Data Stories: A Collection of Case Studies in Data Analytics with Real World Datasets

Chapter 1

Enhancing Fraud Detection in Credit Card Transactions using Data Analysis Approaches

By Lew Sook Ling [0000-0003-4515-1163], Enzo Joaquin Itona Erfe, Michelle Theba & Jessen Tan Kean Tat

Abstract – In today’s world, there is an increase in online payment fraud due to the rise in digital transactions. Improved methods are needed to detect and prevent this fraud. This study examines ways to swiftly identify fraudulent transactions, ensuring the safety of digital financial systems and increasing public trust. By employing data analysis, particularly utilizing the Python programming language, the data was analysed to anticipate future occurrences. Statistical methods, including correlation analysis, were used to identify relationships between transaction timing and fraudulence. Visualisations such as time-series graphs were created to gain deeper insights into the data and make informed decisions. Bar charts and column charts were used to compare the frequency and distribution of fraudulent versus genuine transactions across various categories, such as transaction categories and time periods. Line charts were employed to observe trends and patterns in transaction activities over time, highlighting anomalies that might indicate fraud. Additionally, regression models were utilized to analyse the relationship between variables. The examination of a real-world dataset containing both genuine and fraudulent credit card transactions across various time periods was conducted. By analysing the correlation between transaction timing and fraudulence and testing various hypotheses, innovative strategies for fraud detection, enhancing the security of digital transactions for all users were proposed.

Cite this chapter as:

Lew, S. L., Erfe, E. J. I., Theba, M., & Tan, J. K. T. (2025). Enhancing Fraud Detection in Credit Card Transactions Using Data Analysis Approaches. In S. L. Lew, M. C. Leow, & T. H. Liew, Data Stories: A Collection of Case Studies in Data Analytics with Real World Datasets (pp. 3-32). MMU Press.

Chapter 2

Predicting Netflix Stock Prices using Data Analytics for Market Insights and Investment Strategies

By Lew Sook Ling [0000-0003-4515-1163], Wong Kah Chung, Lau Wen Xiang, Leong Wei Tong & Lee Jia Wei

Abstract– Predicting stock prices is still a major challenge for stakeholders, investors, and financial experts in the present day of data-driven decision-making. The purpose of this study is to determine whether it is possible to predict Netflix’s stock price by using historical data and trend analysis. While stock prices are unpredictable, it can be difficult to make decisions about investments. For this reason, analysts and stakeholders need to know how predictable stock prices during a specific period. This study focuses on conducting analysis of a five-year dataset of Netflix stock prices obtained from Kaggle. It aims to examine the predictability of Netflix’s stock prices and analyse the impact of market dynamics on its performance. This study utilised data analytics techniques, including time series analysis, seasonal decomposition, autoregressive modelling, and correlation analysis to predict Netflix stock prices. Visual tools like heatmaps, scatter plots and line charts, along with Moving Averages (MA) and regression models, identified trends, relationships and performance metrics offering insights for market and investment strategies. With a focus on market trends and performance, this study aims to provide insights into Netflix’s stock price prediction through in-depth analysis and predictive modelling. The results improve understanding of the unpredictability of Netflix’s stock price and how it affects investment decisions through historical data and market trends. This study contributes to stock market prediction and data-driven financial decision-making.

Cite this chapter as:

Lew, S.L., Wong, K.C., Lau, W.X., Leong, W.T. & Lee, J.W. (2024). Netflix Stock Price Prediction. In S. L. Lew, M. C. Leow, & T. H. Liew, Data Stories: A Collection of Case Studies in Data Analytics with Real World Datasets (pp. 28-48). MMU Press.

Chapter 3

Forecasting and Preventing Heart Failure using Data Analytics Approaches

By Lew Sook Ling [0000-0003-4515-1163], Chin Yi Sin, Yeu Nian Yong, Go Hui Ling & Ignatius Ng Wei Siong

Abstract – Cardiovascular disease is a major public health concern that affects people of all ages and includes illnesses such as high blood pressure, heart attacks, and myocardial infarction. Effective preventive strategies are necessary due to the complexity of cardiac disease. For prompt treatment, accurate assessment of the incidence and severity of cardiac disease is essential. This study uses a variety of indicators to evaluate each person’s risk for heart disease and applies data analytics techniques to determine underlying reasons. Personalised risk assessment helps to customise treatment and preventative strategies. The study begins with exploratory data analysis (EDA), followed by feature selection and data scaling to enhance model performance. The Decision Tree Classifier, known for its computational efficiency, interpretability and ability to handle non-linear relationships, is used to predict heart failure. The model’s clear decision rules make it suitable for quick implementation, offering competitive accuracy while maintaining simplicity. The findings suggest that the Decision Tree Classifier, alongside other machine learning models, effectively predicts cardiovascular disease, enabling timely treatment decisions and contributing to improved public health.

Cite this chapter as:

Lew, S. L., Chin, Y. S., Yong, Y. N., Go, H. L., & Ng, I. W. S. (2025). Forecasting and Preventing Heart Failure Using Data Analytics Approaches. In S. L. Lew, M. C. Leow, & T. H. Liew, Data Stories: A Collection of Case Studies in Data Analytics with Real World Datasets (pp. 57-85). MMU Press

Chapter 4

Investigating Relationships and Implications of Age, Body Mass Index (BMI) and Individual Medical Charges through a Case Study Analysis

By Lew Sook Ling [0000-0003-4515-1163], Liow Fong Zhi, Liew Zhao Lun & Ding Zhen Man

Abstract – Age, body mass index (BMI), and personal medical costs are known to have complex relationship that are crucial in today’s changing healthcare system. Medical costs and BMI are also correlated with age. The objective of this study is to provide insight into the complex relationships between these factors and how they affect overall health and wealth. Analysing how BMI relates to age reveals normal physiological changes that affect body composition and general health. Age-related health biases highlight how important BMI is for determining the current state of health. Moreover, the relationship between age and BMI has financial implications in addition to healthy ones since aging and BMI fluctuations are also correlated with higher medical costs. The availability of affordable healthcare is restricted by these interrelated issues. Developing effective prevention initiatives requires an understanding of these relationships. This study has provided data analytic approaches that address health and financial issues by evaluating the correlations between age, BMI and medical costs.

Cite this chapter as:

Lew, S. L., Liow, F. Z., Liew, Z. L., & Ding, Z. M. (2025). Investigating Relationships and Implications of Age, Body Mass Index (BMI) and Individual Medical Charges through a Case Study Analysis. In S. L. Lew, M. C. Leow, & T. H. Liew, Data Stories: A Collection of Case Studies in Data Analytics with Real World Datasets (pp. 87-109). MMU Press.

Chapter 5

Identifying Worldwide Patterns in Life Expectancy using Data Analytics

By Lew Sook Ling [0000-0003-4515-1163], Wong Jing Hong, Louis Law Jing Xin, Au Jun Jie & Shing Yu Bei

Abstract – Identifying trends that deviate from global patterns is crucial, as it highlights issues requiring focused policies. Examining patterns in human longevity is essential for developing targeted interventions to address health challenges and improve outcomes in specific communities or regions. This case study aims to analyse global trends in life expectancy, with a focus on understanding the growth of average life spans for men and women. The primary goal is to examine these patterns and assess the factors influencing life expectancy worldwide. To achieve this, data analytics techniques are applied to the “Average Life Span Trends for Males and Females” dataset, exploring both local and global trends in life expectancy. By examining these patterns, the study provides insights into key aspects of human civilization, particularly as it relates to life expectancy. The findings highlight the power of data analytics in uncovering hidden truths about social developments and contributing to scientific knowledge in the field of public health.

Cite this chapter as:

Lew, S. L., Wong, J. H., Law, L. J. X., Au, J. J., & Shing, Y. B. (2025). Identifying Worldwide Patterns in Life Expectancy Using Data Analytics. In S. L. Lew, M. C. Leow, & T. H. Liew, Data Stories: A Collection of Case Studies in Data Analytics
with Real World Datasets (pp. 111-134). MMU Press.

Chapter 6

Analysing the Relationships Among Physical Attributes, Eating Patterns and Obesity using Data Analytics Approaches

By Lew Sook Ling [0000-0003-4515-1163], Wong Siong Yee, Nicholas Wong Ding Xuan, Lim Jian Hang & Tan Jun
Hong

Abstract – Obesity is a complicated illness is influenced by genetic, environmental and behavioural variables, nutrition and exercise levels. Obesity rates are increased by modern lifestyles that are marked by lack of exercise, easy and affordable high-calorie foods. Therefore, understanding how food habits contribute to obesity is necessary for efficient prevention. Weight gain is caused by a few factors including eating meals high in energy, large portions and at irregular times. However, the relationship between diet and obesity is further complicated by psychological, economic and cultural factors. While exercise has a big impact on weight control and energy balance, modern society and technology contribute to the increase in passive lifestyles. These phenomenon raises the risk of obesity and related chronic diseases with reducing energy expenditure and causing metabolic disorders. Data analytics is necessary to understand the complicated relationships that exist between physical activity, eating patterns and the rising rate of obesity. By using data analytics, this study has identified the trends and differences that help developing evidence-based public health policies and focused prevention strategies. This study has identified the variables influencing obesity rates, examined their relationships and proposed practical prevention and management techniques.

Cite this chapter as:

Lew, S. L., Wong, S. Y., Wong, N. D. X., Lim, J. H., & Tan, J. H. (2025). Analysing the Relationships Among Physical Attributes, Eating Patterns and Obesity Using Data Analytics Approaches. In S. L. Lew, M. C. Leow, & T. H. Liew, Data
Stories: A Collection of Case Studies in Data Analytics with Real World Datasets (pp. 135-151). MMU Press.