Design and Evaluation of AI-Driven Personal Finance Management Tools

Thesis Abstract

The rapid proliferation of digital financial services has underscored the necessity for innovative tools that enhance individual financial management through artificial intelligence (AI), addressing persistent issues of financial illiteracy, poor budgeting skills, and ineffective savings behavior. This study aims to design, develop, and evaluate AI-driven personal finance management tools that can effectively assist users in making informed financial decisions, optimizing budgeting, savings, and expenditure patterns. The specific objectives include identifying user needs and preferences, designing an AI-based prototype aligned with behavioral economic principles, assessing usability and user satisfaction, and measuring the impact of the tool on users' financial behaviors over time. The research adopts a mixed-methods approach, beginning with a qualitative exploratory phase involving focus group discussions with 30 diverse financial technology users to gather insights into user requirements and expectations. This is followed by the development of a functional prototype based on machine learning algorithms, including predictive analytics, recommendation systems, and adaptive interfaces. The quantitative phase employs a quasi-experimental research design involving a sample of 200 adult users recruited via stratified random sampling from banking institutions and financial advisory platforms. Participants are divided equally into test and control groups; the test group interacts with the AI-driven tool for a period of three months, while the control group continues with traditional financial management methods. Data collection instruments comprise structured questionnaires measuring financial literacy levels, usability, and satisfaction, complemented by objective financial data extracted through the tool’s analytics dashboards. The reliability and validity of these instruments are established through Cronbach’s alpha tests (above 0.85) and pilot testing with 30 respondents. Data analysis employs descriptive statistics to profile the sample, followed by inferential techniques such as paired t-tests and ANCOVA to evaluate changes in financial behaviors and satisfaction levels pre-and post-intervention. Structural equation modeling (SEM) is utilized to examine the relationships among user engagement, perceived utility, and financial outcomes, guided by the theoretical framework of the Theory of Planned Behavior and the Behavioral Economics Theory. Expected findings indicate that the AI-driven financial management tool significantly enhances user engagement, financial literacy, and prudent spending patterns compared to traditional methods. The study anticipates discovering complex interactions between user characteristics, system usability, and behavioral outcomes, with predictive models demonstrating substantial accuracy in financial forecasting and personalized advice provision. These results are expected to contribute to the extant literature by illustrating the efficacy of integrating AI technologies with behavioral theories to foster sustainable financial habits. The study's primary contribution to knowledge lies in providing empirical evidence on the design and operational effectiveness of AI-based personal finance tools, establishing a framework for future development and policy formulation in digital financial consultancy. It also offers insights into user-centered design processes that cater to diverse demographic groups, emphasizing the importance of personalization and behavioral cues in fintech applications. The main conclusion underscores that AI-driven tools can serve as viable solutions to improve financial decision-making processes; however, their success depends on user-centric design, ease of use, and behavioral reinforcement mechanisms. Based on these findings, the study recommends that financial institutions and fintech providers incorporate adaptive AI systems tailored to individual user profiles and financial goals. It further advocates for ongoing user engagement and feedback mechanisms to refine tool functionalities continually. For future research, it suggests exploring longitudinal impacts of AI financial tools across different socioeconomic groups and integrating advanced natural language processing features for improved user interaction.