ANALYSIS AND SEMANTIC INTERPRETATION METHODS OF DECISION-MAKING PROCESS IN NEURAL NETWORK SUPERVISED LEARNING CLASSIFICATION MODELS
Article Sidebar
Main Article Content
Abstract
The paper focuses on the problem of interpreting the behavior of a feedforward neural network classifier model after learning, and proposes solutions to analyze and describe model output based on local linear approximations.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
Goodfellow, I. Deep learning / I. Goodfellow, B. Yoshua, C. Aaron. – Cambridge : The MIT Press, 2016. – 775 p.
Курочкин, А.В. Оптимизация процесса принятия решений в медицинских экспертных системах на базе нечеткой логики с использованием исторических данных / А.В. Курочкин, В.С. Садов, О.М. Демиденко // Проблемы физики, математики и техники. – 2019. – № 1 (38). – с. 78–84.
Ioannou, Y. Decision Forests, Convolutional Networks and the Models in-Between / Y. Ioannou, D. Robertson, D. Zikic, [et al.] // arXiv:1603.01250[cs] [Электронный ресурс]. – 2016. – Режим доступа: https://arxiv.org/abs/1603.01250. – Дата доступа: 08.10.2019.
Lundberg, S.M. A Unified Approach to Interpreting Model Predictions / S.M. Lundberg, S.-I. Lee // Advances in Neural Information Processing Systems. – 2017. – № 30. – P. 4765–4774.
Knowledge Acquisition for Expert Systems : A Practical Handbook / ed. by A.L. Kidd. – Springer Science & Business Media, 2012. – 208 p.
Most read articles by the same author(s)
- K. HALAVATAYA, V. SADAU, MODEL OF IMAGE ACQUISITION FOR 3D SCENE RECONSTRUCTION FROM VIDEOENDOSCOPIC IMAGING DATA, Vestnik of Polotsk State University. Part C. Fundamental Sciences: No. 12 (2019)