Skip to content

vinod-polinati/Uber-Data-Analytics

Repository files navigation

Uber Data Analytics | Data Engineering GCP Project

This project uses GCP Storage, Python, Compute Instance, Mage Data Pipeline Tool, BigQuery, and Looker Studio for performing data analytics on Uber data and building a Dashboard.

Architecture

Architecture

Technology Used

  • Programming Language: Python

Google Cloud Platform:

  1. Google Storage
  2. Compute Instance
  3. BigQuery
  4. Looker Studio

Modern Data Pipeline Tool: Mage

Dataset Used

TLC Trip Record Data includes pick-up/drop-off dates and times, locations, trip distances, fares, rate types, payment types, and passenger counts.

For more information about the dataset:

  1. Website
  2. Data Dictionary

Data Model

Uber Data Model

Workflow: Mage -> GCP (BigQuery) -> Looker Studio



Dashboard

Uber Dashboard