Hadiza Ali Umar

Hadiza Ali Umar

Auto-generated profile, based on publicly available information

Hadiza Ali Umar

American University of Nigeria, Yola, Nigeria

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
1 December 2019
Verified
Conference Paper
Closed access
HUHadiza Ali Umar
Hadiza Ali Umar
Rajesh PrasadRajesh Prasad
Rajesh Prasad
MFMathias Fonkam
Mathias Fonkam
Electricity or power has become an essential service which is instrumental to many social, economic and technological developments worldwide. However, power utilities competing in emerging markets are confronted with challenges of shortfall in revenue due to collection losses known as Non-Technical losses (NTLs). Assessing and plummeting NTL to increase revenue, and reliability of the distribution of power remains a top priority for the utilities and regulators. The amount of data generated by power utilities offers exceptional opportunities owing to Machine Learning (ML) techniques to better understand the payment and consumption patterns of consumers. Consumers on the same profile can be segmented into clusters of similar behaviour through Customer segmentation. This paper applies k-means and DBSCAN clustering algorithms to segment customers by a measure of their bill payments, consumption and tariff plans then grouped into clusters. The clusters encountered, shows different categories of consumers in relation to their obligation profiles and tariff plans. K-means gives a better visualization in terms of cluster assignment. Dense clusters and outliers were viewed in DBSCAN.

21 February 2022 in International Journal of Information Technology
Verified
Journal Article
Closed access
PDF file
HUHadiza Ali Umar
Hadiza Ali Umar
MFMathias Fonkam
Mathias Fonkam
Rajesh PrasadRajesh Prasad
Rajesh Prasad

1 October 2021 in Recent Advances in Computer Science and Communications
Verified
Journal Article
Closed access
PDF file
HUHadiza Umar
Hadiza Umar
Rajesh PrasadRajesh Prasad
Rajesh Prasad
MFMathias Fonkam
Mathias Fonkam
Background: A resolve engaged by energy regulators worldwide to curtail the liability of Non-Technical Losses (NTLs) in power is the installation of Smart Meters to measure consumed power. However, power regulators in developing countries are confronted with huge metering gap in an era of momentous energy theft which has resulted in deficits in revenue, debts and ultimately power cuts. Objective: The objective of this research is to predict whether the customers’ who are unmetered are eligible to be metered by identifying worthy and unworthy customers for metering given their bill payment history. Methods: The approach analyses the performance accuracy of some machine learning algorithms on small datasets by exploring the classification abilities of Deep learning, Naïve Bayes, Support Vector Machine and Extreme Learning Machine using data obtained from an electricity distribution company in Nigeria. Results: The performance analysis shows that Naïve Bayes classifier outperformed the Deep Learning, Support Vector Machine and Extreme Learning Machine algorithms. Experiments in deep learning have shown that the alteration of batch sizes has an effect on the outputs. Conclusion: This paper presents a data-driven methodology for the prediction of consumers’ eligibility to be metered. The research has shown the performance of deep learning, Naive Bayes, SVM and ELM on small dataset. The research will help utility companies in developing countries with large populations and huge metering gaps to prioritise the installation of smart meters based on consumer’s payment history.