Apartment Price Prediction

This repository contains code for predicting apartment prices using machine learning techniques. The dataset used is a collection of apartment listings with various features such as size, number of bedrooms, bathrooms, location, and amenities.

Dataset

The dataset contains the following columns:

• id: A unique identifier for each apartment listing (not used as a feature)
• category: The category of the listing (not used as a feature)
• title: The title of the listing (not used as a feature)
• body: The description of the listing (not used as a feature)
• amenities: The amenities offered by the apartment (not used as a feature)
• bathrooms: The number of bathrooms in the apartment (feature)
• bedrooms: The number of bedrooms in the apartment (feature)
• currency: The currency of the listed price (not used as a feature)
• fee: Additional fees associated with the listing (not used as a feature)
• has_photo: Indicates whether the listing has a photo (not used as a feature)
• pets_allowed: Indicates whether pets are allowed in the apartment (feature)
• price: The listed price of the apartment (target variable)
• price_display: The displayed price format (not used as a feature)
• price_type: The type of price (e.g., monthly, weekly) (not used as a feature)
• square_feet: The area of the apartment in square feet (feature)
• address: The address of the apartment (not used as a feature)
• cityname: The city where the apartment is located (feature)
• state: The state where the apartment is located (feature)
• latitude: The latitude coordinate of the apartment (feature)
• longitude: The longitude coordinate of the apartment (feature)
• source: The source of the listing (not used as a feature)
• time: The time when the listing was posted (not used as a feature)

Data Preprocessing - Subject to Change

The code provided in the Python file performs the following data preprocessing steps:

1. Loading the dataset from a CSV file.
2. Dropping irrelevant columns.
3. Handling missing values in the bathrooms, bedrooms, latitude, longitude, pets_allowed, cityname, and state columns.
4. One-hot encoding categorical features (pets_allowed, cityname, and state).
5. Splitting the data into training and test sets.
6. Ensuring no NaN values are present in the training and test sets.

Models

The repository currently includes the following models:

1. Linear Regression

A simple linear regression model is implemented as a baseline. The code evaluates the model's performance using mean squared error (MSE) and R² score.

2. Random Forest Regression

A random forest regression model is also implemented, and its performance is evaluated using MSE and R² score. The code demonstrates that the random forest model outperforms the linear regression model for this dataset.

Getting Started

1. Clone the repository:

git clone https://github.com/your_username/apartment-price-prediction.git

2. Install the required Python packages:

pip install -r requirements.txt

3. Navigate to the desired notebook (e.g., paste.txt) and run the cells.

Future Work

Future work may include:

• Exploring other machine learning models, such as gradient boosting or neural networks.
• Conducting more extensive feature engineering and selection.
• Investigating the impact of different hyperparameters on model performance.
• Deploying the best-performing model as a web application or API.

Acknowledgments

• https://archive.ics.uci.edu/dataset/555/apartment+for+rent+classified