Fraud Detection System for Effective Healthcare Administration in Nigeria using Apache Hive and Big Data Analytics: Reflection on the National Health Insurance Scheme

Nigerian researchers have shown that the lack of adequate mechanisms for fraud detection has impaired both providers and beneficiaries of this scheme. This work develops a fraud detection program for Nigeria's National Health Insurance Scheme (NHIS). Nigeria's National Health Insurance Sch...

Full description

Saved in:
Bibliographic Details
Published in:2022 5th Information Technology for Education and Development (ITED) pp. 1 - 6
Main Authors: Ogala, Justin Onyarin, Mughele, Ese Sophia, Chiemeke, Stella Chinnye
Format: Conference Proceeding
Language:English
Published: IEEE 01.11.2022
Subjects:
Big Data
Detection Apache Hive Deep Learning
File systems
Fraud
Healthcare Insurance
Insurance
Maintenance engineering
Medical services
NHIS
Organizations
Unified modeling language
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Nigerian researchers have shown that the lack of adequate mechanisms for fraud detection has impaired both providers and beneficiaries of this scheme. This work develops a fraud detection program for Nigeria's National Health Insurance Scheme (NHIS). Nigeria's National Health Insurance Scheme (NHIS) and Health Maintenance Organizations (HMOs) are the subjects of this study. The study was conducted using available data from NHIS-registered healthcare facilities and HMOs. Unified Modeling Language (UML) tools were used to create the framework. The framework was built with Apache Derby DB, Hadoop Distributed File System (HDFS), and Apache MapReduce as the big data processing platform. Using Apache Hive and Big Data Analytics, a system for detecting healthcare fraud is developed. This system used data from the Nigerian National Health Insurance Scheme (NHIS), which was broken down into three categories: enrolment, referral, and claim data. The analysis of current healthcare investigative methods is conducted, and a new framework is proposed.
DOI:10.1109/ITED56637.2022.10051541