Forecasting Street Speed using Alternative Data

Team: Levente Szabo, Lingbo Ji, Yaowei Zong

This repository contains code, data and analysis regarding traffic forecasting.

code: contains all data preprocessing and pyspark scripts for performing forecasting

data-processing folders include scripts used to process raw data
2019-2020_nodes.csv  validation street speed time series
Road_id.txt          validation street ids
data-etl-validation.py python script for extracting validation set
data-etl.py            python script for extracting arbitrary sized dataset
spark-model.py         python script for performing modelling directly on data in HDFS
weather-2019.csv       weather data 

data: contains variety of background literature

traffic-forecasting-modelling.ipynb notebook used to prototype models

traffic-forecasting.pdf project article and analysis of results