PIA property development project

Start year: 2024

Summary: The project consist of two scopes: Scope 1: LLM for Enhancing Real Estate License Examination for Contract. The real estate license examination is a critical hurdle for aspiring real estate professionals. Traditional study methods include textbooks, classroom learning, and manual practice tests. However, the rapid evolution of legal and regulatory frameworks necessitates a more dynamic approach to learning. Large Language Models (LLMs), with their extensive databases and ability to simulate complex scenarios, represent a potential tool for enhancing the effectiveness of study practices. This project aims to investigate the integration of LLMs into real estate exam preparation, seeking to create a more dynamic, thorough, and adaptable learning experience. Scope 2: Large-Language-Model-Based Intent Classification for Property Management In today’s fast-paced business environment, automation has become essential for improving operational efficiency and customer satisfaction. Property management and investment companies, like PIA, handle a significant volume of inquiries and communications from landlords and tenants. Correctly classifying the intent of communications leads to a more efficient and friction-free experience for both external customers and internal property managers. The ability to interpret what the communication, regardless of its wording, is the key to solving customer problems quickly. Previously UTS team has developed a classification model to identify communications with maintenance purpose. This research proposal aims to extend the previous study by developing a robust intent classification model for multiple classes leveraging advanced AI technologies such as large language model.