Edwin574 / Traffic-Prediction-Model

Creating a model that can be used to predict traffic in a particular road

2 stars 0 forks source link

Feature Engineering #2

Open Allanvince opened 1 year ago

Allanvince commented 1 year ago

Goal: Enhance predictive performance of time series analysis by implementing feature engineering techniques.

Techniques to Implement:

[ ] 1. Rolling Window Statistics: Calculate moving averages, rolling sums, or rolling standard deviations over a specific time window.
[ ] 2. Date/Time Features: Extract day of the week, hour of the day, month.

Purpose and Impact:

Rolling Window Statistics provide insights into trend and variability, capturing temporal patterns in the traffic volume data.
Date/Time Features capture seasonality and temporal patterns, enabling models to capture daily, weekly, and monthly variations.

Rontim commented 1 year ago

Checkout in your local repository

git fetch origin
git checkout 2-feature-engineering