🏡 Property Advisor AI Chatbot

A real estate AI tool designed to assist property investors and buyers in Lagos by evaluating and recommending the best properties based on their preferences. The chatbot utilizes OpenAI's GPT-4 to provide ranked property suggestions along with reasoning, making property decision-making efficient and data-driven.

Project Overview

The Property Advisor AI Chatbot helps real estate investors and homebuyers quickly identify suitable properties in Lagos using AI. By processing user input (budget, location, property type, and number of bedrooms), the chatbot matches this information with available property data and ranks the most suitable properties based on price, location, and amenities.

Problem Statement

Manual property research can be time-consuming, inconsistent, and error-prone. Buyers often face challenges such as fluctuating prices, multiple listings, and varying property quality, making the decision-making process complicated.

Solution

This AI-driven property advisor integrates OpenAI’s GPT-4 to analyze a dataset of properties and deliver top recommendations. Users provide their preferences, and the AI ranks the most suitable properties, offering detailed reasoning for each recommendation. The system aims to help users make quicker, more informed decisions.

---

Features

• Property Recommendation: Input preferences (budget, location, property type, bedrooms) and get the top 3 properties ranked by GPT-4 based on your input.
• Natural Language Chat: Engage in a dynamic conversation with the AI to refine property preferences.
• User-Friendly Interface: Simple, intuitive design built with Gradio for easy interaction.
• Real-Time Results: Immediate suggestions based on real-time property data (static dataset used for demo purposes).

---

Dataset

• Source: Due to Hugging Face’s 10MB size limitation, the dataset is hosted on Google Sheets for demo purposes. The dataset is synthetic but representative of typical property listings in Lagos.
• Dataset Link: Property Dataset - Google Sheets
• Columns Include:
  • Title
  • Price
  • Location
  • Bedrooms
  • Bathrooms
  • Size
  • Property Type
  • Score (based on a custom scoring algorithm)

---

Workflow

Data Processing

1. Initial Data Processing:

   • First, an IPython notebook (.ipynb) was used to clean, process, and structure the dataset for the chatbot's requirements. The notebook included steps such as handling missing values, filtering data, and transforming the dataset to make it ready for AI analysis.
   • The processed dataset was then saved and used for further analysis in subsequent steps.

2. Transition to Python Script:

   • After completing the data preprocessing, the IPython notebook was converted into a Python script (app.py) for deployment in Hugging Face, which does not support Jupyter notebooks natively.
   • The Python script now runs the data processing, interacts with GPT-4, and serves the chatbot application.

---

Installation & Requirements

Prerequisites

Before getting started, ensure you have the following installed:

• Python 3.9+ (Recommended: 3.9 or later)
• Git (To clone the repository)

Step-by-Step Installation

1. Clone the repository:

   First, clone the repository to your local machine using Git. Open your terminal or command prompt and run:

   git clone https://github.com/yourusername/property-advisor-ai.git

   Then, navigate into the project directory:

 cd property-advisor-ai

2. Install the required dependencies:

Next, install all the required dependencies listed in the requirements.txt file using pip. Run the following command:

pip install -r requirements.txt

This will install the necessary Python libraries, including openai, pandas, gradio, and python-dotenv.

3. Set up environment variables:

Create a .env file in the root directory of your project and add your OpenAI API key. The .env file should contain: Replace your_openai_api_key with your actual OpenAI API key.

OPENAI_API_KEY=your_openai_api_key

Run the app:

Once the environment is set up, you can run the app locally by executing:

python app.py

Usage

Once the app is running, you can interact with the Property Advisor AI Chatbot as follows:

Input your property preferences:

• Budget (₦)
• Location (e.g., Lagos)
• Property Type (e.g., Apartment)
• Number of Bedrooms

Engage with the AI:

The chatbot here will provide you with the top 3 property recommendations based on your preferences and the available dataset.

Ask Follow-up Questions:

You can engage in follow-up questions or adjust your preferences, and the chatbot will refine the recommendations accordingly.

Project Architecture

This project integrates the following technologies:

• OpenAI GPT-4: Used for natural language understanding and generating property recommendations.
• Gradio: Used for building the user interface, allowing users to interact with the chatbot.
• Pandas: Used for loading and manipulating the property dataset.
• Python-dotenv: Manages API keys securely.