Fast weighted horizontal visibility algorithms can construct a complex network from a time series in linear time.

• Reference

Guohun Zhu, Yan Li, Peng (Paul) Wen, Epileptic seizure detection in EEGs signals using a fast weighted horizontal visibility algorithm, Computer Methods and Programs in Biomedicine, Available online 15 April 2014, ISSN 0169-2607, http://dx.doi.org/10.1016/j.cmpb.2014.04.001. (http://www.sciencedirect.com/science/article/pii/S0169260714001266)

Zhu, G., Li, Y., Wen, P.P. and Wang, S., 2014. Analysis of alcoholic EEG signals based on horizontal visibility graph entropy. Brain informatics, 1(1-4), pp.19-25.

• Robustness agaist noise
• Constructing a graph in linear time
• Extract the mean degree, mean strengths in linear time
• Applied in classifying the epileptic EEG siganls

If you want to test on long time series, please install the R package as fellows.

• Download the R software for windows, please access R software
• Download the R package FWHVA 1.1.0, please access here FWHVA R package
• Download the R package FWHVA 1.2.0, please access here FWHVA R package
• Downloand the eiplepsy EEG sets for this paper, please access small epilepsy database

• Installation and Running -------------------------
• 1. Prepare the files

• (1)Install R and the package FWHVA, e1071 of R
• (2)install the R package FWHVA

• There exists a possible problem during install the FWHVA The R software is not in 3.0.2 version package FWHVA was built before R 3.0.0: please re-install it
• (3)Decompress epilepsy EEG data .
• The five EEG sets should be decompress into a same directory.

• (4)decompress all R source code

• (5)Change the InputPath defined in all r files for example, if your path of the EEG data is c:\ then InputPath="c:/"

• 2. Stage 1 : Feature extraction Execute the features by run getTableI_figures 3 and 4.r

• 3. Conducts the classfication with R Execute the R and run the code in the R files. For example, to get the HVG and SampEN (one dimension) classfication (TableI in the paper), execute the R source code get_Figure5.r

• 4. If you meet some problems on downloading or runing, please send an email to me g.zhu@uq.edu.au or guohun.zhu.phd@ieee.org with your questions, I will give your feedback as soon as possible.