Data Analysis and Machine Learning for Financial Insights
Name
Kea Kohv
Abstract
In banking, automated data analysis and machine learning tools can support high-level decision making and reduce the burden of manual work. In addition, automated data analysis can provide up-to-date insights for both the bank and its customers. The goal of this Capstone project was to explore what meaningful data-driven insights about Estonian corporate customers can be derived from transactional and socio-demographic data. The project involved developing relevant queries, enriching the dataset with open data, aggregating and visualizing, decomposing time series, detecting trends and change points, time series forecasting, and segmenting customers. By providing the corporate customer’s identification code and data period, the created tool automatically generates a report specific to that corporate customer. The report covers two perspectives: firstly, an analysis of the business's financials; secondly, profile of its customers. Only aggregated statistics are used to describe the household customers of the business. This tool benefits customer managers by saving time and enabling a better understanding of the business and the needs of the corporate customer.
Graduation Thesis language
English
Graduation Thesis type
Master's exam - Data Science
Supervisor(s)
Üllar Rannik, Jaak Vilo
Defence year
2024