An Improved Defragmentation Model for Distributed Customer’s Bank Transactions


  • Chukwuemeka. K. Onwuamaeze University of Port Harcourt, Choba, Rivers State, Nigeria
  • Christopher. I. Ejiofor Edwin Clark University


Defragmentation, distributed bank transaction, financial transactions classifier, recommender System


Authors: K. Onwuamaeze and C. I. Ejiofor

Received 17 April 2020/Accepted 20 May 2020

The application of Information and Communication Technologies (ICTs) in trade and commerce has changed the ways trading transactions are carried out with the sole aim of accomplishing the growing expectations of organizational clients. In Nigeria nowadays, monetary transactions are made over disparate channels by bank clients.  The exchanges can be heterogeneous, energetic, inter-related and as often as possible dispersed over numerous platforms. The variety, volume and velocity of the transactions can be very cumbersome for manual computations.  In most cases, it is difficult to make informed decisions from the trend and patterns of the transactional datasets.  However, a proper analysis on the datasets generated from the banking transactions of a customer can help in profiling the customer, target recommended solutions and achieve customer loyalty. This project implemented an improved defragmentation model for distributed customer’s banks transactions that can be used by bank customers. It employed Naïve Bayes machine learning and collaborative filtering techniques to separate multiple transactions across numerous payment channels and deploy recommendations for the customer. The Prototype software methodology was adopted in the design. At the implementation of the research work, we utilized test cases to prove that customer’s bank transactions over dispersed channels can be classified based on the user’s query. Customers can now classify their transactions based on purpose of the transactions, the benefiting bank accounts, the beneficiary’s name and their account numbers. The bank customer can quickly obtain a digital statement of account from all her bank accounts with the software.


Download data is not yet available.

Author Biographies

Chukwuemeka. K. Onwuamaeze, University of Port Harcourt, Choba, Rivers State, Nigeria

Department of Computer Science

Christopher. I. Ejiofor, Edwin Clark University

Department of Mathematical Sciences


Agrawal, V., Kumar, V, Ghodke, V. & Temdhare, K. (2014) Multibanking Transaction Interface using MVC2 Architecture. International Journal of Computer Science & Engineering Technology (IJCSET), 5, 4, pp. 284-288

Aleskerov, E.,, Freisleben, B. & Rao, B. (1997) CARDWATCH: A Neural Network Based Database Mining System for Credit Card Fraud Detection. In: Proceedings of Computational Intelligence for Financial Engineering (CIFEr), 220--226,

Ogwueleka, F. N., Sanjay, M., Ricardo, C. P. &, Luis, F. Neural Network and Classification Approach in Identifying Customer Behaviour in the Banking Sector: A Case Study of an International Bank, Human Factors and Ergonomics in Manufacturing and Setvice Industries, 25, 1, pp. 28-42.

Amit C, howdhry (2016) How Clinc Developed The Most Advanced A.I. for the banking Industry available at -amitchowdhry/

Krichene, A. (2017). Using a naive Bayesian classifier methodology for loan risk assessment: Evidence from a Tunisian commercial bank., Journal of Economics, Finance and Administrative Science, 22, 42, pp. 3-24.

Adebayo, A, O., Agbola, I . S., Ayangbade, A. O. & Obajimi, O. O. (2015). Bank Products Recommender. The International Journal of Engineering and Science, 4, 7, pp. 41-52.

Raamani, R. (2016). Recommender system based on customer segmentation (RSCS). Kybernetes, 45, 6, pp. 946-961.