Skip to content

beak-insights/conversational-banking

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Conversational Banking -

Large Language Model Mobile Chatbot for banks in Zimbabwe

Abstract

This study addresses challenges in Zimbabwe’s banking sector, specifically the accessibility and efficiency of customer service. The primary aim is to develop a mobile banking chatbot powered by Large Language Models (LLMs) to streamline access to financial services and support user-friendly interactions. The research uses a mixed-methods approach, incorporating quantitative and qualitative techniques to assess the chatbot’s effectiveness, reliability, and scalability. Methodologies include prompt engineering, API mockups, and usability testing through customer feedback. The conclusions indicate that the LLM chatbot can significantly improve customer satisfaction by automating routine inquiries and providing personalized responses. Key contributions of the study include demonstrating how LLMs can enhance service accessibility, operational efficiency, and customer engagement in a banking context. The project also highlights the benefits of integrating digital tools to drive financial inclusion, making banking services more accessible, especially to underserved populations. Recommendations for future development include refining the chatbot’s contextual understanding and exploring integration with additional financial services to meet evolving customer needs.

Objectives

  • Integrate LLM: Understand and process complex customer queries.
  • API Mockups: Simulate key banking functionalities.
  • Mobile Interface: Seamless integration with the chatbot.
  • System Prompt Engineering: Guide the LLM's responses.

Key Findings

  • The LLM chatbot significantly improves customer satisfaction.
  • Automates routine inquiries and provides personalized responses.
  • Enhances service accessibility, operational efficiency, and customer engagement.

Future Development Recommendations

  • Integrate Additional Banking Services: Expand capabilities to include loan applications, investment advice, and insurance offerings.
  • Multilingual Capabilities: Support multiple languages to increase accessibility.
  • AI Advancements: Integrate sentiment analysis and machine learning-based personalization.
  • Third-Party Service Integration: Offer value-added services like bill payments and e-commerce transactions.
  • Voice Interaction Capabilities: Provide a hands-free banking experience.
  • Continuous User Feedback: Regularly collect feedback to guide future updates.

Releases

No releases published

Packages

No packages published