MetaHeuristic Optimization and Keystroke Dynamics for Authentication of Smartphone Users
Abstract
:1. Introduction
2. Literature Review
3. Keystroke Dynamics
3.1. Touch Information
3.1.1. Dwell Time
3.1.2. Flight Time
3.1.3. Pressure
3.1.4. Coordinates
3.1.5. Motion Data
3.1.6. Accelerometer
3.1.7. Angular Velocity
3.1.8. Rotation Vector
4. The Proposed Methodology
4.1. Bidirectional Recurrent Neural Network (BRNN)
4.2. Dipper Throated Optimization (DTO)
Algorithm 1 The Dipper Throated Optimization algorithm 

4.3. The Proposed Dynamic Weighted DTO Algorithm
Algorithm 2 The Proposed DWDTO Algorithm 

4.3.1. Exploration Group
4.3.2. Exploitation Group
4.3.3. Balance between Exploration and Exploitation
4.3.4. Binary Optimizer
Algorithm 3 The proposed feature selection algorithm (binary bDWDTO) 

5. Experimental Results
5.1. Evaluation Criteria
5.2. Results of the First Scenario
5.3. Results of the Second Scenario
5.4. Classification Results
5.5. Comparison with Other Methods
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
 Abualigah, L.; Elaziz, M.A.; Khodadadi, N.; Forestiero, A.; Jia, H.; Gandomi, A.H. Aquila Optimizer Based PSO Swarm Intelligence for IoT Task Scheduling Application in Cloud Computing. In Integrating MetaHeuristics and Machine Learning for RealWorld Optimization Problems; Springer: Berlin/Heidelberg, Germany, 2022; pp. 481–497. [Google Scholar]
 Sharma, R.; Sharma, V.K.; Singh, A. A Review Paper on Facial Recognition Techniques. In Proceedings of the 2021 Fifth International Conference on ISMAC (IoT in Social, Mobile, Analytics and Cloud) (ISMAC), Palladam, India, 11–13 November 2021; pp. 617–621. [Google Scholar]
 Ali, M.M.; Mahale, V.H.; Yannawar, P.; Gaikwad, A.T. Overview of fingerprint recognition system. In Proceedings of the 2016 International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT), Chennai, India, 3–5 March 2016; pp. 1334–1338. [Google Scholar]
 Daniel, D.M.; Monica, B. Person authentication technique using human iris recognition. In Proceedings of the 2010 9th International Symposium on Electronics and Telecommunications, Timisoara, Romania, 11–12 November 2010; pp. 265–268. [Google Scholar]
 Kaveh, A.; Eslamlou, A.D.; Khodadadi, N. Dynamic water strider algorithm for optimal design of skeletal structures. Period. Polytech. Civ. Eng. 2020, 64, 904–916. [Google Scholar] [CrossRef]
 Kaveh, A.; Talatahari, S.; Khodadadi, N. Stochastic paint optimizer: Theory and application in civil engineering. Eng. Comput. 2022, 38, 1921–1952. [Google Scholar] [CrossRef]
 Kaveh, A.; Khodadadi, N.; Talatahari, S. A comparative study for the optimal design of steel structures using CSS and ACSS algorithms. Int. J. Optim. Civ. Eng 2021, 11, 31–54. [Google Scholar]
 Khodadadi, N.; Abualigah, L.; Mirjalili, S. Multiobjective Stochastic Paint Optimizer (MOSPO). Neural Comput. Appl. 2022, 2022, 1–24. [Google Scholar] [CrossRef]
 Khodadadi, N.; Mirjalili, S. Truss optimization with natural frequency constraints using generalized normal distribution optimization. Appl. Intell. 2022, 52, 10384–10397. [Google Scholar] [CrossRef]
 Kaveh, A.; Talatahari, S.; Khodadadi, N. The Hybrid Invasive Weed OptimizationShuffled Frogleaping Algorithm Applied to Optimal Design of Frame Structures. Period. Polytech. Civ. Eng. 2019, 63, 882–897. [Google Scholar] [CrossRef]
 Ryu, Y.S.; Koh, D.H.; Aday, B.L.; Gutierrez, X.A.; Platt, J.D. Usability Evaluation of Randomized Keypad. J. Usability Study 2010, 5, 65–75. [Google Scholar]
 Spillane, R. Keyboard apparatus for personal identification. IBM Tech. Discl. Bull. 1975, 17, 3346. [Google Scholar]
 Umphress, D.; Williams, G. Identity verification through keyboard characteristics. Int. J. ManMach. Stud. 1985, 23, 263–273. [Google Scholar] [CrossRef]
 Campisi, P.; Maiorana, E.; Lo Bosco, M.; Neri, A. User authentication using keystroke dynamics for cellular phones. IET Signal Process. 2009, 3, 333. [Google Scholar] [CrossRef]
 Lee, H.; Hwang, J.Y.; Kim, D.I.; Lee, S.; Lee, S.H.; Shin, J.S. Understanding Keystroke Dynamics for Smartphone Users Authentication and Keystroke Dynamics on Smartphones BuiltIn Motion Sensors. Secur. Commun. Netw. 2018, 2018, 2567463. [Google Scholar] [CrossRef]
 Cockell, R.; Halak, B. On the Design and Analysis of a Biometric Authentication System Using Keystroke Dynamics. Cryptography 2020, 4, 12. [Google Scholar] [CrossRef]
 Alsultan, A.; Warwick, K.; Wei, H. Nonconventional keystroke dynamics for user authentication. Pattern Recognit. Lett. 2017, 89, 53–59. [Google Scholar] [CrossRef]
 Kim, J.; Kang, P. Freely typed keystroke dynamicsbased user authentication for mobile devices based on heterogeneous features. Pattern Recognit. 2020, 108, 107556. [Google Scholar] [CrossRef]
 Kiyani, A.T.; Lasebae, A.; Ali, K.; Rehman, M.U.; Haq, B. Continuous User Authentication Featuring Keystroke Dynamics Based on Robust Recurrent Confidence Model and Ensemble Learning Approach. IEEE Access 2020, 8, 156177–156189. [Google Scholar] [CrossRef]
 Porwik, P.; Doroz, R.; Wesolowski, T.E. Dynamic keystroke pattern analysis and classifiers with competence for user recognition. Appl. Soft Comput. 2021, 99, 106902. [Google Scholar] [CrossRef]
 Saini, B.S.; Singh, P.; Nayyar, A.; Kaur, N.; Bhatia, K.S.; ElSappagh, S.; Hu, J.W. A ThreeStep Authentication Model for Mobile Phone User Using Keystroke Dynamics. IEEE Access 2020, 8, 125909–125922. [Google Scholar] [CrossRef]
 Jalaly Bidgoly, A.; Jalaly Bidgoly, H.; Arezoumand, Z. A survey on methods and challenges in EEG based authentication. Comput. Secur. 2020, 93, 101788. [Google Scholar] [CrossRef]
 Ingale, M.; Cordeiro, R.; Thentu, S.; Park, Y.; Karimian, N. ECG Biometric Authentication: A Comparative Analysis. IEEE Access 2020, 8, 117853–117866. [Google Scholar] [CrossRef]
 Maiti, A.; Crager, K.; Jadliwala, M.; He, J.; Kwiat, K.; Kamhoua, C. RandomPad: Usability of Randomized Mobile Keypads for Defeating Inference Attacks. In Proceedings of the IEEE Euro Workshop on Innovations in Mobile Privacy & Security (IMPS), Paris, France, 29 April 2017; pp. 1–10. [Google Scholar]
 Benjapatanamongkol, N.; Bhattarakosol, P. A Preliminary Study of Finger Area and Keystroke Dynamics Using Numeric Keypad With Random Numbers on Android Phones. In Proceedings of the 2019 23rd International Computer Science and Engineering Conference (ICSEC), Phuket, Thailand, 30 October–1 November 2019; pp. 30–34. [Google Scholar]
 Yu, E.; Cho, S. GASVM wrapper approach for feature subset selection in keystroke dynamics identity verification. In Proceedings of the International Joint Conference on Neural Networks, 2003, Portland, OR, USA, 20–24 July 2003; Volume 3, pp. 2253–2257. [Google Scholar]
 Azevedo, G.L.F.B.G.; Cavalcanti, G.D.C.; Carvalho Filho, E.C.B. An approach to feature selection for keystroke dynamics systems based on PSO and feature weighting. In Proceedings of the 2007 IEEE Congress on Evolutionary Computation, Singapore, 25–28 September 2007; pp. 3577–3584. [Google Scholar]
 Karnan, M.; Muthuramalingam, A.; Kalamani, A. Feature subset selection in keystroke dynamics using ant colony optimization. J. Eng. Technol. Res. 2009, 1, 72–80. [Google Scholar]
 Karnan, M.; Akila, M. Personal Authentication Based on Keystroke Dynamics Using Soft Computing Techniques. In Proceedings of the 2010 Second International Conference on Communication Software and Networks, Singapore, 26–28 February 2010; pp. 334–338. [Google Scholar]
 Solami, E.A.; Boyd, C.; Clark, A.; Ahmed, I. Userrepresentative feature selection for keystroke dynamics. In Proceedings of the 2011 5th International Conference on Network and System Security, Milan, Italy, 6–8 September 2011; pp. 229–233. [Google Scholar]
 ElKenawy, E.S.M.; Mirjalili, S.; Alassery, F.; Zhang, Y.D.; Eid, M.M.; ElMashad, S.Y.; Aloyaydi, B.A.; Ibrahim, A.; Abdelhamid, A.A. Novel MetaHeuristic Algorithm for Feature Selection, Unconstrained Functions and Engineering Problems. IEEE Access 2022, 10, 40536–40555. [Google Scholar] [CrossRef]
 Abdelhamid, A.A.; ElKenawy, E.S.M.; Alotaibi, B.; Amer, G.M.; Abdelkader, M.Y.; Ibrahim, A.; Eid, M.M. Robust Speech Emotion Recognition Using CNN+LSTM Based on Stochastic Fractal Search Optimization Algorithm. IEEE Access 2022, 10, 49265–49284. [Google Scholar] [CrossRef]
 Sami Khafaga, D.; Ali Alhussan, A.; Elkenawy, E.S.M.; Takieldeen, A.E.; Hassan, T.M.; Hegazy, E.A.; Eid, E.A.F.; Ibrahim, A.; Abdelhamid, A.A. Metaheuristics for Feature Selection and Classification in Diagnostic BreastCancer. Comput. Mater. Contin. 2022, 73, 749–765. [Google Scholar]
 Choi, M.; Lee, S.; Jo, M.; Shin, J.S. Keystroke dynamicsbased authentication using unique keypad. Sensors 2021, 21, 2242. [Google Scholar] [CrossRef] [PubMed]
 Sami Khafaga, D.; Ali Alhussan, A.; Elkenawy, E.S.M.; Ibrahim, A.; Abd Elkhalik, S.H.; ElMashad, S.Y.; Abdelhamid, A.A. Improved Prediction of Metamaterial Antenna Bandwidth Using Adaptive Optimization of LSTM. Comput. Mater. Contin. 2022, 73, 865–881. [Google Scholar] [CrossRef]
 Abdel Samee, N.; ElKenawy, E.S.M.; Atteia, G.; Jamjoom, M.M.; Ibrahim, A.; Abdelhamid, A.A.; ElAttar, N.E.; Gaber, T.; Slowik, A.; Shams, M.Y. Metaheuristic Optimization Through Deep Learning Classification of COVID19 in Chest XRay Images. Comput. Mater. Contin. 2022, 73, 4193–4210. [Google Scholar]
 Nasser AlEisa, H.; Elkenawy, E.S.M.; Ali Alhussan, A.; Saber, M.; Abdelhamid, A.A.; Sami Khafaga, D. Transfer Learning for Chest Xrays Diagnosis Using Dipper Throated Algorithm. Comput. Mater. Contin. 2022, 73, 2371–2387. [Google Scholar]
 MEUMobile KSD Data Set. Available online: https://archive.ics.uci.edu/ml/datasets/MEUMobile+KSD (accessed on 28 June 2022).
 RHU KeyStroke Dynamics Benchmark Dataset. Available online: https://www.coolestech.com/rhukeystroke/ (accessed on 28 June 2022).
Ref.  Methodology  Result 

[12]  Keystroke dynamics  False FRR = 12%, FAR = 6% 
[15]  Keystroke dynamics  EER = 15% 
[16]  Motion sensors  EER = 8.94% 
[25]  Random keypad  EER = 10% 
[34]  Unique Keypad  EER = 4.15% 
Metric  Equation 

Average fitness  $\frac{1}{M}{\sum}_{i=1}^{M}{g}_{*}^{i}$ 
Worst Fitness  $ma{x}_{i=1}^{M}{g}_{*}^{i}$ 
Best fitness  $mi{n}_{i=1}^{M}{g}_{*}^{i}$ 
Average Error  $\frac{1}{M}{\sum}_{j=1}^{M}\frac{1}{N}{\sum}_{i=1}^{N}mse({C}_{i},{L}_{i})$ 
Average select size  $\frac{1}{M}{\sum}_{i=1}^{M}size\left({g}_{*}^{i}\right)$ 
Standard deviation  $\sqrt{\frac{1}{M1}{\sum}_{i=1}^{M}{\left(\right)}^{{g}_{*}^{i}}2}$ 
Accuracy  $\frac{TP+TN}{TP+TN+FP+FN}$ 
Nvalue (NPV)  $\frac{TN}{TN+FN}$ 
pvalue (PPV)  $\frac{TP}{TP+FP}$ 
Sensitivity (TPR)  $\frac{TP}{TP+FN}$ 
Specificity (TNR)  $\frac{TN}{TN+FP}$ 
F1Score  $\frac{TP}{TP+0.5(FP+FN)}$ 
Algorithm  Avg. Error  Avg. Select Size  Avg. Fitness  Best Fitness  Worst Fitness  Std Fitness 

bDWDTO  0.510  0.654  0.537  0.442  0.637  0.345 
bGWO  0.523  0.718  0.573  0.462  0.656  0.364 
bGWO_PSO  0.516  0.713  0.556  0.539  0.617  0.348 
bPSO  0.514  0.848  0.560  0.462  0.675  0.372 
bSFS  0.522  0.672  0.574  0.523  0.597  0.364 
bWAO  0.511  0.943  0.561  0.500  0.675  0.359 
bMGWO  0.520  0.764  0.539  0.490  0.656  0.355 
bMVO  0.511  0.818  0.561  0.520  0.636  0.352 
bSBO  0.528  0.833  0.568  0.520  0.636  0.360 
bGWO_GA  0.532  0.793  0.532  0.520  0.636  0.357 
bFA  0.517  0.853  0.567  0.500  0.695  0.363 
bGA  0.511  0.813  0.561  0.462  0.636  0.363 
Algorithm  D1  D2 

bDWDTO  12.534  12.952 
bGWO  13.178  14.883 
bGWO_PSO  12.77  14.02 
bPSO  12.86  14.455 
bSFS  14.26  14.21 
bWAO  12.667  13.788 
bMGWO  12.95  13.49 
bMVO  13.121  14.395 
bSBO  13.59  14.42 
bGWO_GA  13.31  14.69 
bFA  13.888  14.472 
bGA  13.134  14.408 
Metric  NN  KNN  BRNN 

Accuracy  0.917  0.922  0.939 
Sensitivity (TPR)  0.862  0.870  0.901 
Specificity (TNR)  0.980  0.980  0.980 
Pvalue (PPV)  0.980  0.980  0.980 
Nvalue (NPV)  0.862  0.870  0.901 
Fscore  0.917  0.922  0.939 
Time (seconds)  137  125  102 
Metric  DWDTO + BRNN  GWO  WOA  PSO  GA  GSA 

Num. Values  20  20  20  20  20  20 
Minimum  0.9889  0.9612  0.9378  0.9598  0.9523  0.9563 
25%  0.9900  0.9712  0.9578  0.9685  0.9623  0.9563 
Median  0.9900  0.9712  0.9578  0.9685  0.9623  0.9563 
75%  0.9900  0.9712  0.9578  0.9685  0.9623  0.9563 
Maximum  0.9927  0.9812  0.9698  0.9798  0.9723  0.9763 
Range  0.0038  0.0200  0.0320  0.0200  0.0200  0.0200 
Mean  0.9901  0.9712  0.9574  0.9686  0.9623  0.9582 
Std.  0.0007  0.0032  0.0053  0.0033  0.0032  0.0050 
Std. Error  0.0001  0.0007  0.0012  0.0007  0.0007  0.0011 
Skewness  3.289  5.703 × 10${}^{14}$  −2.171  1.263  0  3.014 
Kurtosis  14.79  9.5  11.74  10.25  9.5  9.335 
Sum  19.8  19.42  19.15  19.37  19.25  19.16 
Metric  DWDTO + BRNN  GWO  WOA  PSO  GA  GSA 

Theo. median  0  0  0  0  0  0 
Act. median  0.99  0.9712  0.9578  0.9685  0.9623  0.9563 
Num. Values  20  20  20  20  20  20 
Sum ranks  210  210  210  210  210  210 
Sum +ranks  210  210  210  210  210  210 
Sum −ranks  0  0  0  0  0  0 
p value  <0.0001  <0.0001  <0.0001  <0.0001  <0.0001  <0.0001 
Significance  Yes  Yes  Yes  Yes  Yes  Yes 
Discrepancy  0.99  0.9712  0.9578  0.9685  0.9623  0.9563 
SS  DF  MS  F (DFn, DFd)  p Value  

Treatment  0.01246  4  0.003115  F (4, 95) = 256.8  p < 0.0001 
Residual  0.001152  95  0.00001213  
Total  0.01361  99 
Algorithm  Avg. Error  Avg. Select Size  Avg. Fitness  Best Fitness  Worst Fitness  Std Fitness 

bDWDTO  0.447  0.449  0.459  0.407  0.558  0.345 
bGWO  0.460  0.579  0.494  0.424  0.568  0.355 
bGWO_PSO  0.461  0.603  0.461  0.449  0.517  0.351 
bPSO  0.488  0.803  0.521  0.458  0.576  0.346 
bSFS  0.467  0.627  0.467  0.414  0.600  0.387 
bWAO  0.473  0.644  0.507  0.433  0.593  0.355 
bMGWO  0.449  0.567  0.491  0.457  0.576  0.350 
bMVO  0.482  0.784  0.515  0.416  0.559  0.352 
bSBO  0.468  0.743  0.468  0.441  0.543  0.350 
bGWO_GA  0.507  0.737  0.507  0.492  0.602  0.359 
bFA  0.478  0.800  0.512  0.407  0.610  0.360 
bGA  0.468  0.703  0.502  0.441  0.619  0.360 
Metric  NN  KNN  BRNN 

Accuracy  0.932  0.941  0.955 
Sensitivity (TPR)  0.857  0.895  0.895 
Specificity (TNR)  0.989  0.989  0.993 
pvalue (PPV)  0.984  0.988  0.988 
Nvalue (NPV)  0.900  0.900  0.938 
Fscore  0.916  0.939  0.939 
Time (seconds)  118  107  97 
Metric  DWDTO + BRNN  GWO  WOA  PSO  GA  GSA 

Num. Values  20  20  20  20  20  20 
Minimum  0.9899  0.9689  0.9465  0.9599  0.9700  0.9471 
25%  0.9934  0.9789  0.9665  0.9689  0.9800  0.9571 
Median  0.9934  0.9789  0.9665  0.9689  0.9800  0.9571 
75%  0.9934  0.9789  0.9680  0.9689  0.9800  0.9571 
Maximum  0.9934  0.9889  0.9767  0.9729  0.9900  0.9771 
Range  0.0034  0.0200  0.0301  0.0130  0.0200  0.0300 
Mean  0.9932  0.9788  0.9673  0.9686  0.9800  0.9581 
Std.  0.0008  0.0033  0.0061  0.0023  0.0032  0.0055 
Std. Error  0.0002  0.0007  0.0014  0.0005  0.0007  0.0012 
Skewness  −4.472  0.0496  −1.694  −2.964  −5.703 × 10${}^{14}$  2.164 
Kurtosis  20  9.379  7.402  13.36  9.5  8.21 
Sum  19.86  19.58  19.35  19.37  19.6  19.16 
Metric  DWDTO + BRNN  GWO  WOA  PSO  GA  GSA 

Theo. median  0  0  0  0  0  0 
Act. median  0.9934  0.9789  0.9665  0.9689  0.98  0.96 
Num. Values  20  20  20  20  20  20 
Sum ranks  210  210  210  210  210  210 
Sum +ranks  210  210  210  210  210  210 
Sum −ranks  0  0  0  0  0  0 
pvalue  <0.0001  <0.0001  <0.0001  <0.0001  <0.0001  <0.0001 
Significance  Yes  Yes  Yes  Yes  Yes  Yes 
Discrepancy  0.9934  0.9789  0.9665  0.9689  0.98  0.96 
SS  DF  MS  F (DFn, DFd)  p Value  

Treatment  0.008733  4  0.002183  F (4, 95) = 170.9  p < 0.0001 
Residual  0.001214  95  0.00001278  
Total  0.009946  99 
Metric  D1  D2 

Accuracy  0.990182803  0.993208829 
Sensitivity (TRP)  0.946547884  0.965909091 
Specificity (TNP)  0.998003992  0.998003992 
pvalue (PPV)  0.988372093  0.988372093 
Nvalue (NPV)  0.990491284  0.994035785 
FScore  0.967007964  0.977011494 
Time (seconds)  77  59 
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. 
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
ElKenawy, E.S.M.; Mirjalili, S.; Abdelhamid, A.A.; Ibrahim, A.; Khodadadi, N.; Eid, M.M. MetaHeuristic Optimization and Keystroke Dynamics for Authentication of Smartphone Users. Mathematics 2022, 10, 2912. https://doi.org/10.3390/math10162912
ElKenawy ESM, Mirjalili S, Abdelhamid AA, Ibrahim A, Khodadadi N, Eid MM. MetaHeuristic Optimization and Keystroke Dynamics for Authentication of Smartphone Users. Mathematics. 2022; 10(16):2912. https://doi.org/10.3390/math10162912
Chicago/Turabian StyleElKenawy, ElSayed M., Seyedali Mirjalili, Abdelaziz A. Abdelhamid, Abdelhameed Ibrahim, Nima Khodadadi, and Marwa M. Eid. 2022. "MetaHeuristic Optimization and Keystroke Dynamics for Authentication of Smartphone Users" Mathematics 10, no. 16: 2912. https://doi.org/10.3390/math10162912