Study on Contribution of Different Journal Evaluation Indicators to Impact Factor Based on Machine Learning

Table 10

The loss of each indicator in input perturbation method based on deeper neural network.

Journal evaluation indicator	MSE	MAE
Origin	0.2572	0.2631
Number of regions	0.43305	0.534394
Ratio of funded paper	0.43330	0.534522
Average number of authors	0.43288	0.534569
Average number of citations	0.43093	0.533971
Citing half-life	0.43286	0.534738
H-index	0.42829	0.533240
Other-cite ratio	0.43302	0.534505
Immediacy index	0.43288	0.534471
Disciplinary impact index	0.43304	0.534521
Disciplinary diffusion index	0.43256	0.534354
Citation num	0.42343	0.534203
Total citations	0.43234	0.553242
Cited half-life	0.43291	0.534099
Literature selection rate	0.43302	0.534546
Number of institutions	0.43838	0.541738
Number of papers	0.43440	0.555592
Ratio of international paper	0.43311	0.534530

Origin means that no indicator is perturbed.