Institute for Web Science and Technologies · Universität Koblenz - Landau
Institute WeST

Sigmentation: Univariate Data Segmentation using Genetic Algorithm

[go to overview]
Zeyd Boukhers

Segmenting univariate data is an important task in several domains. This task is more challenging when the data is acquired from sensors due to the lack of clues to find different patterns in sequential data. In this talk, I will present a segmentation approach that uses Fourier transform and genetic algorithm. Fourier transform is powerful to understand the frequencies in the data. Therefore, it is used in this approach to computing the overall optimization cost. Genetic algorithm is used to find the segments whose amount is not initially defined. Therefore, this task is a multivariable optimization task.

06.02.20 - 10:15
B 016