Over ons

Consortium

Team

Vlaams Beleidsplan AI

Maak kennis met onze Alumni

Contact

<h5>Description:</h5>
<p>GENDIS extracts shapelets, i.e. a collection of subsequences, from a time series dataset that are very informative in classifying each of the time series into categories. GENDIS searches for this set of shapelets through evolutionary computation, a paradigm mostly unexplored within the domain of time series classification, which offers several benefits, namely (a) evolutionary algorithms are gradient-free, allowing for an easy configuration of the optimization objective, which does not need to be differentiable and allows escaping local optima more easily, (b) no brute-force search is required, making the algorithm scalable and allowing easy control over the run-time of the algorithm; (c) the total amount of shapelets and the length of each of these shapelets are evolved jointly with the shapelets themselves, alleviating the need to specify this beforehand; (iv) entire sets are evaluated at once as opposed to single shapelets, which results in smaller final sets with fewer similar shapelets that result in similar predictive performances; and (v) the discovered shapelets do not need to be a subsequence of the input time series. Moreover, the proposed technique has a computational complexity that is multiple orders of magnitude smaller than the current state-of-the-art while outperforming it in terms of predictive performance, with much smaller shapelet sets.</p>

<h5>More Information:</h5>
<ul>
<li><strong>Grand Challenge/Workpackage(s)</strong>: <a href="/nl/onderzoek/onderzoeksuitdagingen-2019-2023/ai-driven-data-science"[entry:58:/onderzoek/onderzoeksutidagingen-2019-2023/ai-driven-data-science]data-entry-id="58">GC1</a> - WP7-T7.3</li>
<li><strong>Responsible Research Lead</strong>: <a href="mailto:Femke.Ongenae@ugent.be%20" target="_blank" rel="noopener">Femke Ongenae </a>&nbsp;</li>
<li><strong>Authors</strong>: <a href="mailto:gilles.vandewiele@ugent.be" target="_blank" rel="noopener">Gilles Vandewiele</a>&nbsp;</li>
<li><strong>Programming Language</strong>: Python</li>
<li><a href="https://github.com/IBCNServices/GENDIS" target="_blank" rel="noopener">Link to the repository</a></li>
<li><strong>Categories</strong>: Time Series, Supervised learning</li>
</ul>

<h5>Publications:</h5>
<ul>
<li>G. Vandewiele, F. Ongenae, F. De Turck, "<a href="https://www.mdpi.com/1424-8220/21/4/1059" target="_blank" rel="noopener">GENDIS: GENetic Discovery of Shapelets</a>", Sensors, Vol 21, No. 4, 2021,&nbsp;</li>
</ul>
<p>&nbsp;</p>

<p>Library for shapelet extraction from time series using a genetic algorithm.</p>