Multientropy Approaches: Combining Different Entropy Measures to Exploit Possible Synergies in Time Series Classification

Dear Colleagues,

A myriad entropy measures have been proposed in the last years. Thousands of studies have used them successfully for time series classification. However, there are still applications where classification performance is still poor or not significant in comparison with the results obtained employing other feature extraction techniques.

In this regard, many improvements to the current methods have also been suggested: more robustness to noise, better sensitivity to time series differences, less dependence on input parameters or number of data, multiscale approaches, etc.

A line of research that has not been exploited so profusely yet is using more than a single entropy measure simultaneously. The rationale of this approach is the fact that the current state of the art of entropy measures includes algorithms based on histograms in terms of amplitude patterns, ordinal patterns, symbolic patterns, and even sorting patterns. Therefore, it can be hypothesized that different patterns look at different aspects of the time series dynamics, and a combination of them, as parameters of a single model, or by means of an algorithmic integration, can provide a more complete picture of the underlying structure of the time series.

In this Special Issue, we would like to collect papers focusing on the simultaneous application of more than one entropy measure for time series classification, not just for performance comparative purposes, but to exploit possible synergies and improve the classification performance using a single measure in an isolated manner. Any kind of entropy measure and of time series can be considered.

The main topics of this Special Issue include (but are not limited to) theoretical developments and practical applications with regard to:

• Time series classification using statistical models or equations including more than one entropy measure;
• Multimeasure algorithm integration. New entropy methods based on the evolution of current methods by merging different approaches in a single algorithm. 

Dr. David Cuesta-Frau
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Entropy is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• entropy
• amplitude patterns
• ordinal patterns
• symbolic patterns
• sorting patterns
• classification
• clustering
• machine learning
• pattern recognition