Lagos State House Pricing Model

Project description

The proposed project, titled "Lagos State House Pricing Modle" seeks to address a pressing issue within the Lagos real estate market. This project is pivotal as it aims to provide an AI-driven solution for precisely predicting building prices. By doing so, it empowers buyers, sellers, and real estate professionals with data-driven insights to make informed decisions. The primary challenge at hand is the unpredictable nature of building prices in Lagos, Nigeria. The real estate market is subject to various factors, making it arduous for stakeholders to accurately estimate property values. As Ayodele and Olaleye (2022) emphasized, this complexity makes precise predictions a formidable task.

Currently, there is no comprehensive AI-based solution tailored for predicting building prices in Lagos. Traditional methods, including manual appraisal and historical data analysis, fall short in providing the accuracy and efficiency that AI can offer. Our proposed project represents an innovative approach to addressing this issue. Our project follows the established Data Science/Machine Learning (DS/ML) pipeline, including Data Sourcing, Data Cleaning and Preparation, ML Model Development, Model Evaluation, and Model Deployment.

1. Data Sourcing: • Collected historical building price data from the Kaggle Housing Prices in Lagos, Nigeria dataset, which includes comprehensive information on location, size, features, and past sale prices (Kaggle Housing Prices in Lagos, Nigeria Dataset).
2. Data Cleaning and Prep: • Carefully cleaned, formatted, and preprocessed the dataset to ensure its suitability for model training.
3. ML Model Development: • Utilized a range of machine learning algorithms, including decision trees, and neural networks. The choice of these algorithms is based on their suitability for capturing the complex relationship between property features and market dynamics in Lagos.
4. Model Evaluation: • Assessed the model's accuracy, reliability, and overall effectiveness using metrics like mean squared error and R-squared.
5. Model Deployment: • The project was deployed locally, and the source codes have been added to this repository

Getting Started

The project uses a Flask web framework, a machine learning model stored using Pickle, and it is recommended to use the Python Anaconda interpreter for pre-installed dependencies, This repo also contains a requirement.txt file specifying the necessary dependencies.

Prerequisites • Anaconda installed on your machine • Git installed on your machine

Clone the Repository

Create a virtual environment

1. conda create --name your_env_name python=3.8

2. Activate the virtual environment conda activate your_env_name

Install dependencies

1. Make sure you're in the directory containing requirement.txt
2. Install dependencies from requirement.txt using pip: pip install -r requirements.txt

Run the Flask Application

1. Ensure you're in the directory contain the main.py file
2. Run the flask application: python main.py This command will start the development server, and you should see output indicating that the server is running.
3. Open a web browser and navigate to http://127.0.0.1:[portNumber]/ to view the running Flask application.