Claim Missing Document
Check
Articles

Found 12 Documents
Search

Using A*Algorithm and Google Maps API for Web-Based Path Optimisation Public Vehicles Routes in Medan City Faridawaty; Arnita; Dewi, Sri
Jurnal Penelitian Pendidikan IPA Vol 10 No SpecialIssue (2024): Science Education, Ecotourism, Health Science
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v10iSpecialIssue.8388

Abstract

Public transport optimum route search is a problem to find a route between two points with the minimum number of weights.  The research method that can be used to solve the problem of finding the optimum route is divided into two stages, namely first designing a model device using the A* algorithm and Google Maps API, second designing an android-based application. The purpose of this research is to develop an android-based system that can contain information on the optimum route of public transport in Medan City.  The A* algorithm is a computer algorithm that uses distance estimation using the search for the closest path to reach the destination and has a heuristic function that is used as a basis for consideration to determine the choice of a number of alternatives to achieve the target effectively. The output of this research is the application of optimum route information for Medan city public transport based on android. The level of readiness of this research technology is of the Software type at the subsystem module validation status in a laboratory environment with indicators of integrated basic software components working together.
OPTIMALISASI DETEKSI KECURANGAN PADA TRANSAKSI E-WALLET MENGGUNAKAN ALGORITMA ISOLATION FOREST BERBASIS BIG DATA Anggi Silalahi; Azhara Amelia H; Sabrina Akva; Desni Paramitha Purba; Fanny Ramadhani; Arnita
PROSISKO: Jurnal Pengembangan Riset dan Observasi Sistem Komputer Vol. 12 No. 3 (2025): Prosisko Vol. 12 No. 3 November 2025
Publisher : Pogram Studi Sistem Komputer Universitas Serang Raya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30656/prosisko.v12i3.10738

Abstract

Kemajuan teknologi finansial telah mendorong adopsi layanan dompet digital (e-wallet) secara luas. Namun, peningkatan volume transaksi juga membawa risiko keamanan yang tinggi, khususnya terkait aktivitas kecurangan. Penelitian ini bertujuan membangun sistem deteksi kecurangan pada transaksi e-wallet menggunakan algoritma Isolation Forest, yang mampu mengidentifikasi anomali secara efisien tanpa memerlukan data berlabel. Dataset yang digunakan terdiri dari 6.362.620 transaksi e-wallet yang mencakup atribut numerik dan kategorikal. Proses penelitian meliputi tahapan preprocessing data, pelatihan model, dan evaluasi kinerja dengan metrik precision, recall, dan F1-score. Hasil evaluasi menunjukkan bahwa meskipun akurasi model mencapai 99%, recall terhadap transaksi fraud masih rendah, yaitu sebesar 4%, menandakan perlunya pendekatan lanjutan untuk meningkatkan sensitivitas model. Penelitian ini menunjukkan potensi Isolation Forest dalam mendeteksi pola transaksi anomali pada data berukuran besar serta memberikan dasar untuk pengembangan sistem keamanan finansial berbasis data.