Intelligent Citizenship Identity through Family Pedigree Using Graph-Signature Based Random-Forest Model

  • Adedayo D. Adeniyi
  • Semiu O. Oladejo
  • Tiwalade M. Usman

Abstract

There has been a global upsurge of interest in the topic of citizenship identity over the past decades, specifically in the world dominated by profound insecurity, inequalities, proliferation of identities, and rise of identity politics,engendered by capitalism. However finding effective solution to these problems has been rendered difficult. To alleviate these problems, this paper presents an analytical Machine learning model that suitably combined the graph signature with random forest techniques. This study presents the design and realization of a novel Intelligent Citizenship Identity through family pedigree using Graph Signature based random forest (GSB-RF) model. The study also showcases the development of a novel graph signature technique referred to as Canonical Code Signature(CCS) method. The CCS method is used at the pre-processing stage of the identification process to build signature for any given tuple. Performance comparisim between the present system and the baseline techniques which includes: the K-Nearest Neighbour and the traditional Random Forest shows that the present system outperformed the baseline method studied. The proposed system shows capability to perform continuous re-identification of Citizens based on their family pedigree with ability to select best sample with low computational complexity, high identification accuracy and speed. Our experimental result shows that the precision rate and identification quality of our system in most cases are equal to or greater than 70%. Therefore, the proposed Citizenship Identification machine is capable of providing usable, consistent, efficient, faster and accurate identification, to the users, security agents, government agents and institutions on-line, real-time and at any-time.

Published
2021-09-13
Section
Articles

Journal Identifiers


eISSN: 2579-0617
print ISSN: 2579-0625