Silvano Sugidamayatno
Master Program of Computer Science, FMIPA UGM, Yogyakarta

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : IJCCS (Indonesian Journal of Computing and Cybernetics Systems)

Outlier Detection Credit Card Transactions Using Local Outlier Factor Algorithm (LOF) Silvano Sugidamayatno; Danang Lelono
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 13, No 4 (2019): October
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.46561

Abstract

Threats or fraud for credit card owners and banks as service providers have been harmed by the actions of perpetrators of credit card thieves. All transaction data are stored in the bank's database, but are limited in information and cannot be used as a knowledge. Knowledge built with credit card transaction data can be used as an early warning by the bank. The outlier analysis method is used to build the knowledge with a local outlier factor algorithm that has high accuracy, recall, and precision results and can be used in multivariate data. Testing uses a matrix sample and confusion method with attributes date, categories, numbers, and countries. The test results using 1803 transaction data from five customers, indicating that the average value accuracy of LOF algorithms (96%), higher than the average accuracy values of the INFLO and AFV algorithms (84% and 77%).