Customer Segmentation using RFM Analysis and K-Means Clustering to enhance Marketing Strategies; Analysis of Algorithmic Bias in Customer Segmentation Models
Patel, Harshil, School of Engineering and Applied Science, University of Virginia
Earle, Joshua, EN-Engineering and Society, University of Virginia
Morrison, Briana, EN-Comp Science Dept, University of Virginia
My technical and STS research projects are connected by the idea of customer segmentation using machine learning algorithms. Customer segmentation refers to the process of dividing a customer base into groups based on common characteristics, such as demographics, behaviors, or preferences. Machine learning algorithms can help businesses analyze large amounts of customer data and gain insights into customer behavior and preferences, enabling them to develop targeted marketing strategies and improve performance. However, there are potential ethical concerns related to the use of machine learning algorithms in customer segmentation, as these algorithms may introduce biases that could have negative consequences for some customer segments.
In my technical report, I aim to demonstrate a customer segmentation approach using the k-means clustering algorithm and RFM analysis model to create clusters of customers based on their purchasing behavior. The project's objective is to help businesses target customers effectively, resulting in improved performance and customer services. By utilizing the RFM model and k-means clustering, we can group customers into four distinct clusters based on their recency, frequency, and monetary value. This segmentation allows businesses to identify high-value and low-value customer segments and those that require targeted marketing efforts. The project methodology comprises five stages: business understanding, data understanding, data preparation, modeling, and evaluation. The results show that the k-means clustering algorithm, coupled with the RFM model, is effective in clustering customers based on their purchasing behavior. The insights from the model enable businesses to analyze their customers, allowing them to develop targeted marketing strategies that enhance customer retention and acquisition. Future work of this project involves incorporating additional data sources such as tenure, enhancing the web model's features, and conducting additional testing and evaluation to improve the clustering accuracy.
In my STS research project, I explore the impact of AI-powered customer segmentation models on businesses and consumers. I analyzed various approaches to customer segmentation models that were meant to define target markets and precisely identify customer segments, but biases affected the system along the way. I also evaluate the limitations of these models within the lens of the Actor-Network Theory and Social Construction of Technology. The study highlights the need for companies to prioritize diversity and inclusivity in their data collection and analysis processes to avoid biases being introduced into these customer segmentation models, which can have negative consequences. Ultimately, the responsible integration of AI in customer segmentation models can help companies build trust with their customers, create personalized experiences, and drive growth in the increasingly competitive business landscape.
By exploring these two projects simultaneously, I gained an understanding of why and how machine learning algorithms can be used to create customer segments and the potential ethical concerns associated with their use. While the technical report focused on developing an effective customer clustering approach using k-means clustering and the RFM model, the STS research paper evaluated the potential limitations and ethical considerations related to the use of AI technology in customer segmentation models. By examining these two projects side by side, I realized the importance of considering the ethical implications of machine learning algorithms in developing effective marketing strategies. The responsible use of AI in customer segmentation models can lead to better customer experiences and enhanced business performance while avoiding potential ethical concerns.
BS (Bachelor of Science)
Customer segmentation, RFM analysis, K-means clustering, Machine learning algorithms, Marketing strategies, Business performance, Algorithmic bias, Data collection, Data analysis, Actor-Network Theory, Social Construction of Technology, Personalized experiences, Targeted marketing, Customer retention, Customer acquisition
Thank you for taking the time to read my thesis on customer segmentation using RFM analysis and K-means clustering to enhance marketing strategies, as well as the analysis of algorithmic bias in customer segmentation models. Through this thesis, I aim to demonstrate how businesses can utilize machine learning algorithms to effectively analyze customer data and gain insights into customer behavior and preferences, while also highlighting the importance of considering ethical implications and potential biases associated with these algorithms.
I would like to take a moment to thank my technical advisor, Briana Morrison, and my STS advisor, Joshua Earle, for their invaluable guidance and support throughout the research process. Their insights, feedback, and encouragement have been instrumental in the completion of this thesis.
I hope that my research inspires readers to think critically about the role of AI in marketing and the potential ethical considerations associated with its use. I also hope that my findings will help businesses develop more effective marketing strategies that enhance customer experiences, retention, and acquisition while minimizing potential negative consequences.
Once again, thank you for your interest in my research.
Harshil Anishkumar Patel
All rights reserved (no additional license for public reuse)