OJIT Home

2025 - Volume 8 - Number 2

Reduction of Concept Drift Effect on Mobile Network Fraud Detection Using Deep Learning Markov Transition Encoding Dynamic LSTM Approach

James Mundia * ORCID: 0009-0000-5074-6650
University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA

Evans Kirimi Miriti
University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA

Stephen Mburu
University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA

Andrew Mwaura Kahonge
University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA

Christopher Kipchumba Chepken
University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA

Open Journal for Information Technology, 2025, 8(2), 21-52 * https://doi.org/10.32591/coas.ojit.0802.01021m
Received: 22 July 2025 ▪ Revised: 11 October 2025 ▪ Accepted: 5 November 2025

LICENCE: Creative Commons Attribution 4.0 International License.

ARTICLE (Full Text - PDF)

ABSTRACT:
Background: Mobile network infrastructure has exponentially advanced in the past 20 years, resulting in tremendous evolution of the capabilities/services that the network can deliver. With these advancements, fraud activities have in equal measure metamorphosed into a complex web involving and spanning multiple touchpoints. Although different machine learning approaches have been developed and adopted, mobile network fraud remains dynamic and continually changes in form and means with time, resulting in degraded efficiency of these ML models. The phenomenon resulting in this degradation of efficiency with time is known as concept drift, which can be described as changes in the conditional distribution of the target variable, which is also referred to as output given the input features or the inputs, while the distribution of the inputs may still stay unchanged. This study aimed to analyze the effect of concept drift in relation to mobile network fraud detection, propose and develop an approach that reduces the concept drift effect when classifying and detecting mobile network fraud. Method: A quantitative research methodology was adopted using experiments as the core method. The dataset used for the study was an extract of DataStream from a Kenyan telco company and contains about 400,000 events or records generated by a process that randomly selected 65000 events for each month (Jan 2023 to June 2023) from the complete dataset. A deep learning model known as MTED-LSTM was developed and trained, then evaluated using the 6-month data stream. The other four commonly used approaches, i.e., Random Forest, Naïve Bias, Support Vector Machine, and Logistic Regression, were also developed and evaluated using the same data stream. Eight evaluation criteria were used to evaluate the effectiveness of the models. The data stream was used to evaluate whether the models were affected by the drift and how good they were in handling the concept drift effect. Results: Using each evaluation criterion, the models were tested through each month's data stream without retraining the model. The result indicated a gradual reduction of the effectiveness of the models with time, attributed to the concept drift effect. MTED-LSTM model showed to be more effective through the six-month data stream and least affected by concept drift as compared to other models developed in the study. Conclusion: The MTED-LSTM model effectively managed the effects of concept drift better due to its ability to store the previous model state on the time series DataStream, which is also referred to as memory. This is evidenced by the results, where all the performance indicators show that the model is least affected by the drift with time. It is worth noting that between the 4th and 5th months, all the other models drastically reduce their effectiveness, but the MTED-LSTM model recorded improved effectiveness during this time frame, resulting in a stable and smooth performance.

KEY WORDS: mobile network, fraud, concept drift, long short-term memory, deep learning.

CORRESPONDING AUTHOR:
James Mundia, University Of Nairobi, Faculty of Science, Department of Computing and Informatics, Nairobi, KENYA.

REFERENCES:

Abdul et al., 2024, Transfer-learning enabled adaptive framework for load forecasting under concept-drift challenges in smart-grids across different-generation-modalities. Available at: https://www.sciencedirect.com/science/article/pii/S2352484724006176.

Adele et al., 2011, Random Forests. Available at: https://link.springer.com/chapter/10.1007/978-1-4419-9326-7_5.

Ajay et al., 2019, Review of Deep Learning Algorithms and Architectures. Available at:  https://ieeexplore.ieee.org/abstract/document/8694781/.

Ajith Abraham, 2005, Artificial Neural Networks. Available at: http://wsc10.softcomputing.net/ann_chapter.pdf.

Alberto et al., 2018, SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary. Available at: https://www.jair.org/index.php/jair/article/view/11192.

Angshuman et al., 2017, Improved Random Forest for Classification. Available at: https://ieeexplore.ieee.org/abstract/document/8357563/.

Apapan et al., 2018, Credit Card Fraud Detection using Deep Learning based on Auto-Encoder and Restricted Boltzmann Machine. Available at: https://thesai.org/Downloads/Volume9No1/Paper_3-Credit_Card_Fraud_Detection_Using_Deep_Learning.pdf.
Baena-Garcia et al,2006, Early drift detection method. Available at: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=2747577a61c70bc3874380130615e15aff76339e.

Basseville et al., 1994, Early warning of slight changes in systems. Available at: https://www.sciencedirect.com/science/article/pii/0005109894902313.

Biju, 2022, SMS Fraud Detection Using Machine Learning. Available at: https://www.researchgate.net/profile/Soumya-Prusty/publication/360371905_SMS_Fraud_Detection_Using_Machine_Learning/links/628781c5cd5c1b0b34e95817/SMS-Fraud-Detection-Using-Machine-Learning.pdf.

C. Alippi et al., 2008. An effective just-in-time adaptive classifier for gradual concept drifts. Available at: https://ieeexplore.ieee.org/abstract/document/6033426/.

Campanharo et al. 2011, Duality between Time Series and Networks. Available at: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0023378.

Chong et al., 2018, Classification algorithmof Chinese sentiment orientation based on dictionary and LSTM. Available at: https://dl.acm.org/doi/abs/10.1145/3291801.3291835.

Delany et al., 2005, A Comparison of Ensemble and Case-Base Maintenance Techniques for Handling Concept Drift in Spam Filtering. Available at: https://cdn.aaai.org/FLAIRS/2006/Flairs06-067.pdf.

Dennis et al., 2018, Adapting to Concept Drift in Credit Card Transaction Data Streams Using Contextual Bandits and Decision Trees. Available at: https://ojs.aaai.org/index.php/AAAI/article/view/11411.

Distributions. Available at: https://epubs.siam.org/doi/abs/10.1137/1.9781611972771.1

Dries et al., 2009. Adaptive concept drift detection. Available at: https://onlinelibrary.wiley.com/doi/abs/10.1002/sam.10054.

Duda et al., 200, Book review: " Pattern classification". Available at: http://www.worldscinet.com/google/S1469026801000251.pdf.

Evgeny et al., 2018, Modern approaches to prevent fraud in mobile communications networks, Available at: https://ieeexplore.ieee.org/abstract/document/8317111.

Fidelis et al., 2024, Enhancing the Random Forest Model via Synthetic Minority Oversampling Technique for Credit-Card Fraud Detection. Available at: https://www.researchgate.net/profile/Margaret-Okpor/publication/379324623_Enhancing_the_Random_Forest_Model_via_Synthetic_Minority_Oversampling_Technique_for_Credit-Card_Fraud_Detection/links/6630201b7091b94e93e7086b/Enhancing-the-Random-Forest-Model-via-Synthetic-Minority-Oversampling-Technique-for-Credit-Card-Fraud-Detection.pdf.

Firas et al., 2023, DA-LSTM: A dynamic drift-adaptive learning framework for interval load forecasting with LSTM networks. Available at: https://www.sciencedirect.com/science/article/pii/S0952197623006644.

Gallego et al., 2017, A survey on data preprocessing for data stream mining: Current status and future directions. Available at: https://www.sciencedirect.com/science/article/pii/S0925231217302631.

Gama et al., 2013, A Survey on Concept Drift Adaptation. Available at:  https://dl.acm.org/doi/abs/10.1145/2523813.

Gama et al., 2004, Concept Drift in Decision Trees Learning from Data Streams. Available at: https://www.researchgate.net/profile/Joao-Gama/publication/236007671_Concept_Drift_in_Decision-Tree_Learning_from_Data_Streams/links/00b7d53959ecb5e3e2000000/Concept-Drift-in-Decision-Tree-Learning-from-Data-Streams.pdf.

Gao et al., 2007, A General Framework for Mining Concept-Drifting Data Streams with Skewed Enzo et al,2007, Introduction to artificial neural networks. Available at: https://journals.lww.com/eurojgh/FullText/2007/12000/Introduction_to_artificial_neural_networks.5.aspx.

Geoffrey et al., 2016, Characterizing Concept Drift. Available at: https://arxiv.org/pdf/1511.03816

Georgios et al., 2019, Augment to Prevent: Short-Text Data Augmentation in Deep Learning for Hate-Speech Classification. Available at:  https://dl.acm.org/doi/abs/10.1145/3357384.3358040.

Gerard et al., 2015, Random Forest Guided Tour. Available at: https://link.springer.com/article/10.1007/s11749-016-0481-7.

Gers, et al, 2000, Learning to forget: Continual prediction with LSTM. Available at: https://ieeexplore.ieee.org/abstract/document/6789445/.

Hochreiter et al., 1997, Long short-term memory. Available at: https://ieeexplore.ieee.org/abstract/document/6795963/.

Hochreiter et al., 2001, Gradient flow in recurrent nets: the difficulty of learning long-term dependencies. A Field Guide to Dynamical Recurrent Neural Networks. Available at: https://ieeexplore.ieee.org/document/5264952.

H. Raza et al., 2015, Adaptive learning with covariate shift-detection for motor imagery-based brain–computer interface. Available at: https://link.springer.com/article/10.1007/s00500-015-1937-5.

Huiying et al.,2018, Adaptive Fraud Detection System Using Dynamic Risk Features. Available at: https://arxiv.org/abs/1810.04654.

JOAO et al., 2014, A Survey on Concept Drift Adaptation. Available at: https://dl.acm.org/doi/abs/10.1145/2523813.

Jie Lu et al, 2020, Data-driven decision support under concept drift in streamed big data. Available at: https://link.springer.com/article/10.1007/s40747-019-00124-4.

Judi, 2002, Dealing with Missing Data. Available at: https://mro.massey.ac.nz/handle/10179/4355.

J. Shao et al., 2014, Prototype-based learning on concept-drifting data streams. Available at: https://dl.acm.org/doi/abs/10.1145/2623330.2623609.

Kala, 2021, Assessment of SIM Box Fraud: An Approach to National Security Threat. Available at: https://www.researchgate.net/profile/Kala-Baskar/publication/356914651_Assessment_of_SIMBox_Fraud_An_Approach_to_National_Security_Threat/links/61b71797a6251b553ab51ccc/Assessment-of-SIMBox-Fraud-An-Approach-to-National-Security-Threat.pdf.

Kelly et al., 1999, The impact of changing populations on classifier performance. Available at: https://dl.acm.org/doi/pdf/10.1145/312129.312285.

Kenji, 2011, Artificial neural networks: methodological advances and biomedical applications. Available at: https://books.google.com/books?hl=en&lr=&id=JuaODwAAQBAJ&oi=fnd&pg=PR11&dq=Kenji,+2011+ann&ots=XWjQ_S9KXd&sig=Ph8jFONuvXvX08yM8OzSZpi-v8.

K.-I. Kim et al., 2015 Toll Fraud Detection of VoIP Service Networks in Ubiquitous Computing Environments, Int. J. Distrib. Sens. Networks, vol. 2015, 2015. Available at: https://journals.sagepub.com/doi/full/10.1155/2015/276408.

Lakshmi et al., 2018, Machine Learning For Credit Card Fraud Detection System. Available at: https://www.academia.edu/download/106080716/ijaerv13n24_18.pdf.

Liu et al, 2017, Regional Concept Drift Detection and Density Synchronized Drift Adaptation. Available at: https://opus.lib.uts.edu.au/handle/10453/126374.

Mais et al., 2018, Detection of Wangiri Telecommunication Fraud Using Ensemble Learning. Available at: https://www.researchgate.net/profile/George-Sammour/publication/333232206_Detection_of_Wangiri_Telecommunication_Fraud_Using_Ensemble_Learning/links/5d07d7f5a6fdcc35c155c208/Detection-of-Wangiri-Telecommunication-Fraud-Using-Ensemble-Learning.pdf.

Metz et al., 2016, Discrete sequential prediction of continuous actions for deep rl. Available at: https://arxiv.org/abs/1705.05035.

Mukta et al., 2009, Neural networks and statistical techniques: A review of applications. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0957417407004952

Mundia et al., 2024, Assessing Mobile Network Fraud Threats and Prevention Strategies in Kenya. Available at: https://www.journals.eanso.org/index.php/eajit/article/view/2212.

Nassir, 2020, Study to use NEO4J to analysis and detection SIM-BOX fraud. Available at: https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Study+to+use+NEO4J+to+analysis+and+detection+SIM-BOX+fraud&btnG=.

Nitesh et al., 2002, SMOTE: synthetic minority over‐sampling technique. Available at: https://cir.nii.ac.jp/crid/1370017282217666450.

Niveditha et al., 2019, Credit Card Fraud Detection Using Random Forest Algorithm. Available at: https://www.academia.edu/download/65294900/CSEIT195261.pdf.

N. Lu et al., 2014, Concept drift detection via competence models. Available at: https://www.sciencedirect.com/science/article/pii/S0004370214000034.

N. Lu et al., 2016, An online competence-based concept drift detection algorithm. Available at: https://link.springer.com/chapter/10.1007/978-3-319-50127-7_36.

Ogwueleka, 2009, fraud detection in mobile communications networks using user profiling and classification techniques. Available at: https://core.ac.uk/download/pdf/37370566.pdf.

Paliwal, M., & Kumar, U. A., 2009. Neural Networks and Statistical Techniques: A Review of Applications. Expert System with Application. 36: 2–17. Available at: https://www.sciencedirect.com/science/article/pii/S0957417407004952.

Philip et al., 2015, On arbitration, arbitrage and arbitrariness in financial markets and their governance: unpacking LIBOR and the LIBOR scandal. Available at: https://www.researchgate.net/profile/Philip-Ashton-2/publication/275556489_On_Arbitration_Arbitrage_and_Arbitrariness_in_Financial_Markets_and_Their_Governance_Unpacking_LIBOR_and_the_LIBOR_Scandal/links/5b10028faca2723d99775354/On-Arbitration-Arbitrage-and-Arbitrariness-in-Financial-Markets-and-Their-Governance-Unpacking-LIBOR-and-the-LIBOR-Scandal.pdf.

P. Saravanan et al., 2014 Data mining approach for subscription-fraud detection in telecommunication sector, Contemp. Eng.Sci., vol. 7, no. 9–12, pp. 515–522, 2014. Available at: http://m-hikari.com/ces/ces2014/ces9-12-2014/subramaniyaswamyCES9-12-2014.pdf.

Raghavendra et al., 2011, Credit Card Fraud Detection Using Neural Network, Available at: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=0419c275f05841d87ab9a4c9767a4f997b61a50e.

Roger, 2023, SIM Swapping Attacks for Digital Identity Theft: A threat to financial services and beyond. Available at:  https://www.researchgate.net/profile/Roger-Hallman/publication/376612643_SIM_Swapping_Attacks_for_Digital_Identity_Theft_A_threat_to_financial_services_and_beyond/links/658091b20bb2c7472bf3dd84/SIM-Swapping-Attacks-for-Digital-Identity-Theft-A-threat-to-financial-services-and-beyond.pdf.

Rok et al., 2013, SMOTE for high-dimensional class-imbalanced data. Available at: https://link.springer.com/article/10.1186/1471-2105-14-106.

Roselina et al., 2016, Potential ANN prediction model for multiperformances WEDM on Inconel 718. Available at: https://link.springer.com/article/10.1007/s00521-016-2796-4.

R. Sallehuddin et al, 2015. Detecting SIM box fraud by using support vector machine and artificial neural network, J. Teknol., vol. 74, no. 1, pp. 137–149, 2015. Available at: https://journals.utm.my/jurnalteknologi/article/view/2649.

Ruinan et at., 2018, Sequential Behavioral Data Processing Using Deep Learning and the Markov Transition Field in Online Fraud Detection. Available at: https://arxiv.org/abs/1808.05329.

Salganicoff, 1997, Tolerating Concept and Sampling Shift in Lazy Learning Using Prediction Error Context Switching. Available at: https://link.springer.com/chapter/10.1007/978-94-017-2053-3_5.

Samira et al., 2018, A Survey on Deep Learning: Algorithms, Techniques,and Applications. Available at: https://dl.acm.org/doi/abs/10.1145/3234150.

Sheo et al.,2022, Credit Card Fraud Detection Using Support Vector Machine, Available at: https://link.springer.com/chapter/10.1007/978-981-16-6407-6_3.

S. Xu et al., 2017, Concept drift learning with alternating learners. Avalable at: https://ieeexplore.ieee.org/abstract/document/7966109/.

Snehal et al., 2019, Awareness of Sim Swap Attack. Available at: https://d1wqtxts1xzle7.cloudfront.net/59921743/215_Awareness_of_Sim_Swap_Attack20190703-77135-1etachf-libre.pdf?1562159287=&response-content-disposition=inline%3B+filename%3DAwareness_of_Sim_Swap_Attack.pdf&Expires=1718043928&Signature=MK9zvJb6Gps287s3jEyjpkvO9~9bEfb~dwP4l3Co2O9oqWO96DrTn7wzCH29Ua3Bm~1tOmFAxlD9IdYAgkXsfaXIgMW4okfjaR1dxmelM7qAVaWuU8VjCxgp2dOGUmc5K43DmebUySV~6s-29e6NQOmI5deRql02yvQzRlMnanNZ-x5p7KR7ldVxQ0cidc6XhUCGCxtzVO8x~aie3Wgt0a2QCLyqi~4oCG1c8Tr7mlRFRb2mo0bB1lGWqXxTM0qjuJschVcxs-gzD2qydq-6jb1cuCwqux6~LSABs9LtazrW1UkQ3EuPbq1eXy7kw7xonn2Q1Rd4zkGcUx-bllV6Gg__&Key-Pair-Id=APKAJLOHF5GGSLRBV4ZA.

S. Priya1et al., 2023, Deep learning framework for handling concept drift and class imbalanced complex decision‑making on streaming data. Available at: https://link.springer.com/article/10.1007/s40747-021-00456-0.

Steven et al.,2018, Artificial Neural Networks. Available at: https://www.igi-global.com/chapter/artificial-neural-networks/183727.

Suryani et al., 2022, Fraud Triangle Perspective: Artificial Neural Network Used in Fraud Analysis. Available at: https://search.proquest.com/openview/fb2104c325ba267b9ca22ab48cb7589b/1?pq-origsite=gscholar&cbl=1046413&casa_token=sgH0Ka6YncsAAAAA:iWgnQ9x6rTbZFsbi5YDMykuJ4M1M2x_3sCg0qG_u-Z_Iw4rnq1slaWFMXi2LB4F-sSQFDgkjUt4.

S. Yu, et al., 2017, Concept drift detection with hierarchical hypothesis testing. Available at: https://epubs.siam.org/doi/abs/10.1137/1.9781611974973.86.

Tajwar et al., 2024, LSTMDD: an optimized LSTM-based drift detector for concept drift in dynamic cloud computing. Available at: https://peerj.com/articles/cs-1827/.

T.J. Watson Research Center, 2001, An empirical study of the naive Bayes classifier. Available at: https://www.cc.gatech.edu/home/isbell/classes/reading/papers/Rish.pdf.

Tsymbal, 2004, The problem of concept drift: definitions and related work. Available at: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=30eac73e9b482bc28b5b68cd585557de48d0618f.

Tung-Duong et al., 2021, Customs Fraud Detection in the Presence of Concept Drift. Available at: https://ieeexplore.ieee.org/abstract/document/9679911/.

Udayakumar et al., 2023, Integrated SVM-FFNN for Fraud Detection in Banking Financial Transactions. Available at: https://www.researchgate.net/profile/Pbharath-Kumar-Chowdary/publication/376179858_Integrated_SVM-FFNN_for_Fraud_Detection_in_Banking_Financial_Transactions/links/657365f26610947889aec12a/Integrated-SVM-FFNN-for-Fraud-Detection-in-Banking-Financial-Transactions.pdf.

Vapnik V. 1995. The Nature of Statistical Learning. Available at: https://cir.nii.ac.jp/crid/1370846644385113856.

W. Heng et al., 2015, Concept Drift Detection for Streaming Data. Available at: https://ieeexplore.ieee.org/abstract/document/7280398/.

Widmer, 1993, Learning in the presence of concept drift and hidden contexts. Available at: https://link.springer.com/article/10.1023/A:1018046501280

Xizhao, et al., 2020, Recent advances in deep learning. Available at: https://link.springer.com/article/10.1007/s13042-020-01096-5.

XIN et al., 2024, Markov transition fields and auto-encoder-based preprocessing for event recognition of Φ-OTDR. Available at: https://yadda.icm.edu.pl/baztech/element/bwmeta1.element.baztech-5c079b94-fce1-42c9-a840-612843b9a04f.

X. Song et al., 2007, Learning under concept drift: an overview. Available at: https://arxiv.org/abs/1010.4784.

Yang et al., 2018, Mining Fraudsters and Fraudulent Strategies in Large-Scale Mobile Social Networks, Available at: https://ieeexplore.ieee.org/document/8744319.

Yoshua, 2015, Deep Learning and its application to CV and NLP, Available at:  http://udrc.eng.ed.ac.uk/sites/udrc.eng.ed.ac.uk/files/attachments/Deep%20Learning%20and%20its%20application%20to%20CV%20and%20NLP_0.pdf.

Y. Zhang et al., 2017, Three-layer Concept Drifting Detection in Text Data Streams. Available at: https://www.sciencedirect.com/science/article/pii/S0925231217307981.