Identifying Key Factors in Human-Computer Interaction Through Feature-Based Analysis and Sensitivity Evaluation

Distribution	Variables	Indicators
Min	Max	Average	Standard deviation
Normal	Session length	–2.41	69.26	30.06	9.96
	Clicks per minute	–2.84	12.06	4.98	2.02
	Scroll depth percent	7.49	100.00	74.18	18.14
	Mouse distance Px	0.70	3239.54	1017.62	482.40
Interface complexity	1.00	10.00	5.02	1.95
Task completion time	0.05	640.97	300.42	98.93
Age	18.00	64.00	41.32	13.51
		Lower bound		Upper bound
Uniform	Keyboard activity level	0		2
	Errors made	0		10
	Help requested	0		1
	Device type	0		2
Experience level	0		2
Gender	0		2
Satisfaction score	1		5

Table 2.

Obtained results of the fine-tuning values.

Hyperparameters	Model
Hyperparameters	DTNO	DTFO	RFNO	RFFO
N_estimators	–	–	876	378
Max_depth	33	48	145	55
Min_samples_split	7	32	24	31
Min_samples_leaf	5	21	12	15
Max_leaf_nodes	11	3	–	–

Table 3.

Performance of the developed models in the training and testing phases.

	Model	Index values
Accuracy	Precision	Recall	F1-score
Train	DTC	0.952	0.952	0.952	0.952
	DTNO	0.969	0.969	0.969	0.969
	DTFO	0.982	0.982	0.982	0.982
RFC	0.960	0.960	0.960	0.960
RFNO	0.980	0.980	0.980	0.979
RFFO	0.992	0.992	0.992	0.992
Test	DTC	0.933	0.934	0.933	0.933
	DTNO	0.936	0.936	0.936	0.936
	DTFO	0.974	0.974	0.974	0.974
RFC	0.937	0.937	0.937	0.937
RFNO	0.940	0.940	0.940	0.940
RFFO	0.984	0.984	0.984	0.984

Table 4.

Comparison between the models in five classes.

	DT models			RF models
	DTC			RFC
Class	Precision	Recall	F1-score	Precision	Recall	F1-score
1	0.95	0.95	0.95	0.96	0.95	0.96
2	0.95	0.95	0.95	0.96	0.96	0.96
3	0.95	0.95	0.95	0.95	0.96	0.95
4	0.96	0.94	0.95	0.96	0.96	0.96
5	0.94	0.95	0.95	0.95	0.96	0.95
	DTNO			RFNO
	Precision	Recall	F1-score	Precision	Recall	F1-score
1	0.97	0.96	0.96	0.97	0.97	0.97
2	0.97	0.96	0.97	0.97	0.97	0.97
3	0.95	0.96	0.95	0.98	0.98	0.98
4	0.96	0.96	0.96	0.97	0.96	0.97
5	0.96	0.97	0.97	0.98	0.97	0.98
	DTFO			RFFO
	Precision	Recall	F1-score	Precision	Recall	F1-score
1	0.98	0.98	0.98	0.99	0.99	0.99
2	0.98	0.98	0.98	0.99	0.99	0.99
3	0.98	0.98	0.98	0.99	1.00	0.99
4	0.98	0.98	0.98	0.99	0.99	0.99
5	0.98	0.98	0.98	0.99	0.99	0.99

Confusion matrix for the accuracy of the developed models.

Comparison of TPR versus FPR for the studied models.

Sensitivity analysis of input variables using the CAM, showing that mouse distance, clicks per minute, and interface complexity are the three most influential factors affecting model performance.

References

Authors

Metrics

Articles in this issue

DOI: https://doi.org/10.2478/msr-2026-0029 | Journal eISSN: 1335-8871

Journal RSS Feed

Language: English

Page range: 231 - 244

Submitted on: Feb 7, 2026

Accepted on: Jun 30, 2026

Published on: Aug 19, 2026

Published by: Slovak Academy of Sciences, Institute of Measurement Science

In partnership with: Paradigm Publishing Services

Publication frequency: Volume open

Keywords:

Human-Computer Interaction,

machine learning,

Random Forest Classification,

Decision Tree Classification,

Fox Optimizer,

Northern Goshawk Optimization

Related subjects:

Engineering,

Electrical engineering,

Control engineering, metrology and testing

© 2026 Liyun Liu, Hui Shi, Yi Wang, published by Slovak Academy of Sciences, Institute of Measurement Science
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Volume 26 (2026): Issue 5 (October 2026)

Identifying Key Factors in Human-Computer Interaction Through Feature-Based Analysis and Sensitivity Evaluation

Figures & Tables

Table 1.

Fig. 1.

Table 2.

Table 3.

Table 4.

Fig. 2.

Fig. 3.

Fig. 4.

Paradigm

My account