mlmts1.1.2 package

Machine Learning Algorithms for Multivariate Time Series

dis_2dsvd

Constructs a pairwise distance matrix based on two-dimensional singula...

dis_cor

Constructs a pairwise distance matrix based on auto and cross-correlat...

dis_dtw_1

Constructs a pairwise distance matrix based on multivariate dynamic ti...

dis_dtw_2

Constructs a pairwise distance matrix based on multivariate dynamic ti...

mts_forecasting

A forecasting procedure for MTS based on lag-embedding matrices

dis_eros

Constructs a pairwise distance matrix based on the Eros distance measu...

dis_eucl

Constructs a pairwise distance matrix based on the Euclidean distance

dis_frechet

Constructs a pairwise distance matrix based on the Frechet distance

dis_gcc

Constructs a pairwise distance matrix based on the generalized cross-c...

dis_hwl

Constructs a pairwise distance matrix based on feature extraction

mts_plot

Constructs a plot of a MTS

dis_lpp

Constructs a pairwise distance matrix based on locality preserving pro...

dis_mahalanobis_dtw

Constructs a pairwise distance matrix based on a dissimilarity combini...

dis_mahalanobis

Constructs a pairwise distance matrix based on the Mahalanobis distanc...

dis_mcc

Constructs a pairwise distance matrix based on maximal cross-correlati...

dis_modwt

Constructs a pairwise distance matrix based on the maximum overlap dis...

dis_pca

Constructs a pairwise distance matrix based on Principal Component Ana...

dis_ppca

Constructs a pairwise distance matrix relying on a piecewise represent...

dis_qcd

Constructs a pairwise distance matrix based on the quantile cross-spec...

dis_qcf

Constructs a pairwise distance matrix based on the quantile cross-cova...

dis_spectral

Constructs a pairwise distance matrix based on estimated spectral matr...

dis_swmd

Constructs a pairwise distance matrix based on VPCA and SWMD

dis_var_1

Constructs a pairwise distance matrix based on the estimated VAR coeff...

dis_var_2

Model-based dissimilarity proposed by Maharaj (1999)

mc2pca_clustering

Performs the crisp clustering algorithm of Li (2019)

dis_www

Constructs a pairwise distance matrix based on feature extraction

dis_zagorecki

Constructs a pairwise distance matrix based on feature extraction

f4_classifier

Constructs the F4 classifier of López-Oriona and Vilar (2021)

knn_classifier

Constructs a nearest neighbours-based classifier and returns the predi...

mlmts

mlmts: Machine Learning Algorithms for Multivariate Time Series.

outlier_detection

Constructs the outlier detection procedure of López-Oriona and Vilar (...

plot_2d_scaling

Constructs a 2-dimensional scaling plot based on a given dissimilarity...

vpca_clustering

Performs the fuzzy clustering algorithm of He and Tan (2020).

An implementation of several machine learning algorithms for multivariate time series. The package includes functions allowing the execution of clustering, classification or outlier detection methods, among others. It also incorporates a collection of multivariate time series datasets which can be used to analyse the performance of new proposed algorithms. Some of these datasets are stored in GitHub data packages 'ueadata1' to 'ueadata8'. To access these data packages, run 'install.packages(c('ueadata1', 'ueadata2', 'ueadata3', 'ueadata4', 'ueadata5', 'ueadata6', 'ueadata7', 'ueadata8'), repos='<https://anloor7.github.io/drat/>')'. The installation takes a couple of minutes but we strongly encourage the users to do it if they want to have available all datasets of mlmts. Practitioners from a broad variety of fields could benefit from the general framework provided by 'mlmts'.

  • Maintainer: Angel Lopez-Oriona
  • License: GPL-2
  • Last published: 2024-08-18