Northern Kentucky University
Minimized Bank Transaction Analysis through Machine Learning
Grade Level at Time of Presentation
Junior
Major
Computer Science
Minor
Data Science and Mathematics
Institution 23-24
Northern Kentucky University
Faculty Advisor/ Mentor
Dr. Junxiu Zhou; Dr. Yangyang tao
Department
College of Informatics
Abstract
In an increasingly data-centric world, examining bank data becomes crucial for gaining insights into spending habits, making informed decisions, tailoring marketing strategies, and improving customer satisfaction. This project aims to using data analysis and machine learning technologies to analyze the bank transaction data. The dataset contains about 15 million user transaction records collected by fifth-third bank, covering main aspects from transaction amount to merchant’s category and high-level spending categories. To ensure privacy protection, the personal information of customers has been expunged from the dataset. Starting with the basics, this project explores how often transactions occur, unveiling the rhythm of financial interactions. It then delves into regional spending patterns, conducting a geographic analysis to identify key cities where transactions predominantly take place. This project also highlights preferred merchants, showcasing the top frequently visited establishments. These data analysis work acts as a roadmap, guiding the understanding of where preferred merchants and financial priorities lie. Later, machine learning methods are used to discover hidden purchase pattern among the data. This could enable predicting future spending patterns, detecting anomalies, and making more data-driven financial decisions. In essence, this ongoing research project is not just a snapshot of financial transactions but a call to continually explore data-driven insights for improved financial planning and decision-making in the future.
Minimized Bank Transaction Analysis through Machine Learning
In an increasingly data-centric world, examining bank data becomes crucial for gaining insights into spending habits, making informed decisions, tailoring marketing strategies, and improving customer satisfaction. This project aims to using data analysis and machine learning technologies to analyze the bank transaction data. The dataset contains about 15 million user transaction records collected by fifth-third bank, covering main aspects from transaction amount to merchant’s category and high-level spending categories. To ensure privacy protection, the personal information of customers has been expunged from the dataset. Starting with the basics, this project explores how often transactions occur, unveiling the rhythm of financial interactions. It then delves into regional spending patterns, conducting a geographic analysis to identify key cities where transactions predominantly take place. This project also highlights preferred merchants, showcasing the top frequently visited establishments. These data analysis work acts as a roadmap, guiding the understanding of where preferred merchants and financial priorities lie. Later, machine learning methods are used to discover hidden purchase pattern among the data. This could enable predicting future spending patterns, detecting anomalies, and making more data-driven financial decisions. In essence, this ongoing research project is not just a snapshot of financial transactions but a call to continually explore data-driven insights for improved financial planning and decision-making in the future.