dbt Portfolio: Airbnb Dataset Analysis with Snowflake 🏠 + ❄

Project Overview 🚀

This project demonstrates the use of dbt (data build tool) for transforming and modeling data using the Airbnb dataset, which contains information about hosts, reviews, and listings. The project is built on Snowflake, utilizing SQL to create and manage data models.

Dataset ℹ️

The Airbnb dataset consists of three main tables 📑:

• Listings: Information about properties available on Airbnb.
• Hosts: Details about the hosts offering properties.
• Reviews: Customer feedback on the listed properties.

Architecture 🏙

The data pipeline is structured in the following way:

1. Source Tables: Raw data ingestion into Snowflake from the Airbnb dataset.
2. Staging Tables: Cleaned and transformed data using dbt's staging models.
3. Final Models:
   • Views: Derived data tables for analytical purposes.
   • Ephemeral Models: Temporary transformations that do not persist in the database.
   • Tables: Finalized materialized tables that persist in the Snowflake database.

Key Features 🔑

• dbt Configurations: Applied various dbt configurations (e.g., materialization strategies like view, table, and ephemeral) to optimize data processing.
• Data Models: Created and documented models for easy reuse.
• Snowflake as Data Warehouse: Integrated dbt with Snowflake for data transformation and storage.

Project Structure

dbt_portfolio/
├── models/
│   ├── src/
│   ├── staging/
│   ├── marts/
├── seeds/
├── snapshots/
├── logs/
├── dbt_project.yml
├── README.md
└── .gitignore

• models/src/: Contains raw source models.
• models/staging/: Contains transformed staging models.
• models/marts/: Final business logic models.

How to Use

Prerequisites

• Snowflake Account: Make sure you have access to a Snowflake account.
• dbt Installed: Install dbt using:

pip install dbt

Setup

git clone https://github.com/Mehranmzn/dbt_portfolio.git

Model Overview

• Source Tables: Extract raw data and prepare it for further processing.
• Staging Models: Clean and enrich raw data into a standardized format.
• Final Models: These models are used for analytics, reporting, or machine learning.

Example Models

• stg_listings.sql: Cleans and normalizes the listings data.
• final_hosts.sql: Aggregates host data for analytics.
• ephemeral_reviews.sql: A temporary transformation of reviews.

Contributions

Feel free to fork this repository and create a pull request to add or improve models.