Garuda - Garba Rujukan Digital

PENERAPAN MACHINE LEARNING UNTUK MENENTUKAN KELAYAKAN KREDIT MENGGUNAKAN METODE SUPPORT VEKTOR MACHINE Syafi'i Syafi'i; Odi Nurdiawan; Gifthera Dwilestari
JURSIMA (Jurnal Sistem Informasi dan Manajemen) Vol 10 No 2 (2022): Jursima Vol. 10 No. 2, Agustus Tahun 2022
Publisher : INSTITUT TEKNOLOGI DAN BISNIS INDOBARU NASIONAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47024/js.v10i2.422

Credit is one of the services provided by banks, credit risk that occurs in the provision of credit loans, in the case that the customer is unable to pay the loan received is always considered by the bank, and supervises the customer to reduce risk. The main risk for banks and financial institutions is to differentiate creditors who have the potential for bad loans, this crisis is a concern for financial institutions about credit risk. SUPPORT VEKTOR MACHINE algorithm is an algorithm used to form a decision tree. The decision tree is a very powerful and well-known classification and prediction method. The richer the information or knowledge contained by the training data, the accuracy of the decision tree will increase. The SUPPORT VEKTOR MACHINE algorithm classification method can determine the credit worthiness of the national civil capital capitals as evidenced by the performance table data consisting of the AUC results, Acuracy results. The results of the application of machine learning using the vector machine support algorithm against cooperative data in KPRI "RUKUN" SMKN 1 Lemahabang to determine creditworthiness based on the results of the Performance Vector from the Support Vector Machine algorithm resulted in smooth prediction, smooth true 130, prediction of jammed, true jam 72, current prediction true jam 41, prediction of jammed true jam 332. The accuracy rate of the performance vector of the support vector algorithm is 80.34%. .

RANCANG BANGUN SISTEM INFORMASI PENDATAAN INVENTARIS BERBASIS WEB PADA SERVER PT JASA MARGA (PERSERO) TBK. CABANG PALIKANCI nur syarief abdullah; Arif Rinaldi Dikananda; Saeful Anwar; Odi Nurdiawan
JURSIMA (Jurnal Sistem Informasi dan Manajemen) Vol 10 No 3: Jursima Vol.10 No.3
Publisher : INSTITUT TEKNOLOGI DAN BISNIS INDOBARU NASIONAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47024/js.v10i3.464

PT Jasa Marga (Persero) Tbk. The Palikanci branch is a toll gate support company, but computer Inventory Data Collection which includes servers and is still considered manual data collection so that it consumes a lot of paper and is very vulnerable to being lost and takes a lot of time to collect lost or torn data. To emphasize and learn in understanding the problems as described, the formulation of the problem that researchers can explain is to design a computerized information system for data collection of Jasa Marga's server inventory, create a database for data collection of server inventory items for managers to carry out their work. The purpose of the research is to find out, develop and create an ongoing asset collection application system into the PHP and HTML programming language using the MySQL database. So that researchers can draw conclusions in processing inventory data collection server by implementing applications that have been designed and built in a systematic and structured manner, so that the level of damage in the process of carrying out data collection for seafarers can be resolved.

ANALISA PEMBELIAN SEPEDA MENGGUNAKAN ALGORITMA APRIORI PADA TOKO SEPEDA BRADEN BIKE Dicky Miftakhul Rizki; Odi Nurdiawan; Saeful Anwar
JURSIMA (Jurnal Sistem Informasi dan Manajemen) Vol 10 No 3: Jursima Vol.10 No.3
Publisher : INSTITUT TEKNOLOGI DAN BISNIS INDOBARU NASIONAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47024/js.v10i3.465

The store is a place for trading activities that provide all daily necessities with a special type of goods. Braden Bike Shop is a store that sells a variety of bicycle products and accessories, but the data collection of sales transactions for goods that have been sold is usually written on sheets of paper and collected paper that has been sold and rewrites items that have been sold manually to new paper to record sales reports every month with the current system, The purpose of this study is to find the rules of the combination of items by looking at the relationships of two or more variables, The method used is the A priori Algorithm Method in data mining techniques, namely the association rule or association rule used using a minimum support of 10% and a minimum of confidence of 50%, The results obtained are 12 rules 2 itemsets and 2 rules 3 itemssets following sales for 1 year using a priori algorithms, namely categories Aviator_GN, Exotic_GN, Interbike_GN, Fastron_GN, Polygon_GN, Seat Covers, Anti-Slip_AS Grips and Bell_AS. Results obtained based on manual calculations and using Rapid Miner software have results above the minimum support of 10%and confidence of 50%.

RANCANG BANGUN SISTEM INFORMASI PERSEDIAAN BARANG BERBASIS WEB PADA PT PARAGON FURNITAMA INDUSTRY Arif Rinaldi Dikananda; Shofian Yunus; Saeful Anwar; Odi Nurdiawan
JURSIMA (Jurnal Sistem Informasi dan Manajemen) Vol 10 No 3: Jursima Vol.10 No.3
Publisher : INSTITUT TEKNOLOGI DAN BISNIS INDOBARU NASIONAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47024/js.v10i3.474

PT. Paragon Furnitama Industry is one of the companies engaged in the production of fabric and leather sofas, back seats, and chair cushions, at this time the inventory process is still done manually because it still uses records in books and Microsoft Excel, the process sometimes finds several problems including data redundancy, discrepancies in stock of goods with records, and providing long reports because data validation is needed first. So that the information received by the parties concerned is very difficult to obtain quickly. To emphasize and study the problems as described, the problem formulation that researchers can explain is to design an inventory information system so that the company's performance is getting better. The Design and Build of this Goods Inventory Information System is built based on a website. The design of the information system uses the Software Development Life Cycle (SDLC) with the waterfall method so that this design system is expected to improve performance and performance, especially those related to processing inventory data to making inventory reports at PT. Paragon Furnitama Industry.

Irvan Himawan PREDIKSI HARGA SAHAM DENGAN ALGORITMA REGRESI LINIER DENGAN RAPIDMINER Irvan Himawan; Odi Nurdiawan; Gifthera Dwilestari
JURSIMA (Jurnal Sistem Informasi dan Manajemen) Vol 10 No 3: Jursima Vol.10 No.3
Publisher : INSTITUT TEKNOLOGI DAN BISNIS INDOBARU NASIONAL

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47024/js.v10i3.475

Stock investment in the capital market is very important for every company in the world. Stock prices in the capital market move very randomly, the highs and lows of stock prices are influenced by many factors. Therefore, it is necessary to predict the stock price so that it can help investors to see investment prospects in the future. In this study, the prediction of the stock price of BRI Bank with the BBRI stock code will be carried out, using an algorithm, namely Linear Regression on rapid miners. This Linear Regression Algorithm is the best algorithm to use because it is the most complex compared to other algorithms. Based on signaling theory, which are information signals needed by investors, the value of forecasting results that have been obtained can be used to consider investors' decisions that the stock has high or low risk in the future. Based on the theory of risk, this forecasting analysis helps investors to minimize losses. Stock prediction is one of the technical analysis. Stock buying and selling transactions without technicalities are gambling behavior and contain gharar or ambiguity. The impact of not using this technical analysis clearly resulted in transactions containing maisir and gharar which were clearly prohibited. The historical stock data used in the test was obtained from the finance.yahoo.com web page with the category PT. Bank Rakyat Indonesia Tbk, or with the issuer code BBRI shares. What will be used is annual data for the last 5 years in the form of time series accompanied by open, high, low and volume variables as independent variables and close as dependent variables. The algorithm used is multiple linear regression.

ANALISA KANKER PARU PARU DENGAN MENGGUNAKAN ALGORITMA K-NEARST NEIGHBOR Teguh Abdi Mangun; Odi Nurdiawan; Ade Irma Purnamasari
Jurnal Teknik Industri, Sistem Informasi dan Teknik Informatika Vol. 2 No. 2 (2023): Jurnal Tinsika
Publisher : Universitas Bakti Indonesia

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

Kanker paru adalah salah satu jenis kanker yang paling mematikan di dunia. Menurut data dari World Health Organization (WHO), kanker paru merupakan jenis kanker yang paling banyak menyebabkan kematian di seluruh dunia. Karena itu, pengembangan metode klasifikasi yang akurat dan efektif untuk kanker paru sangat penting dalam upaya untuk meningkatkan deteksi dini dan pengobatan yang tepat. Tahapan penelitian analisis kanker paru-paru menggunakan algoritma k-Nearest Neighbor (k-NN), Tahap awal dalam penelitian ini adalah mengumpulkan data terkait kanker paru-paru, Setelah data terkumpul, langkah berikutnya adalah melakukan preprocessing data untuk membersihkan, Pemilihan nilai k, yaitu jumlah tetangga terdekat, merupakan langkah krusial dalam analisis kanker paru-paru, Setelah nilai k terpilih, model K-NN dilatih menggunakan data pelatihan untuk mempelajari hubungan antara atribut dan status kanker paru-paru. Hasil pada penelitian ini yaitu hasil akurasi yang didapat yaitu sebesar 80.40%.

Peningkatan Pemahaman Akuntansi Dengan Menggunakan Software Zahir Fidya Arie Pratama; Odi Nurdiawan
Edunomic : Jurnal Ilmiah Pendidikan Ekonomi Fakultas Keguruan dan Ilmu Pendidikan Vol 7 No 2 (2019): EDISI SEPTEMBER
Publisher : FKIP Unswagati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33603/ejpe.v7i2.2551

STMIK IKMI have a student academic test when the students enter to STMIK IKMI, the result of the test is low. Low in this case will explains by categories the students get D 25% C 37% B 0% and all the students can not get A. The research uses kuasi eksperiments method with time series design that collaborate classroom action research. The result of research shows about pre test 1 until 4, the score of pre test 1 about 54,129, the score of pre test 2 about 55,548, the score of pre test 3 about 56,032, the score of pre test 4 about 56,097. Based on Kruskall Walls Test shows Asymp Sig score about 0,986 it means there is no significance differences between student perception for the first time and students understanding the materials. In second part of research the students learn accounting that use zahir accounting software for 6 meetings. In third part of research the students has a post test for 4 meetings and the results are the score of post test 1 62,6456, the results are the score of post test 2 70,065, the results are the score of post test 3 80,032, the results are the score of post test 4 86,742. The research analyze statistic test focus on pre test and post test by Kruskall Wall Test that shows Asymp Sig score about 0,000 it means there is a differences between the result of pre test and post test. This reality shows accounting learning by Zahir software gives the positive effect for improving (upgrading) student understanding about accounting.

Bibliometrik Analysis: Signal Preprocessing Techniques for Kualitas Sinyal Electrogram Odi Nurdiawan; Dadang Sudrajat; Fathurrohman; Ade Rizki Rinaldi
Prosiding SISFOTEK Vol 8 No 1 (2024): SISFOTEK VIII 2024
Publisher : Ikatan Ahli Informatika Indonesia

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

This study explores electroencephalogram (EEG) signal preprocessing techniques used in the early detection and diagnosis of epilepsy, aiming to enhance the quality and reliability of data used in clinical applications. Effective signal preprocessing techniques are crucial for minimizing artifacts and noise, which can obscure critical information in EEG signals. More accurate EEG signal processing allows for the identification of abnormal patterns associated with various neurological conditions, such as epilepsy, which heavily relies on this signal analysis for precise diagnosis. This study conducted a bibliometric analysis using a descriptive approach to identify research trends, geographic distribution, institutional contributions, and key authors in this field. Data was collected from the Scopus database using the keywords "electroencephalogram AND signal AND processing AND epilepsy". The analysis results show a significant increase in the number of publications related to EEG signal preprocessing techniques over the past five years, with major contributions from countries like China, India, and the United States, reflecting the high global interest and focus on this topic. Additionally, deep learning and machine learning techniques emerged as the most dominant methods in this research, indicating future trends in the development of increasingly sophisticated EEG signal processing technologies. The findings also suggest that using techniques such as artificial neural networks, convolutional neural networks (CNN), and deep learning can enhance the accuracy of epilepsy diagnosis and prediction, making a significant contribution to modern clinical practice. Moreover, this study emphasizes the importance of developing and integrating more advanced preprocessing techniques to improve the effectiveness of EEG signal detection and classification, which is expected to enhance diagnostic outcomes and patient management with neurological disorders. This study provides valuable contributions to the development of medical diagnostic technologies, particularly for neurological disorders such as epilepsy, and highlights the need for further research to optimize these techniques for broader clinical application.

PENINGKATAN MODEL ANALISIS SENTIMEN MELALUI ALGORITMA NAIVE BAYES BERDASARKAN DATA KOMENTAR YOUTUBE Syarif Hidayat, Deden; Odi Nurdiawan; Fadhil M.Basysyar; Muhamad Sulaeman
Jurnal Manajemen Informatika dan Sistem Informasi Vol. 8 No. 1 (2025): MISI Januari 2025
Publisher : LPPM STMIK Lombok

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36595/misi.v8i1.1413

Penerapan kebijakan subsidi Bahan Bakar Minyak (BBM) berbasis QR Code untuk kendaraan roda empat, yang dimulai pada 1 Oktober 2024, telah memunculkan berbagai tanggapan dari masyarakat. Kebijakan ini bertujuan untuk memastikan distribusi BBM bersubsidi lebih tepat sasaran, namun menghadapi kritik terutama terkait kompleksitas proses pendaftaran QR Code dan pembatasan kriteria kendaraan yang memenuhi syarat. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat terhadap kebijakan tersebut dengan menggunakan data komentar dari video YouTube berjudul "Simak! Aturan Baru Kriteria Penggunaan BBM Bersubsidi - SIP 02/09" yang diunggah pada kanal YouTube Seputar iNews RCTI. Metode yang digunakan adalah analisis sentimen berbasis algoritma Naive Bayes. Proses preprocessing data mencakup pembersihan teks, tokenisasi, penghilangan stopwords, dan stemming untuk memastikan data yang dianalisis bersih dan terstruktur. Dataset dibagi menjadi data pelatihan (70%) dan data uji (30%) untuk membangun serta mengevaluasi model. Model menunjukkan akurasi sebesar 79,40%, dengan performa yang lebih baik dalam mengenali sentimen negatif dibandingkan positif. Hasil penelitian menunjukkan bahwa mayoritas komentar memiliki sentimen negatif, mencerminkan ketidakpuasan masyarakat terhadap kebijakan ini. Penelitian ini menyoroti pentingnya strategi komunikasi yang lebih efektif dari pemerintah untuk meningkatkan pemahaman dan penerimaan masyarakat terhadap kebijakan yang diimplementasikan. Selain itu, hasil penelitian ini juga membuka peluang untuk pengembangan lebih lanjut dalam pemanfaatan analisis sentimen berbasis komputasi untuk mendukung pengambilan keputusan dalam studi kebijakan publik.

Peningkatan Pemahaman Akuntansi Dengan Menggunakan Software Zahir Fidya Arie Pratama; Odi Nurdiawan
Edunomic : Jurnal Ilmiah Pendidikan Ekonomi Fakultas Keguruan dan Ilmu Pendidikan Vol 7 No 2 (2019): EDISI SEPTEMBER
Publisher : FKIP Unswagati

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33603/ejpe.v7i2.2551

STMIK IKMI have a student academic test when the students enter to STMIK IKMI, the result of the test is low. Low in this case will explains by categories the students get D 25% C 37% B 0% and all the students can not get A. The research uses kuasi eksperiments method with time series design that collaborate classroom action research. The result of research shows about pre test 1 until 4, the score of pre test 1 about 54,129, the score of pre test 2 about 55,548, the score of pre test 3 about 56,032, the score of pre test 4 about 56,097. Based on Kruskall Walls Test shows Asymp Sig score about 0,986 it means there is no significance differences between student perception for the first time and students understanding the materials. In second part of research the students learn accounting that use zahir accounting software for 6 meetings. In third part of research the students has a post test for 4 meetings and the results are the score of post test 1 62,6456, the results are the score of post test 2 70,065, the results are the score of post test 3 80,032, the results are the score of post test 4 86,742. The research analyze statistic test focus on pre test and post test by Kruskall Wall Test that shows Asymp Sig score about 0,000 it means there is a differences between the result of pre test and post test. This reality shows accounting learning by Zahir software gives the positive effect for improving (upgrading) student understanding about accounting.

Title

Found 31 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 31 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 31 Documents
Search