An efficient stock market prediction model using hybrid feature reduction method based on variational autoencoders and recursive feature elimination

Financial Innovation

Table 10 Classification results using allstock_own features

Stocks	LSTM (basic)		LSTM (with attention)		SVM		LightGBM
Stocks	Acc	F-Mea	Acc	F-Mea	Acc	F-Mea	Acc	F-Mea
ALBRK	0.723	0.632	0.750	0.670	0.738	0.658	0.740	0.673
AKBNK	0.646	0.554	0.674	0.537	0.637	0.547	0.623	0.556
HALKB	0.624	0.568	0.662	0.543	0.637	0.588	0.627	0.593
ISCTR	0.664	0.543	0.686	0.550	0.680	0.573	0.666	0.572
SKBNK	0.737	0.634	0.743	0.641	0.741	0.639	0.744	0.652
TSKB	0.689	0.536	0.694	0.611	0.695	0.603	0.687	0.605
VAKBN	0.594	0.560	0.620	0.590	0.558	0.547	0.562	0.544
YKBNK	0.613	0.555	0.652	0.531	0.632	0.587	0.608	0.584
Average	0.661	0.573	0.685	0.574	0.669	0.594	0.657	0.597