BMC Research Notes

Table 1 The MAPEs and RMSEs of the testing set for all methods applied on fs1: time series and fs2: \(53- weeks-before\_52-first-order-differences\) and fs3: \(n-years-before\_m-weeks-around\) feature space

From: A comparative study on predicting influenza outbreaks using different feature spaces: application of influenza-like illness data from Early Warning Alert and Response System in Syria

(a) The MAPEs%
Models	Average	Naïve	Seasonal naïve	Drift	STL	DHR	TBATS
Features:fs1	11.31	5.69	7.9	5.7	5.72	5.12	4.66
Models	GLM	SVR	GB	RF	3-Layers-LSTM	4-Layers-LSTM
Features:fs2	6.16	5.83	5.94	5.89	4.72	4.9
Models	GLM	SVR	GB	RF	3-Layers-LSTM	4-Layers-LSTM
Features:fs3
Y0W1	5.69	5.68	6.75	6.86	5.66	5.67
Y0W2	5.76	5.72	7.1	6.3	5.34	5.38
Y0W3	5.73	5.73	6.69	6.29	5.36	5.27
Y0W4	5.49	5.61	6.92	6.19	5.44	5.3
Y0W5	5.68	5.71	7.29	5.91	5.47	5.39
Y1W0	6.48	6.72	6.89	9.27	4.5	4.12
Y2W0	6.17	5.85	6.47	8.91	4.67	4.6
Y3W0	5.96	6.4	8.37	9.08	4.82	4.69
Y1W1	5.12	5.23	6.36	6.94	3.94	4.8
Y1W2	5.61	5	5.26	6.48	4.09	4.09
Y1W3	6.22	5.27	6.22	6.89	3.97	3.63
Y1W4	6.22	5.3	5.7	7.11	3.89	3.52
Y1W5	6.08	5.23	6.98	6.18	4.46	3.54
Y2W1	5.26	5.32	5.56	7.22	4.44	4.03
Y2W2	6.01	5.08	4.24	5.78	4.22	4.3
Y2W3	7.43	5.68	6.19	6.17	4.43	3.99
Y2W4	7.33	5.4	6.71	6.36	4.81	4.39
Y2W5	7.23	5.34	5.53	6.44	4.61	4.71
Y3W1	6.01	5.43	6.65	6.9	4.72	4.37
Y3W2	6.86	5.49	5.75	6.22	5.5	4.74
Y3W3	7.74	5.76	6.18	6.34	6.21	5.42
Y3W4	8.24	5.92	6.97	6.24	6.43	5.92
Y3W5	8.56	6.14	6.24	6.48	6.78	6.13

(b) The RMSEs
Models	Average	Naïve	Seasonal naïve	Drift	STL	DHR	TBATS
Features:fs1	0.05796	0.03925	0.04454	0.03931	0.03672	0.03179	0.03096
Models	GLM	SVR	GB	RF	3-Layers-LSTM	4-Layers-LSTM
Features:fs2	0.03743	0.03643	0.03687	0.03699	0.02294	0.0237
Models	GLM	SVR	GB	RF	3-Layers-LSTM	4-Layers-LSTM
Features:fs3
Y0W1	0.03852	0.03866	0.04389	0.04245	0.02744	0.02767
Y0W2	0.03933	0.03923	0.04525	0.03868	0.02598	0.02593
Y0W3	0.03754	0.03861	0.04286	0.03915	0.02631	0.02549
Y0W4	0.03689	0.03755	0.04355	0.0392	0.02666	0.02566
Y0W5	0.03686	0.03798	0.04528	0.03825	0.02611	0.02565
Y1W0	0.03781	0.03874	0.04163	0.04927	0.02113	0.01985
Y2W0	0.03685	0.03534	0.04096	0.04761	0.02252	0.02189
Y3W0	0.03629	0.03965	0.04526	0.05006	0.02312	0.02323
Y1W1	0.03341	0.03361	0.03851	0.0396	0.01929	0.02296
Y1W2	0.03654	0.03272	0.02968	0.03826	0.01938	0.0198
Y1W3	0.03742	0.03309	0.03434	0.03768	0.01873	0.01763
Y1W4	0.03745	0.03265	0.03369	0.03851	0.01825	0.01662
Y1W5	0.03652	0.03175	0.04083	0.0335	0.02139	0.01682
Y2W1	0.03472	0.034	0.03515	0.04029	0.02108	0.01984
Y2W2	0.03886	0.03352	0.0248	0.03327	0.01994	0.02276
Y2W3	0.04222	0.03381	0.03401	0.03705	0.02094	0.01926
Y2W4	0.04159	0.03249	0.03787	0.03909	0.02288	0.02122
Y2W5	0.04276	0.03291	0.03502	0.03929	0.02148	0.02276
Y3W1	0.0362	0.03305	0.03476	0.03988	0.02253	0.02121
Y3W2	0.04038	0.03156	0.03382	0.03729	0.02659	0.02319
Y3W3	0.04366	0.03357	0.03517	0.03822	0.0313	0.02588
Y3W4	0.04539	0.03405	0.04228	0.03822	0.03226	0.02919
Y3W5	0.05098	0.03592	0.03747	0.03928	0.03388	0.03064

Italics number indicate best result

Back to article page

ISSN: 1756-0500

Contact us

Submission enquiries: bmcresearchnotes@biomedcentral.com
General enquiries: ORSupport@springernature.com