Detection of hidden patterns in time series data via multiple-time FOD method

Date

2019-06-23

Author

Erkuş, Ekin Can
Purutçuoğlu Gazi, Vilda

Metadata

Show full item record

Item Usage Stats

172
views

0
downloads

The periodicity in time series data can be detected by several frequency domains’ methods, especially, by the Fourier transform (FT). FT is a non-parametric method to convert the time domain data into the frequency domain and it is used in many engineering and data science applications. Recently this method which has been used to detect outliers in time series observations where the data may also include some systematic patterns is called “outlier detection via Fourier transform” (FOD). From our previous analyses via real and simulated time-course datasets, it has been shown that FOD is a promising technique to find periodic and non-periodic outliers in the data, particularly, when the sample size increases and the data are far from normal distribution. On the other hand, it has been observed that the multiple application of FOD is successful in order to detect the hidden patterns in real electrocardiogram (ECG) datasets since the pattern of FOD signals indicates differences between control and various types of heart diseases. Therefore, we consider that they can be applied for the pre-diagnosis of certain heart illnesses. In this study, we aim to extend the multiple time FOD by evaluating its performance comprehensively under distinct Monte Carlo scenarios such as different sample sizes, distributions and percentage of outliers. We consider that these analyses can be helpful to detect outliers and hidden patterns in distinct time-series data including ECG datasets.

Subject Keywords

Medical applications, Computational biology, Bioinformatics and medicine, Mathematical programming

URI

https://hdl.handle.net/11511/72007
https://www.euro-online.org/conferences/program/#abstract/2683

Conference Name

30th European Conference on Operational Research (EURO 2019) (23 - 26 Haziran 2019)

Collections

Graduate School of Natural and Applied Sciences, Conference / Seminar

Suggestions

OpenMETU
Core

Detection of abnormalities in heart rate using multiple Fourier transforms Erkuş, Ekin Can; Purutçuoğlu Gazi, Vilda (2019-09-01) Fourier transform (FT) is one of the transformation techniques to convert the time-domain signal into the frequency-domain signal. Due to its easy usage, it is applied in many engineering approaches where the data are made of periodic components. Electrocardiography (ECG) is an imaging modality which represents the data of electrophysiological activities of heart. ECG data are gathered from the electrodes that are placed on the specific locations on chest, and electrical activities of heart generally produc...
Using artificially generated spectral data to improve protein secondary structure prediction from Fourier transform infrared spectra of proteins Severcan, M; Haris, PI; Severcan, Feride (Elsevier BV, 2004-09-15) Secondary structures of proteins have been predicted using neural networks from their Fourier transform infrared spectra. To improve the generalization ability of the neural networks, the training data set has been artificially increased by linear interpolation. The leave-one-out approach has been used to demonstrate the applicability of the method. Bayesian regularization has been used to train the neural networks and the predictions have been further improved by the maximum-likelihood estimation method. T...
Feature extraction of hidden oscillation in ECG data via multiple-FOD method Purutçuoğlu Gazi, Vilda; Erkuş, Ekin Can (null; 2019-10-30) Fourier transform (FT) is a non-parametric method which can be used to convert the time domain data into the frequency domain and can be used to find the periodicity of oscillations in time series datasets. In order to detect periodic-like outliers in time series data, a novel and promising method, named as the outlier detection via Fourier transform (FOD), has been developed. From our previous studies, it has been shown that FOD outperforms most of the commonly used approaches for the detection of outliers...
Estimation of protein secondary structure from FTIR spectra using neural networks Severcan, M; Severcan, Feride; Haris, PI (Elsevier BV, 2001-05-30) Secondary structure of proteins have been predicted using neural networks (NN) from their Fourier transform infrared spectra. Leave-one-out approach has been used to demonstrate the applicability of the method. A form of cross-validation is used to train NN to prevent the overfitting problem. Multiple neural network outputs are averaged to reduce the variance of predictions. The networks realized have been tested and rms errors of 7.7% for alpha -helix, 6.4% for beta -sheet and 4.8% for turns have been achi...
ANALYSIS OF MILLIMETER WAVE-GUIDES ON ANISOTROPIC SUBSTRATES USING THE 3-DIMENSIONAL TRANSMISSION-LINE MATRIX-METHOD BULUTAY, C; PRASAD, S (1993-06-01) Three-dimensional condensed asymmetrical node, variable grid, transmission-line matrix (TLM) method has been used in analyzing several millimeter waveguides on anisotropic substrates. The dispersion characteristics of image guides together with field and energy confinement properties at millimeter-wave frequencies have been investigated. Edge coupled microstrip line on a uniaxial substrate is analyzed for the even and odd mode dispersion characteristics. Finally the same analysis is repeated for bilateral f...

Citation Formats

E. C. Erkuş and V. Purutçuoğlu Gazi, “Detection of hidden patterns in time series data via multiple-time FOD method,” presented at the 30th European Conference on Operational Research (EURO 2019) (23 - 26 Haziran 2019), Dublin, İrlanda, 2019, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/72007.