/RAG_LLM_based_RealState_Recommendation_System

The simple Q/A bot is designed to respond to user queries regarding available properties by generating personalized recommendations based on the retrieved property data. The system utilizes state-of-the-art NLP models and few-shot learning techniques to deliver human-like responses tailored to the user's needs

Primary LanguageJupyter Notebook

RAG_LLM_based_RealState_Recommendation_System

The chatbot is designed to respond to user queries regarding available properties by generating personalized recommendations based on the retrieved property data. The system utilizes state-of-the-art NLP models and few-shot learning techniques to deliver human-like responses tailored to the user's needs

Real Estate Chatbot with GPT-2 and LangChain

This repository contains a real estate recommendation chatbot built using Hugging Face's GPT-2 and LangChain. The chatbot is designed to respond to user queries regarding available properties by generating personalized recommendations based on the retrieved property data. The system utilizes state-of-the-art NLP models and few-shot learning techniques to deliver human-like responses tailored to the user's needs.

Key Features

  • Property Query Handling: Users can query the system in natural language (e.g., "I need a villa in OBOUR") and receive property recommendations.
  • GPT-2 for Response Generation: GPT-2 generates human-like, context-aware responses, providing detailed property information based on user queries.
  • LangChain for Prompt Engineering: LangChain is used to structure prompts with few-shot examples to improve response consistency and relevance.
  • Real-Time Property Matching: The system simulates retrieving property listings based on the user's preferences, including details like price, location, and property type.
  • Flexible Integration: The chatbot can be integrated into real estate websites or mobile apps to assist users in finding properties based on their criteria.

Technologies

  • Transformers (Hugging Face): Pre-trained GPT-2 model for generating natural language responses.
  • LangChain: Framework for creating structured prompts and enabling few-shot learning for improved response generation.
  • Pandas: For managing and manipulating property data.
  • Python: Core programming language for developing the chatbot.