Classification Algorithms in Adaptive Systems for Neuro-Ergonomic Applications
Adaptive systems typically comprise components that interrelate and interact to enable the whole system to respond and adjust to changes in the environment, operator, and task in order to regulate or maintain a level of performance or homeostasis. In so doing, they enable a degree of individualization and customization for many technological innovations such as managing the use of automation. Adaptive systems often involve some kind of feedback or closed-loop which requires a criteria for determining invoking thresholds, as well as some type of classification algorithm that models the type of changes to which the system has to adapt. This paper outlines the issues, considerations, and challenges associated with classification in adaptive systems, and reviews several algorithms that implement the feedback loop in neuro-ergonomic applications. These include logistic regression, Naïve-Bayes, artificial neural networks (ANN), and support vector machines (SVM) techniques.
KeywordsClassification Adaptive systems Neuro-ergonomics
This research was sponsored by the Army Research Laboratory and was accomplished under Cooperative Agreement Number W911 NF-14-2-0021. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied of the Army Research Laboratory of or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.
- 2.Hannula, M., Huttunen, K., Koskelo, J., Laitinen, T., Leino, T.: Comparison between artificial neural network and multilinear regression models in an evaluation of cognitive workload in a flight simulator. Comput. Biol. Med. 38, 1163–1170 (2008). https://doi.org/10.1016/j.compbiomed.2008.09.007CrossRefGoogle Scholar
- 10.Hockey, G.R.J.: Operator Functional State: The Assessment and Prediction of Human Performance Degradation in Complex Tasks. IOS Press, Amsterdam (2003)Google Scholar
- 12.Dimensionality Reduction Algorithms: Strengths and Weaknesses. https://elitedatascience.com/dimensionality-reduction-algorithms
- 14.Goodness-of-fit tests for Binary Logistic Regression. http://support.minitab.com/en-us/minitab-express/1/help-and-how-to/modeling-statistics/regression/how-to/binary-logistic-regression/interpret-the-results/all-statistics-and-graphs/goodness-of-fit-tests/
- 15.Mehrotra, K., Mohan, C.K., Ranka, S.: Elements of Artificial Neural Networks (Complex Adaptive Systems). MIT Press, Cambridge, MA (1997)Google Scholar
- 16.Ruck, D.W., Rogers, S.K., Kabrisky, M.: Feature selection using a multilayer perceptron. J. Neural Netw. Comput. 2, 40–48 (1990)Google Scholar
- 17.Provost, F., Fawcett, T.: Data Science for Business: What You Need to Know About Data Mining and Data-Analytic Thinking. O’Reilly Media Inc, Sebastopol (2013)Google Scholar
- 18.Besson, P., Dousset, E., Bourdin, C., Bringoux, L., Marqueste, T., Mestre, D.R., Vercher, J.-L.: Bayesian network classifiers inferring workload from physiological features: compared performance. In: 2012 IEEE Intelligent Vehicles Symposium (IV), pp. 282–287. IEEE (2012)Google Scholar
- 21.Understanding Support Vector Machine algorithm from examples (along with code). https://www.analyticsvidhya.com/blog/2017/09/understaing-support-vector-machine-example-code/