Munich Personal RePEc Archive

Black-Box Classification Techniques for Demographic Sequences : from Customised SVM to RNN

Muratova, Anna and Sushko, Pavel and Espy, Thomas H. (2017): Black-Box Classification Techniques for Demographic Sequences : from Customised SVM to RNN. Published in: CEUR Workshop Proceeding , Vol. 1968, No. Experimental Economics and Machine Learning (28 October 2017): pp. 31-40.

[img]
Preview
PDF
paper4.pdf

Download (369kB) | Preview

Abstract

Nowadays there is a large amount of demographic data which should be analysed and interpreted. From accumulated demographic data, more useful information can be extracted by applying modern methods of data mining. The aim of this study is to compare the methods of classification of demographic data by customising the SVM kernels using various similarity measures. Since demographers are interested in sequences without discontinuity, formulas for such sequences similarity measures were derived. Then they were used as kernels in the SVM method, which is the novelty of this study. Recurrent neural network algorithms, such as Simple RNN, GRU and LSTM, are also compared. The best classification result with SVM method is obtained using a special kernel function in SVM by transforming sequences into features, but recurrent neural network outperforms SVM.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.