Skip to content
This repository has been archived by the owner on Jan 8, 2020. It is now read-only.
/ TFG Public archive

Time Series Forecasting using K-Nearest Neighbors Algorithm (Parallel approach)

License

Notifications You must be signed in to change notification settings

Dani-Basta/TFG

Repository files navigation

Abandoned branch

For further releases visit https://github.com/Grasia/knnp

knnp : Time Series Prediction using K-Nearest Neighbors Algorithm (Parallel)

TFG

=========== First release has been developed as an End-of-Degree Project.

Purpose

This package intends to provide R users or anyone interested in the field of time series prediction the possibility of aplying the k-nearest neighbors algorithm to time series prediction problems. Two main functionalities are provided:

  • Time series prediction using this method.
  • Optimization of parameteres k and d of the algorithm.

All the code involved has been optimized to:

  • Parallelize critic components as the process of optimization of parameteres k and d or the calculation of distances.
  • Use memory efficiently.

Authors

  • Daniel Bastarrica Lacalle
  • Javier Berdecio Trigueros

License

AGPL-3