Claim Missing Document
Check
Articles

Found 32 Documents
Search

APLIKASI E-DIAGNOSIS PENYAKIT ENDIMIK BERBASIS ANDROID MENERAPKAN METODE OPTIMASI NAÏVE BAYES Hadiyansyah Hadiyansyah; Diana Diana
Jurnal Ilmiah Matrik Vol 24 No 3 (2022): Jurnal Ilmiah Matrik
Publisher : Direktorat Riset dan Pengabdian Pada Masyarakat (DRPM) Universitas Bina Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33557/jurnalmatrik.v24i3.2002

Abstract

Lack of public understanding of endemic diseases can increase the number of sufferers. This study aims to build an e-diagnosis application to determine the type of endemic disease using the Naïve Bayes Optimization method. This application will be able to provide information about the disease suffered by the patient based on the symptoms entered in the application. The information provided includes a description of the disease, its causes and solutions. The application development stage adopts the stages in the Expert System Development Life Cycle (ESDLC) which includes project initialization, knowledge engineering process and implementation. The application of the Naïve Bayes Oprimization method produces a diagnosis result in the form of the type of disease dan its opportunities. The application can accessed by the public anywhere and anytime because this application is based on Android. Utilization of android can optimze the use of this application.
Penerapan Metode Fuzzy Tsukamoto Dalam Memprediksi Permintaan Barang Dwi Melina; Diana Diana
Journal of Computer System and Informatics (JoSYC) Vol 4 No 3 (2023): May 2023
Publisher : Forum Kerjasama Pendidikan Tinggi (FKPT)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/josyc.v4i3.3194

Abstract

Azthana Digital Media is a digital agency that works with a number of Indonesian vendors, including Pempek Mola, Pempek Kalani, and Eightsy Bag. Azthana oversees distribution and maintains warehousing for all three brands simultaneously. However, due to fluctuations in product demand, there are sometimes gaps in the supply of goods across all brands. So we need a method that can be implemented in inventory management and product development in the future. Forecasting strategies are useful tools for maintaining and expanding stock levels. In this study, researchers used the Fuzzy Tsukamoto technique built into a web-based inventory system to estimate consumer demand. Sales data, product purchases and goods production data are the three main inputs used to formulate demand forecasts. The results of this study can be concluded that the inventory system created can help the related team to find out the exact amount of stock available and to implement the Tsukamoto fuzzy calculations in predicting the demand for goods in the system that is running well as evidenced by the results of manual calculations and the system if the demand value is input equal to 355 with four fuzzy rules that are set together to produce a production value of requests for goods that must be ordered to suppliers of 522 packs.
PENERAPAN MESIN INFERENSI FUZZY METODE MAMDANI PADA SISTEM PREDIKSI JUMLAH PRODUKSI PEMPEK PALEMBANG Diana
JURNAL ILMIAH BETRIK Vol. 14 No. 02 AGUSTUS (2023): JURNAL ILMIAH BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : P3M Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/betrik.v14i02 AGUSTUS.94

Abstract

This study aims to predict the “best’ amount of production so that there is no shortage of inventory or excess inventory. Fluctuations in the number of requests for crispy caused uncertainty about the amount of inventory and the amount of production. The Mamdani fuzzy method as a fuzzy logic inference engine can be used to overcome this uncertainty. The steps taken are to determine the inference system input, fuzzification, implication, aggregation and defuzzification. The set data used is supply (X) and demand (Y) for crispy pempek in December 2022. Furthermore, this data is mapped into the fuzzy membership function values using ascending function, decreasing function and triangular function graphs. At this stage, the fuzzy membership function equations are determined in 3 linguistic variables, each supply and demand input data will be mapped into a few, medium or many sets. The provisions that apply are if the fuzzy membership values are added up, it will produce a value of 1. Furthermore, at the implication, 9 basic rules are determined in IF-THEN rule. AT the aggregation stage use the max min function, where the maximum membership function value will be selected for supply and demand variables. In test data 1, the amount of inventory is included in the medium category with a fuzzy membership function value of 0,914 and the number of demand is included in the medium category with a fuzzy membership function value of 0,860 so that the membership function for the variables number of production is min {0.914, 0.860} = 0.860. The predicted ‘best’ production number in January are 306(LOM) and 271 (MOM). The Mamdani method can be applied to determine forecasting the amount of production so that the amount of inventory is as needed.
PEMBUATAN PROTOTYPE SISTEM E-LAPORAN BERBASIS WEB SEBAGAI PENDUKUNG PROGRAM SATU DATA PADA DINAS PU BINA MARGA DAN TATA RUANG PROVINSI SUMATERA SELATAN Muhammad Kiki Nugraha; Diana Diana
JURNAL AKADEMIK PENGABDIAN MASYARAKAT Vol. 4 No. 4 (2026): JULI
Publisher : CV. KAMPUS AKADEMIK PUBLISING

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61722/japm.v4i4.11500

Abstract

The Department of Public Works for Highways and Spatial Planning (DPUBMTR) of South Sumatra Province is responsible for providing accurate infrastructure feasibility data, in line with Presidential Regulation No. 39 of 2019 on One Data Indonesia. In practice, the existing E-Laporan platform suffered from recurring bugs, a static interface, and insufficient server performance during peak usage. This study aims to redesign (rewrite) a web-based E-Laporan prototype to operate reliably on a local server (localhost). The development method used was the Prototyping model, covering requirements analysis, UI/UX design (Figma), back-end development using Native PHP, MySQL database management, and Bootstrap-based interface styling, followed by functional testing. Black Box Testing results confirmed that all core features—including authentication, CRUD operations, form validation, and dashboard charts—functioned correctly. Two technical issues were resolved during development: a MySQL port conflict (3306 redirected to 3307) and data type precision errors in road length calculations (INT changed to FLOAT). The resulting prototype successfully eliminated data duplication and accelerated the digitalization of physical records. Future development is recommended to deploy the system to a public cloud server.
Data mining M David; Diana Diana
Dinamik Vol 31 No 2 (2026)
Publisher : Universitas Stikubank

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35315/dinamik.v31i2.10418

Abstract

The purpose of this study is to provide recommendations to companies in determining stock levels based on previous sales data predictions using the K-Nearest Neighbor method. The data mining process carried out in this study follows the stages in Knowledge Discovery in Database (KDD) to produce information in accordance with a predetermined sequence consisting of data selection, preprocessing, transformation, data mining, and interpretation or evaluation. The tool used in the application of data mining is Python. The results of this study indicate the factors that influence product sales at PT. Alam Perkasa Lestari Palembang, including product price, sales (marketing), market competition, and others. The product sales prediction results based on the highest accuracy value are k=3168 with an accuracy of 97.00%. Thus, the KNN k=3168 algorithm method can be implemented to predict product sales at PT. Alam Perkasa Lestari Palembang.. Keywords : Data mining, K-Nearest Neighbor, Sales
PELATIHAN APLIKASI CAPCUT DAN TIKTOK UNTUK MENINGKATKAN KINERJA UMKAM DI DESA TANJUNG LAGO M. Rizki Julianto; Diana Diana
NGABDIMAS Vol. 8 No. 01 Juni (2025): Jurnal Ngabdimas Pengabdian Pada Masyarakat
Publisher : PPPM Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/c6c0v737

Abstract

Angka transaksi belanja online semakin meningkat di era ini. Pemanfaatan platform e-commerce memungkinkan terjadi transaksi online yang dapat meningkatkan interaksi antara pelaku UMKM dan konsumen.  Sangat disayangkan apabila peluang baik ini tidak dimanfaatkan oleh pelaku UMKM di Desa Tanjung Lago. Pelatihan ini diikuti oleh 12 orang pelaku UMKM di Desa Tanjung Lago.  Para peserta mengikuti seluruh rangkaian acara dengan baik dan antusias. Setiap materi yang di sampaikan dapat dipahami dan sesuai dengan yang diharapkan. Peningkatan pemahaman materi pelatihan meningkat dilihat dari hasil pre test dan post test yang meningkat rata-rata 11.67 point dari rata-rata 70,83 menjadi 82,5. Berdasarkan hasil pengolahan kuisioner diperoleh nilai rata-rata 4,3 yang bermakna bahwa kualitas pelatihan masuk dalam kategori baik. Pelaksanaan program pelatihan penggunaan platform e-commerce melalui aplikasi CapCut dan TitTok di Desa Tanjung Lago berhasil memberikan dampak positif terhadap peningkatan literasi digital dan kemampuan promosi pelaku UMKM.  Kegiatan ini membekali para pelaku usaha dengan keterampilan dalam membuat konten digital yang menarik serta mengoptimalkan media sosial sebagai sarana pemasaran produk.
PENGUATAN TATA KELOLA PELAYANAN MAHASISWA DALAM IMPLEMENTASI PROGRAM MERDEKA BELAJAR KAMPUS MERDEKA (MBKM) Siti Nurhayati Nafsiah; Mukran Mukran; Emilia Gustini; Merry Agustina; A.Yani Ranius; Diana Diana; Darwin Darwin
JURNAL PENGABDIAN MANDIRI Vol. 5 No. 1 (2026): Januari 2026
Publisher : Bajang Institute

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

Abstract

Implementasi Program Merdeka Belajar Kampus Merdeka (MBKM) memerlukan tata kelola pelayanan mahasiswa yang efektif dan akuntabel. Namun, masih dijumpai kendala berupa keterbatasan pemahaman kebijakan dan belum optimalnya kualitas pelayanan. Kegiatan Pengabdian kepada Masyarakat ini bertujuan memperkuat tata kelola pelayanan mahasiswa dalam mendukung implementasi MBKM melalui pendekatan penguatan kapasitas. Metode pelaksanaan meliputi sosialisasi, diskusi kelompok terarah, dan pendampingan dengan pendekatan partisipatif. Evaluasi keberhasilan kegiatan dilakukan menggunakan kuesioner pre-test dan post-test untuk mengukur peningkatan pemahaman, kualitas, serta akuntabilitas pelayanan. Hasil kegiatan menunjukkan adanya peningkatan pemahaman mitra terhadap kebijakan MBKM dan perbaikan persepsi terhadap kualitas pelayanan mahasiswa. Penggunaan kuesioner terbukti efektif sebagai instrumen evaluasi capaian kegiatan. Kegiatan ini berkontribusi dalam mendukung implementasi MBKM yang lebih optimal serta penguatan tata kelola pelayanan mahasiswa melalui kegiatan pengabdian kepada masyarakat.
Analisis Kinerja Algoritma Kirsch dan Robinson dalam Deteksi Tepi Citra Diana; Septa Riansyah
BETRIK Vol. 17 No. 02 (2026): Jurnal Ilmiah BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : PPPM Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/vcxp9w08

Abstract

One of the important stages in digital image processing is edge detection, which serves to determine the boundaries of objects in an image. This process is very important for various applications, such as image segmentation, pattern recognition, and object analysis. Two very popular algorithms in gradient operator-based edge detection are Kirsch and Robinson. The Kirsch algorithm uses eight kernels with compass directions to detect pixel intensity changes maximally, while the Robinson algorithm uses eight simpler kernels with a lighter computational approach. The aim of this research is to examine how both algorithms detect image edges based on the sharpness of the results, sensitivity to noise, and computational efficiency. The study was conducted by testing both algorithms on several grayscale test images and comparing the visual and quantitative edge detection results using the parameters of accuracy, precision, recall, and F-Measure. The analysis results show that the Robinson algorithm is more efficient in the computation process but produces smoother edges. The Kirsch algorithm, on the other hand, produces sharper and more detailed edge detection but requires longer computation time. Therefore, which algorithm is most suitable for the application depends on whether the priority is on process efficiency or detection quality.
Implementasi Sistem Deteksi Golongan Darah Menggunakan Metode Sobel Diana; Roytuddin
STORAGE: Jurnal Ilmiah Teknik dan Ilmu Komputer Vol. 5 No. 3 (2026): Agustus
Publisher : Yayasan Literasi Sains Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55123/storage.v5i3.8316

Abstract

Interpretasi golongan darah secara manual sering kali menghadapi kendala subjektivitas dan variasi kondisi lingkungan, sehingga diperlukan sistem otomatisasi yang akurat dan efisien. Penelitian ini mengembangkan sistem deteksi golongan darah otomatis melalui integrasi operator Sobel sebagai ekstraktor fitur dan Random Forest sebagai model klasifikasi. Proses diawali dengan pengolahan citra menggunakan kanal hijau untuk memperjelas batas aglutinasi, kemudian nilai Edge Density Ratio (EDR) diekstraksi dari area reagen Anti-A, Anti-B, dan Anti-D sebagai input klasifikasi. Berdasarkan pengujian terhadap 432 data citra, sistem yang diusulkan mencapai akurasi global sebesar 97,92%. Hasil ini menunjukkan peningkatan signifikan dibandingkan pendekatan rule-based murni yang hanya mencapai akurasi 44,68%. Selain itu, metode ini menawarkan efisiensi komputasi yang lebih baik dibandingkan arsitektur Deep Learning konvensional, sehingga menjadi solusi yang lebih praktis untuk implementasi pada perangkat medis portabel berbasis edge computing tanpa mengorbankan performa.
RANCANG BANGUN APLIKASI E-KONSELING BERBASIS WEB DENGAN METODE CERTAINTY FACTOR Muhammad Kevin Al Khahfi; Diana Diana
Jurnal Sistem Informasi dan Informatika (Simika) Vol. 9 No. 2 (2026): Jurnal Sistem Informasi dan Informatika (Simika)
Publisher : Program Studi Sistem Informasi, Universitas Banten Jaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47080/simika.v9i2.4542

Abstract

Incompatibility between a chosen major and students’ interests and talents is often a primary obstacle to academic success in higher education. At the same time, the availability of guidance and counseling teachers for personal assistance remains highly limited. This Research aims to design and develop a web-based e-counseling expert system application capable of automatically identifying major recommendations based on students’ self-potential at SMA Quraniah Palembang. The Research method employed is the Expert System Development Life Cycle (ESDLC) with qualitative approaches. Data collection techniques were conducted through observation, literature review, and in-depth interviews with an expert, specifically a Guidance and Counseling teacher, to acquire the knowledge base. The system was developed using the PHP programming language with the Laravel framework, a MySQL database, and a responsive interface built with Bootstrap. To address diagnostic uncertainty, this study implemented the Certainty Factor (CF) method by combining expert Measure of Belief (MB) and Measure of Disbelief (MD) values and user certainty values. The results indicate that the E-Counseling application accurately performs the identification function, producing outputs as percentage confidence levels for specific types of intelligence and relevant majors. Black-box testing confirmed that all functional features of the system operate correctly in accordance with the design specifications. In conclusion, this application is an effective decision-support tool for students in planning their academic futures. The implication of this Research is the availability of a digital platform that facilitates the efficient identification of student potential without time and place constraints, serving as a foundation for future technology-based counseling development.