Leveraging machine learning insights to optimize marketing strategies
Zemskova, Emilia (2024)
2024
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2024121335424
https://urn.fi/URN:NBN:fi:amk-2024121335424
Tiivistelmä
This thesis focused on the use of machine learning techniques, specifically logistic regression, to optimize marketing strategies for the Business-to-Consumer (B2C) sector. The study utilized a dataset from a Portuguese banking institution's direct marketing campaigns, which provided insights into customer responses to financial product offers. The main objective was to apply data-driven methods to enhance marketing outcomes, including customer segmentation, campaign timing, and personalized communication.
A logistic regression model was developed to classify customer behavior, with a focus on predicting whether clients would subscribe to a bank term deposit. Through this analysis, the thesis explored how machine learning can help marketers better understand customer preferences and tailor their strategies accordingly. One key finding was the importance of using OneHotEncoder to handle categorical data effectively, avoiding bias in model predictions. Additionally, the thesis examined the potential for optimizing marketing efforts by identifying the best times to contact customers based on historical engagement data.
The results of the logistic regression model indicated some success in predicting customer behavior, though challenges such as multicollinearity and model accuracy were noted. The conclusion suggests that further improvements could be made by exploring more advanced machine learning techniques. Overall, the thesis demonstrates the value of integrating machine learning insights into marketing strategies to drive better customer engagement and conversion rates.
A logistic regression model was developed to classify customer behavior, with a focus on predicting whether clients would subscribe to a bank term deposit. Through this analysis, the thesis explored how machine learning can help marketers better understand customer preferences and tailor their strategies accordingly. One key finding was the importance of using OneHotEncoder to handle categorical data effectively, avoiding bias in model predictions. Additionally, the thesis examined the potential for optimizing marketing efforts by identifying the best times to contact customers based on historical engagement data.
The results of the logistic regression model indicated some success in predicting customer behavior, though challenges such as multicollinearity and model accuracy were noted. The conclusion suggests that further improvements could be made by exploring more advanced machine learning techniques. Overall, the thesis demonstrates the value of integrating machine learning insights into marketing strategies to drive better customer engagement and conversion rates.