Joseph Nzabanita

Dr. Joseph Nzabanita

University of Rwanda, Kigali, Rwanda
orcid

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 May 2022 in Future Internet
Verified
Journal Article
Open access
PDF file
Belle Fille MurorunkwereBelle Fille Murorunkwere
Belle Fille Murorunkwere
OTOrigene Tuyishimire
Origene Tuyishimire
DHDOMINIQUE HAUGHTON
DOMINIQUE HAUGHTON
JNJoseph Nzabanita
Joseph Nzabanita
Detecting tax fraud is a top objective for practically all tax agencies in order to maximize revenues and maintain a high level of compliance. Data mining, machine learning, and other approaches such as traditional random auditing have been used in many studies to deal with tax fraud. The goal of this study is to use Artificial Neural Networks to identify factors of tax fraud in income tax data. The results show that Artificial Neural Networks perform well in identifying tax fraud with an accuracy of 92%, a precision of 85%, a recall score of 99%, and an AUC-ROC of 95%. All businesses, either cross-border or domestic, the period of the business, small businesses, and corporate businesses, are among the factors identified by the model to be more relevant to income tax fraud detection. This study is consistent with the previous closely related work in terms of features related to tax fraud where it covered all tax types together using different machine learning models. To the best of our knowledge, this study is the first to use Artificial Neural Networks to detect income tax fraud in Rwanda by comparing different parameters such as layers, batch size, and epochs and choosing the optimal ones that give better accuracy than others. For this study, a simple model with no hidden layers, softsign activation function performs better. The evidence from this study will help auditors in understanding the factors that contribute to income tax fraud which will reduce the audit time and cost, as well as recover money foregone in income tax fraud.
1

21 June 2023 in International Journal of Information Technology
Verified
Journal Article
Closed access
PDF file
EMEmmanuel Masabo
Emmanuel Masabo
JNJoseph Nzabanita
Joseph Nzabanita
INInnocent Ngaruye
Innocent Ngaruye
CRCharles Ruranga
Charles Ruranga
JNJean Paul Nizeyimana
Jean Paul Nizeyimana
Denis NdanguzaDenis Ndanguza
Denis Ndanguza

9 April 2023 in African Journal of Science, Technology, Innovation and Development
Verified
Journal Article
Closed access
Belle Fille MurorunkwereBelle Fille Murorunkwere
Belle Fille Murorunkwere
DHDominique Haughton
Dominique Haughton
JNJoseph Nzabanita
Joseph Nzabanita
FKFrancis Kipkogei
Francis Kipkogei
IKIgnace Kabano
Ignace Kabano
With the advancement in technology, the tax base in Rwanda has become broader, and as a result, tax fraud is growing. Depending on the dataset used, fraud detection experts and researchers have used different methods to identify questionable cases. This paper aims to predict features of tax fraud using the most robust supervised machine-learning model. This research provides a context where a fraud expert can use a machine-learning model, and an implemented model offers instant feedback to the fraud expert. We evaluate supervised machine learning models such as Artificial Neural Network, Logistic Regression, Decision Tree, Random Forest, GaussianNB and XGBoost. Based on different evaluation metrics, Artificial Neural Network was the most robust model for predicting tax fraud. Findings reveal that the time of business that indicates the difference in time from when a business started and the time it was audited, the domestic businesses, taxpayers who import and export goods, those with no losses, those whose businesses are located in the eastern province, and those registered on withholding and Value Added Tax types are more susceptible to tax fraud. This study is among the few to evaluate the effectiveness of multiple supervised machine-learning models for identifying tax fraud factors on an accurate data set with numerous tax types. The evidence generated in the current study will serve as a valuable tool for both tax policymakers and auditors, as well as for enhancing awareness of more robust methods for predicting tax fraud.

21 February 2023 in Information
Verified
Journal Article
Open access
PDF file
Belle Fille MurorunkwereBelle Fille Murorunkwere
Belle Fille Murorunkwere
Felicien IhirweFelicien Ihirwe
Felicien Ihirwe
IKIdrissa Kayijuka
Idrissa Kayijuka
JNJoseph Nzabanita
Joseph Nzabanita
DHDOMINIQUE HAUGHTON
DOMINIQUE HAUGHTON
Tax fraud is a common problem for many tax administrations, costing billions of dollars. Different tax administrations have considered several options to optimize revenue; among them, there is the so-called electronic billing machine (EBM), which aims to monitor all business transactions and, as a result, boost value added tax (VAT) revenue and compliance. Most of the current research has focused on the impact of EBMs on VAT revenue collection and compliance rather than understanding how EBM reporting behavior influences future compliance. The essential contribution of this study is that it leverages both EBM’s historical reporting behavior and actual business characteristics to understand and predict the future reporting behavior of EBMs. Herein, tree-based machine learning algorithms such as decision trees, random forest, gradient boost, and XGBoost are utilized, tested, and compared for better performance. The results exhibit the robustness of the random forest model, among others, with an accuracy of 92.3%. This paper clearly presents our approach contribution with respect to existing approaches through well-defined research questions, analysis mechanisms, and constructive discussions. Once applied, we believe that our approach could ultimately help the tax-collecting agency conduct timely interventions on EBM compliance, which will help achieve the EBM objective of improving VAT compliance.
1

2 August 2017 in Communications in Statistics - Theory and Methods
Verified
Journal Article
Closed access
INInnocent Ngaruye
Innocent Ngaruye
JNJoseph Nzabanita
Joseph Nzabanita
DRDietrich von Rosen
Dietrich von Rosen
Martin SingullMartin Singull
Martin Singull
In this article, small area estimation under a multivariate linear model for repeated measures data is considered. The proposed model aims to get a model which borrows strength both across small areas and over time. The model accounts for repeated surveys, grouped response units, and random effects variations. Estimation of model parameters is discussed within a likelihood based approach. Prediction of random effects, small area means across time points, and per group units are derived. A parametric bootstrap method is proposed for estimating the mean squared error of the predicted small area means. Results are supported by a simulation study.

9 January 2023 in IJID Regions
Verified
Journal Article
Open access
LMLydie Mpinganzima
Lydie Mpinganzima
JNJean Marie Ntaganda
Jean Marie Ntaganda
WBWellars Banzi
Wellars Banzi
JMJean Pierre Muhirwa
Jean Pierre Muhirwa
BNBetty Kivumbi Nannyonga
Betty Kivumbi Nannyonga
IMIsambi Sailon Mbalawata
Isambi Sailon Mbalawata
Mathematical modelling is of interest to study the dynamics of coronavirus disease 2019 (COVID-19), and models such as SEIR (Susceptible-Exposed-Infected-Recovered) have been considered. This article describes the development of a compartmental transmission network model - Susceptible-Exposed-Quarantine-Infectious-Infectious, undetected-Infectious, home-based care-Hospitalized-Vaccinated-Recovered-Dead - to simulate the dynamics of COVID-19 in order to account for specific measures put into place by the Government of Rwanda to prevent further spread of the disease. The compartments of this model are connected by parameters, some of which are known from the literature, and others are estimated from available data using the least squares method. For the stability of the model, equilibrium points were determined and the basic reproduction number R 0 was studied; R 0 is an indicator for contagiousness. The model showed that secondary infections are generated from the exposed group, the asymptomatic group, the infected (symptomatic) group, the infected (undetected) group, the infected (home-based care) group and the hospitalized group. The formulated model was reliable and fit the data. Furthermore, the estimated R 0 of 2.16 shows that COVID-19 will persist without the application of control measures. This article presents results regarding predicted spread of COVID-19 in Rwanda.