Investigating the contribution of author- and publication-specific features to scholars’ h-index prediction

EPJ Data Science

Table 7 Comparison of XGBoost regression model performance to predict the feature h-index in one, five, and ten years (2009, 2014, and 2018) implemented on three datasets (junior, middle, and senior researchers). RMSE, MAPE, and sMAPE are the metrics to assess performance

Feature combination	Metric	Junior			Middle-level			Senior
Feature combination	Metric	2009	2014	2018	2009	2014	2018	2009	2014	2018
1	RMSE	0.62	3.01	5.15	0.68	2.85	4.94	0.75	3	5.09
	MAPE	0.24	0.52	0.62	0.16	0.33	0.45	0.09	0.2	0.28
	sMAPE	0.23	0.39	0.45	0.16	0.29	0.36	0.09	0.19	0.25
2	RMSE	0.61	2.91	4.99	0.67	2.78	4.81	0.75	2.94	4.97
	MAPE	0.24	0.49	0.59	0.16	0.32	0.43	0.09	0.2	0.28
	sMAPE	0.23	0.38	0.43	0.15	0.28	0.35	0.09	0.19	0.25
3	RMSE	0.61	2.85	4.91	0.68	2.75	4.77	0.75	2.9	4.9
	MAPE	0.24	0.5	0.6	0.16	0.33	0.44	0.09	0.2	0.28
	sMAPE	0.23	0.38	0.44	0.15	0.28	0.36	0.09	0.19	0.25
4	RMSE	0.6	2.78	4.81	0.67	2.68	4.67	0.74	2.85	4.8
	MAPE	0.24	0.48	0.57	0.16	0.32	0.43	0.09	0.2	0.27
	sMAPE	0.22	0.37	0.42	0.15	0.28	0.35	0.09	0.19	0.24
5	RMSE	0.67	3.23	5.46	0.72	3.05	5.23	0.78	3.24	5.49
	MAPE	0.28	0.57	0.68	0.17	0.37	0.49	0.09	0.23	0.31
	sMAPE	0.27	0.42	0.47	0.17	0.31	0.39	0.1	0.21	0.28
6	RMSE	1	3.27	5.43	1.87	3.56	5.5	4.04	5.75	7.52
	MAPE	0.37	0.56	0.65	0.41	0.44	0.53	0.41	0.4	0.44
	sMAPE	0.31	0.41	0.46	0.32	0.35	0.4	0.32	0.32	0.34
7	RMSE	0.97	3.19	5.3	1.8	3.48	5.38	3.79	5.47	7.24
	MAPE	0.36	0.54	0.62	0.39	0.43	0.51	0.38	0.38	0.42
	sMAPE	0.31	0.4	0.44	0.31	0.34	0.39	0.31	0.31	0.33
8	RMSE	0.96	2.97	5.02	1.75	3.33	5.23	3.64	5.23	6.93
	MAPE	0.35	0.53	0.62	0.38	0.41	0.5	0.35	0.36	0.4
	sMAPE	0.3	0.4	0.44	0.31	0.33	0.38	0.29	0.29	0.32
9	RMSE	0.94	2.92	4.93	1.69	3.28	5.15	3.47	5.05	6.74
	MAPE	0.34	0.51	0.6	0.36	0.41	0.49	0.34	0.34	0.39
	sMAPE	0.3	0.39	0.43	0.3	0.33	0.38	0.28	0.29	0.32