Movelets: Movelets for Multiple Aspect Trajectory Data Mining

[Publication] [citation.bib] [GitHub] [PyPi]

The present application offers a tool, to support the user in the classification task of multiple aspect trajectories, specifically for extracting and visualizing the movelets, the parts of the trajectory that better discriminate a class. It integrates into a unique platform the fragmented approaches available for multiple aspects trajectories and in general for multidimensional sequence classification into a unique web-based and python library system. Offers both movelets visualization and classification methods.

Created on May, 2023 Copyright (C) 2023, License GPL Version 3 or superior (see LICENSE file)

Main Modules

• Methods: Methods for trajectory classification and movelet extraction;
• Tutorial: Tutorial on how to use Automatise as a Python library.

Available Classifiers (needs update):

• MLP (Movelet): Multilayer-Perceptron (MLP) with movelets features. The models were implemented using the Python language, with the keras, fully-connected hidden layer of 100 units, Dropout Layer with dropout rate of 0.5, learning rate of 10−3 and softmax activation function in the Output Layer. Adam Optimization is used to avoid the categorical cross entropy loss, with 200 of batch size, and a total of 200 epochs per training. [REFERENCE*]
• RF (Movelet): Random Forest (RF) with movelets features, that consists of an ensemble of 300 decision trees. The models were implemented using the Python language, with the keras. [REFERENCE*]
• SVN (Movelet): Support Vector Machine (SVM) with movelets features. The models were implemented using the Python language, with the keras, linear kernel and default structure. Other structure details are default settings. [REFERENCE*]

Installation

Install directly from PyPi repository, or, download from github. (python >= 3.7 required)

    pip install movelets

Citing

If you use automatize please cite the following paper:

Tarlis Tortelli Portela; Jonata Tyska Carvalho; Vania Bogorny. HiPerMovelets: high-performance movelet extraction for trajectory classification, International Journal of Geographical Information Science, 2022. DOI: 10.1080/13658816.2021.2018593.

Bibtex:

@article{Portela2022,
    author = {Tarlis Tortelli Portela and Jonata Tyska Carvalho and Vania Bogorny},
    title = {HiPerMovelets: high-performance movelet extraction for trajectory classification},
    journal = {International Journal of Geographical Information Science},
    volume = {0},
    number = {0},
    pages = {1-25},
    year  = {2022},
    publisher = {Taylor & Francis},
    doi = {10.1080/13658816.2021.2018593},
    URL = {https://doi.org/10.1080/13658816.2021.2018593}
}

Collaborate with us

Any contribution is welcome. This is an active project and if you would like to include your algorithm in movelets, feel free to fork the project, open an issue and contact us.

Feel free to contribute in any form, such as scientific publications referencing movelets, teaching material and workshop videos.

Related packages

• automatize: Automatize: Multiple Aspect Trajectory Data Mining Tool Library;

Change Log

This is a package under construction:

Dec. 2023:

TODO:

• Comments on all public interface funcions and modules