Makassar
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Mathematics Social Sciences Other

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Analisis Kestabilan dan Usaha Pemanenan Model Predator Prey Tipe Holling III dengan Keuntungan Maksimum Didiharyono Didiharyono; Muh Irwan
Jurnal Varian Vol 2 No 2 (2019)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v2i2.373

Abstract

In this paper discussed Stability Analysis and Harvesting Effort at second Predator Prey Populations model Holling Type III with Maximum Profit. The step this research is to determine the equilibrium point, linearize the model, stability analysis of the equilibrium point, and numerical simulation. Result shows that obtained an interior point T that asymptotic stable based on Hurwitz stability test then obtained maximum profit from exploitation harvesting effort of second predator prey populations. This second populations will always exist, even though exploited with harvesting effort done by humans. Harvesting effort of second predator-prey populations given maximum profit that occur on critical points of surface profit function
PENGEMBANGAN BAKAT, KEILMUAN, DAN SEMANGAT BELAJAR SISWA DI SDN LEMBA muh irwan; Hasbi Ibrahim; Rezki Awalia; Muhben Muktar; Nahda Syaidah Marhabatsar; Alda Robiany; Moh. Hidayat
Pangulu Abdi: Jurnal Pengabdian Kepada Masyarakat Vol 2 No 1 (2022): Pangulu Abdi : Jurnal Pengabdian Kepada Masyarakat
Publisher : PPM LP2M UIN Alauddin Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/pangabdi.v2i1.34291

Abstract

Porseni or commonly referred to as sports and arts week, is a forum where we can channel interests, talents and hobbies, especially students, with the aim of strengthening student relationships both from seniors to juniors and teachers, apart from that, the purpose of this research is to train talent development. and enthusiasm for students learning, especially in SD 03 Lemba. This study uses a qualitative approach with a Phenomenological and Observational research model. The findings of this study are as follows: 1) The condition of students’ skills is different because it is influenced by the family because it is influenced by the family environment, especially in the religious field. 2)There is a KKN student work program that can develop students’ scientific talent skills and enthusiasm for learning, including art porn. 3) KKN students hold this activity to develop students’ scientific talent skills and enthusiasm for learning which is proven through changing student behavior to be active and better. 4) The obstacle experienced by the activity was the lack of competition, so the competition was limited because it was still during the covid-19 pandemic. This work program activity was carried out for four days and the efforts made were by good communication, making the concept of a competition about religion interesting, continuing to carry out health protocols, and closing the prosenin activity followed by vaccination in collaboration with Mrs. Bhayangkari, Soppeng Regency. This work program was successfully carried out with full support by education and education staff and was closely monitored by the Covid-19 sitgam Team in Soppeng Regency.
KLASIFIKASI MACHINE LEARNING UNTUK ANEMIA MENGGUNAKAN METODE SUPPORT VECTOR MACHINE DAN RANDOM FOREST Nur Wahida , Zam; Irwan; Irwan, Muh.
Jurnal Edukasi dan Sains Matematika (JES-MAT) Vol. 11 No. 1 (2025): Jurnal Edukasi dan Sains Matematika (JES-MAT)
Publisher : Department of Mathematics Education, Universitas Kuningan

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

Abstract

This research involves the classification of anemia using the support vector machine (SVM) and random forest (RF) methods. SVM has the ability to separate two classes of data by finding the optimal hyperplane. On the other hand, Random Forest can handle incomplete or missing data, reduce overfitting, minimize errors, and efficiently handle large training datasets. The classification of anemia in this study uses a 70% training data and 30% testing data split. The accuracy rate of classification using the SVM method is 94%. In comparison, the classification using the RF method achieves an accuracy rate of 98%.
MEMPREDIKSIKAN HARGA SAHAM BANK RAKYAT INDONESIA (PERSERO) TBK MENGGUNAKAN METODE BIDIRECTIONAL LONG SHORT TERM MEMORY (BI-LSTM) Darja, Rahma; Irwan; Muh. Irwan
Jurnal INSTEK (Informatika Sains dan Teknologi) Vol 10 No 2 (2025): OCTOBER
Publisher : Department of Informatics Engineering, Faculty of Science and Technology, Universitas Islam Negeri Alauddin, Makassar, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24252/instek.v10i2.57438

Abstract

Pasar modal memegang peranan strategis sebagai indikator stabilitas ekonomi suatu negara, di mana saham berfungsi sebagai instrumen vital untuk pertukaran modal yang memengaruhi berbagai sektor keuangan. Mengingat sifat harga saham yang fluktuatif dan non-linear, prediksi yang akurat menjadi sangat krusial bagi investor untuk meminimalisir risiko kerugian. Penelitian ini bertujuan untuk memprediksi harga saham tertinggi (high price) pada PT Bank Rakyat Indonesia (Persero) Tbk menggunakan pendekatan Deep Learning, yaitu metode Bidirectional Long Short-Term Memory (Bi-LSTM). Dalam metodologi penelitian, skenario pembagian data dilakukan dengan komposisi 50% data latih dan 50% data uji. Penelitian ini melakukan serangkaian eksperimen tuning hyperparameter yang meliputi variasi learning rate, jumlah neuron, dan jumlah epoch. Berdasarkan hasil pengujian, arsitektur model paling optimal diperoleh dengan konfigurasi 50 neuron, penggunaan algoritma optimasi Adam, learning rate sebesar 0,001, batch size 1, dan 5 kali epoch. Evaluasi kinerja model menunjukkan hasil yang sangat impresif, dengan nilai Mean Absolute Percentage Error (MAPE) sebesar 1,24998952%. Mengacu pada standar evaluasi di mana nilai MAPE < 10%, kemampuan prediksi model ini dikategorikan sebagai "sangat baik". Hasil ini mengindikasikan bahwa model Bi-LSTM yang dibangun sangat andal dan dapat dimanfaatkan oleh investor sebagai acuan strategis dalam pengambilan keputusan investasi saham BBRI.
Co-Authors Alda Robiany Darja, Rahma Didiharyono, D. Hasbi Ibrahim Irwan Irwan Kasse Moh. Hidayat Muhben Muktar Nahda Syaidah Marhabatsar Nur Wahida , Zam Rezki Awalia