Data

Tools

Indexes

Applications

Experiments

Open Source Science

path_based_traffic_flow_prediction

Forecast future traffic flow on a road network.

https://github.com/stratoskar/path_based_traffic_flow_prediction

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (11.0%) to scientific vocabulary

Keywords

big-data lstm-neural-networks time-series-forecasting xgboost
Last synced: 6 months ago · JSON representation

Repository

Forecast future traffic flow on a road network.

Basic Info
  • Host: GitHub
  • Owner: stratoskar
  • License: mit
  • Language: Jupyter Notebook
  • Default Branch: main
  • Homepage:
  • Size: 159 MB
Statistics
  • Stars: 3
  • Watchers: 2
  • Forks: 3
  • Open Issues: 0
  • Releases: 0
Topics
big-data lstm-neural-networks time-series-forecasting xgboost
Created over 2 years ago · Last pushed over 1 year ago
Metadata Files
Readme License Citation

README.md

Path-based traffic flow prediction

This repository contains the code files produced during an extensive research in the field of mobility data analytics.

Repository overview

  • Objective: We are trying to predict the forthcoming traffic flow volume for specific routes within San Francisco, California, leveraging insights gleaned from historical data.
  • Methodology: We perform time series forecasting using machine and deep learning models. We use the Strict Path Queries algorithm to measure traffic flow in each path accurately.
  • Data: We use traffic flow data of Yellow Taxis that are moving within the city or San Francisco, California. We use this original dataset to generate the final time series data. This entire process is described in the Jupyter notebook files inside the Python_Code folder. Notebooks have increment numbers that define the order of their execution.
  • Number of Paths: A total of 100 paths (or pathways) are used for conducting forecasts.

Dependencies

To run the code in this repository, ensure you have the latest version of Python installed. The required libraries are listed in the Necessary_Libraries.txt file. You can install them using pip or conda commands.

About Me

My name is Efstratios Karkanis, and I have finished my studies in computer science at the University of Piraeus. For any inquiries or to establish contact, please feel free to reach out to me at stratoskarkanis2@gmail.com.

Feel free to explore the code and the insights gained from this project. Contributions and feedback are always welcome!

Owner

  • Name: Efstratios Karkanis
  • Login: stratoskar
  • Kind: user
  • Location: Piraeus, Greece

I am a 21-year-old student and currently, I am studying computer science at the University of Piraeus.

GitHub Events

Total
  • Fork event: 1
Last Year
  • Fork event: 1