Computerized System for Cooperation Model's Selection based on Intelligent Fuzzy Technique

Abstract

In this paper, the current state of the academic and industrial cooperation, as well as intelligent fuzzy technique for solving the task of cooperation model's selection in academia-industry collaboration are analyzed. The choice of cooperation model is an actual and important topic for the constant and substantial progress in increasing the quality of education for higher education systems of the different countries. Successful academia-industry cooperation will promote more active implementation of compatible programs and projects. One of the perspective approach to solve the current problem of cooperation model's selection is based on the implementation of the computerized systems or decision support systems (DSSs). In addition, existing software tools for the development of the DSSs based on intelligent fuzzy technique are analyzed. Among their limitations are: the restrictive numbers of aggregation and defuzzification methods; a limited number of membership functions types; the lack of discrete fuzzy logic output. The authors developed a computerized system (DSS) for cooperation model's selection based on intelligent fuzzy technique, in particular, based on Mamdani-type fuzzy inference engine which allows: (a) more flexible settings for input signals, (b) the use of more wide row of existing methods for aggregation and defuzzification in fuzzy data processing, and (c) outputting results in both discrete and continuous forms. DSS's testing results for several real examples of academia-industry cooperation confirm the correctness and efficiency of cooperation model's selection. Developed based on C# software for intelligent DSS expands the functionality to research, analyze and solve the corresponding task of cooperation model's selection.

Keywords

• academia-industry cooperation
• cooperation model's selection
• DSS
• fuzzy inference engine
• discrete output
• continuous output
• membership function
• aggregation
• defuzzification

-

Selection of cooperation model is and actual and important topic for continuous and considerable development of education quality in higher education institutions of the country, its theoretical and, especially, practical part [ 1-3 ]. Also one cannot disregard the fact that institutions are in the foreground of research, which results are applied directly to the tasks confronted by the industry. The intercommunication between the industry and the scientific community is rather complicated, it has countless number versatile aspects – from cooperation during schooling of a new generation of beginner-level specialists to partnership in research aimed towards solving problems of tomorrow [ 4-7 ]. Successful cooperation is going to assist a more active implementation of joint programs and projects. Besides, within partnership students, teachers and scientific workers of higher education institutions are going to have opportunities to raise their competitiveness in the job market [ 2, 4, 8-10 ].

Creation of an effective network between academic and industrial partners is going to allow the higher education to reach modern specialist training formats which organically develop their theoretical and practical competence [ 1, 4-6 ]. 2

Related Works and Problem Statement

Fuzzy model is meant to be an informational-logical model of a system which is based on the theory of fuzzy multitudes and fuzzy logic [ 11-13 ]. Therefore, the separate stages of fuzzy modelling are [ 14, 15 ]: analysis of problematic situation; structuration of a subject area and building of a fuzzy model; making a computing experiments with the fuzzy model; application of results of the computing experiments; correction or revision if the fuzzy model. Fuzzy techniques can be applied to solve different problems. In the paper [ 16 ] solved the problem of a fuzzy observer development for the clamping force automatic control system of a mobile robot. The paper [ 17 ] discusses the design of modern tactile sensor systems for intelligent (with fuzzy approach) and adaptive robots.

According to the latest research within various consortiums there has been proven that today effectiveness of cooperation depends on the cooperation model between university and industrial partners. Let’s consider four ( m  4 ) cooperation models as alternative decisions (solutions) Ei , i  1,..., m . Herewith the model A1 corresponds to the decision E1 (cooperation between a university and a company to organize education and training, exchange of knowledge, purposeful training of the staff). Model A2 corresponds to the decision E2 (organization and support of the cooperation result certification processes). Model B corresponds to the decision E3 (creation of joint center of scientific research, development of joint scientific projects). Model C corresponds to the decision E4 (creation of scientific groups of students and independent business-oriented companies) [ 2-4, 7, 18 ].

The problem of selecting the cooperation model appears before a university in the beginning of a co-working and in conditions of altering the direction of development. Analysis of literature sources allows to mark emphasize on 27 main factors which influence on the selection of the model, particularly experience level of students, level of their involvement in international programs of exchange, experience level of company staff, education-qualification level of the company etc. [ 2, 4 ]. In the papers [ 18, 19 ] there were researched the problem of selecting of a model of cooperation and were built a system of fuzzy logic inference which is provided in Fig. 1. In this research, authors will lean on the given fuzzy inference system (FIS). The authors developed the fuzzy intelligent DSS (Fig. 1) for choosing the rational model of cooperation of universities and IT companies [ 18, 19 ]. Corresponding DSS includes 27 input coordinates X  x j , j  1,..., 27 , one output y , which are interconnected with fuzzy dependencies yk  f  x1, x2 ,..., x27  , k  1,11 of the relevant rule bases of 11 subsystems. Fig. 1 shows the version of proposed by the authors hierarchically-organized structure of DSS, which was formed from the decomposition of the input coordinate vector with their association to the group combination. In this case, the appropriate DSS subsystems (Fig. 1) in particular FSS1, FSS2 ,..., FSS10 , FSS11 implements the following functional dependencies for structure St   y1, y2 ,..., y10 , y of DSS [ 18, 19 ]:  y1  f1  x1, x2 , x3  , y2  f2  x4 ,..., x7 , y3  f3  x8 ,..., x13 ,     y4  f4  x14 ,..., x17  , y5  f5  x6 , x18 , x19  , y6  f6  x18 ,..., x23 ,  St   .  y7  f7  x24 ,..., x27  , y8  f8  y1, y2  , y9  f9  y3 , y4  , y10  f10  y5 , y6 ,    y  f11  y7 , y8 , y9 , y10 .  To evaluate the input coordinates X  x j , j  1,..., 27 and intermediates y8 , y9 , y10 there were selected three linguistic terms (LTs) with triangular membership function (MF). Some of them: x1 is the level of scientific novelty of diploma (DP) and master's works (MW); x2 is practical significance of DP and MW; x7 is the success of students' learning; x9 is the number of patents; x10 is the number of grants; x11 is the level of scientific publications of the University Chair; x15 is the level of business course teaching; x16 is the experience in the organization of student companies; x19 is the staff experience level of IT companies; x20 is the education level of IT companies; y9 is the scientific and business level of the University Chair. To evaluate the intermediate coordinates y1, y2 , y4 , y5 , y7 there were selected five LTs with triangular MF, including «low - L», «low than medium - LM», «medium - M», «higher than medium - HM», «high - H». Some of them: y2 is the level of professional orientation of students; y4 is the level of business orientation of the University Chair. To evaluate the intermediate coordinates y3 , y6 there were selected seven LTs with triangular MF, where y3 is the level of scientific activity of the University; y6 is the assessment of potential scientific and educational support level of the IT company. To evaluate the output variable y (the model of cooperation) there were selected four LTs with triangular MF, including «A1», «A2», «B», «C» [ 18, 19 ]. 3

Fuzzy Approach for DSS’s Design

Application of fuzzy multitude and fuzzy logic theory in design of DSS allows completing tasks on the intellectual level using expert knowledge bases [ 20-23 ].

An important problem of synthesis of DSS based on fuzzy logic inference is the complexity of making decisions with an altering input data structure of the system. This is related to the necessity of development of effective approaches to the correction of fuzzy knowledge bases. The necessity of corresponding correction or workaround of the rules, considering input signals, which are excluded from the vector of input coordinates during the selection of the person making decisions, appears in a certain application of the DSS in an interactive mode. In such interactive modes, the person can lower the dimensions of input coordinates of the DSS excluding the signals, which are the least important for the DSS and will not take part in the following process of making decisions [ 4, 19, 24-27 ].

Membership function (MF) represents the degree of membership of each element of a space to the given fuzzy multitude. More often, the membership functions from the Table 1 are used. In software application developed by authors, the following MFs were implemented: triangular and Gaussian (symmetrical) [ 24, 28, 29 ].

In fuzzy logic expressions more often as operators of intersection А ∩ В (logical operator AND) various t-norms which define the form of implementation of a corresponding operation are used [ 25, 30 ].

Name Triangular

Gaussian (symmetrical) Trapezoidal Sigmoidal In the Table 2 there are represented some un-parameterized t-norm operators which were implemented by the authors in the software application [ 24, 25 ].  AB  z   MIN  A  x  ,  B  y   AB  z    A  x    B  y  Defuzzification of a fuzzy multitude which is a result of an inference means an operation defining an accurate value y* which would represent this multitude in the most rational way. There are different methods of defuzzification, the following are used more often [ 24, 25 ]: method of an average maximum; method of the first maximum; method of the last maximum; method of the center of heaviness; method of the center of sums; method of altitudes.

Formally, Mamdani [ 31, 32 ] algorithm is implemented by the following stages. Stage 1. Forming of knowledge bases of fuzzy logic inference system.

Stage 2. Fuzzification of input variables using membership functions of corresponding linguistic terms.

Stage 3. Activation of antecedents of rules.

Stage 4. Aggregation of antecedents of rules.

Stage 5. Accumulation of consequents (conclusions) of fuzzy rules.

Stage 6. Defuzzification of fuzzy multitude.

At the moment there exist a few pieces of software which allow to solve the problem of assessment and selection of cooperation model. The most popular ones are Fuzzy Logic Toolbox for the MatLAB packet and FuzzyTECH environment. The limitations of them are absence of some methods of aggregation and defuzzification, few membership functions, absence of discrete fuzzy logic inference [ 24, 33-35 ]. 4

The Structure of Developed Computerized Intelligent DSS

The developed computerized intelligent DSS allows: to make more flexible setting of input parameters comparing to analogs, using existing methods of aggregation and defuzzification, to make logical inference in discrete and continuous forms [ 4, 24 ].

Authors should represent all (6 input, 2 intermediate and 1 output) linguistic variables (LVs) for the developed computerized intelligent DSS: “X1” is the level of professional orientation of students; “X2” is the level of business orientation of the university department; “X3” is the level of scientific activity of the university department; “X4” is the assessment of the possible exchange of knowledge among the personnel of the IT-company; “X5” is the assessment of the possible level of scientific support from the IT-company; “X6” is the assessment of the possible level of educational support from the IT-company; “Y1” is collaboration level of university; “Y2” is collaboration level of IT-company; “Y” is cooperation model.

Let’s review the structure and functions of the developed intelligent DSS for assessment and selection of the cooperation model in academia-industry cooperation on different combinations of settings, specifically, triangular and Gaussian MFs, t-norms MIN and PROD, and the MEAN operator [ 18, 24, 25, 36-39 ].

The process of creation of the first LV “X1” in the range [0, 50] with three LTs of triangular MF shown in Fig. 2. You can see three tabs in this figure. The first tab gives information about the name of the LV and its type. The second tab describes the variable in LTs. The third tab gives descriptive characteristic and comments of the current linguistic variable.

Thus was done 6 input LVs, 2 intermediate and 1 output variables. In the next step there was created a rules block (RB) (Fig. 3a). The window of rules for the first RB with LVs X1, X2 and X3 shown in Fig 3b.

The created computerized intelligent DSS shown in Fig. 4. After the launch of the system and input of the data (X1 = 50, X2 = 5, X3 = 3, X4 = 85, X5 = 8, X6 = 9) for our university and IT-company “ITServ”, the result of cooperation was computed as one of the models (A1, A2, B, C) in discrete and continuous forms of fuzzy inference engine (for example, model “C” with 97 points). It based on the Mamdani algorithm using the t-norm MIN (Fig. 5), t-norm PROD (Fig. 6a) and the MEAN operator (Fig. 6b). The results of research of the influence of changing the defuzzification method (left maximum and the right maximum methods) on the result of cooperation model’s selection while using the Gaussian (symmetrical) MF and the triangular MF are shown in Fig. 8 and Fig. 9. The corresponding variability allows expanding the functionality of the computerized system for cooperation model’s selection based on intelligent fuzzy technique [ 24, 40, 41 ]. Fig. 9. The result of cooperation model’s selection with triangular MF using the defuzzification method of left maximum (a) and the right maximum (b) Therefore, the results of the developed computerized system show that using different membership functions, operators of t-norms, aggregation and defuzzification methods to solve the current task, the best cooperation model is model “C” (creation of scientific groups of students and independent business-oriented companies), but with different output values for further research and comparisons. 5

Conclusions

The authors developed a computerized system (DSS) for cooperation model’s selection based on intelligent fuzzy technique, in particular, based on Mamdani-type fuzzy inference engine which allows: (a) more flexible settings for input signals, (b) the use of more wide row of existing methods for aggregation and defuzzification in fuzzy data processing, and (c) outputting results in both discrete and continuous forms. DSS’s testing results for several real examples of academia-industry cooperation confirm the correctness and efficiency of cooperation model’s selection. Developed based on C# software for intelligent DSS expands the functionality to research, analyze and solve the corresponding task of cooperation model’s selection.

References

1. Bogel , S. , Stieglitz , S. , Meske , C. : A Role Model-Based Approach for Modelling Collaborative Processes . In: Business Process Management Journal , Vol. 20 , No. 4 , pp. 598 - 614 ( 2014 )

2. Kondratenko , G. , Kondratenko , Y. , Sidenko , I. : Fuzzy Decision Making System for Model-Oriented Academia /Industry Cooperation: University Preferences. In: Berger-Vachon , C. , Gil Lafuente , A. , Kacprzyk , J. , Kondratenko , Y. , Merigó , J. , Morabito , C. (eds) Complex Systems: Solutions and Challenges in Economics, Management and Engineering. Studies in Systems, Decision and Control , Vol. 125 . Springer, Cham, pp. 109 - 124 ( 2018 ). doi: 10 .1007/978-3- 319 -69989- 9 _ 7

3. Meerman , A. , Kliewe , T. (Eds.): Fostering University-Industry Relationships, Entrepreneurial Universities and Collaborative Innovations . Good Practice Series ( 2013 )

4. Kondratenko , Y. , Kondratenko , G. , Sidenko , I. : Intelligent Decision Support System for Selecting the University-Industry Cooperation Model Using Modified AntecedentConsequent Method . In: Medina, J. et al. ( eds) Information Processing and Management of Uncertainty in Knowledge-Based Systems . Theory and Foundations. IPMU 2018 . Communications in Computer and Information Science , Vol. 854 . Springer, Cham, pp. 596 - 607 ( 2018 ). doi: 10 .1007/978-3- 319 -91476-3_ 49

5. Kondratenko , Y. , Simon , D. , Atamanyuk , I.: University Curricula Modification Based on Advancements in Information and Communication Technologies . In: Ermolayev, V. et al. (eds) Proceedings of the 12th International Conference on Information and Communication Technologies in Education, Research, and Industrial Application . Integration, Harmonization and Knowledge Transfer , ICTERI' 2016 , CEUR-WS , Vol. 1614 , pp. 184 - 199 . Kyiv, Ukraine ( 2016 ).

6. Caloghirou , Y. , Tsakanikas , A. , Vonortas , N.S. : University-Industry Cooperation in the Context of the European Framework Programmes . In: The Journal of Technology Transfer , Vol. 26 , Iss . 1- 2 , pp. 153 - 161 ( 2001 ). doi: 10 .1023/A:1013025615518

7. Barnes , T. , Pashby , I. , Gibbons , A. : Effective University -Industry Interaction : A Multicase Evaluation of Collaborative R & D Projects . In: European Management Journal , Vol. 20 , Iss . 3, pp. 272 - 285 ( 2002 ).

8. Valentin , E.M.M. : University-Industry Cooperation : A Framework of Benefits and Obstacles . In: Industry and Higher Education , Vol. 14 , Iss . 3, pp. 165 - 172 ( 2000 ). doi: 10.5367/000000000101295011

9. Franco , M. , Haase , H.: University-industry cooperation: Researchers' motivations and interaction channels . In: Journal of Engineering and Technology Management , Vol. 36 , pp. 41 - 51 ( 2015 ). doi: 10 .1016/j.jengtecman. 2015 . 05 .002

10. Seppo , M. , Lilles , A. : Indicators Measuring University-Industry Cooperation. In: Discussions on Estonian Economic Policy , Vol. 20 , Iss . 1, pp. 204 - 225 ( 2012 ).

11. Chen , M.-Y., Linkens , D.A. : Rule-base Self-generation and Simplification for Data-driven Fuzzy Models . In: Fuzzy Sets and Systems , Vol. 142 , Iss . 2, pp. 243 - 265 ( 2004 ). doi: 10 .1016/S0165- 0114 ( 03 ) 00160 -X

12. Cornejo , M.E. , Diaz-Moreno , J.C , Medina , J.: Multi-adjoint Relation Equations: A Decision Support System for Fuzzy Logic . In: International Journal of Intelligent Systems , Vol. 32 , Iss . 8, pp. 778 - 800 ( 2017 ). doi: 10 .1002/int.21889

13. Drozd A. , Lobachev M. , Drozd J.: The problem of on-line testing methods in approximate data processing . In: Proc. 12th IEEE International On-Line Testing Symposium , pp. 251 - 256 . Como, Italy ( 2006 ). doi: 10 .1109/IOLTS. 2006 .61

14. Kacprzyk , J. , Zadrozny , S. , Tre , G.: Fuzziness in Database Management Systems: Half a Century of Developments and Future Prospects . In: Fuzzy Sets and Systems , Vol. 281 , pp. 300 - 307 ( 2015 ). doi: 10 .1016/j.fss. 2015 . 06 .011

15. Lodwick , W.A. , Untiedt , E.: Introduction to Fuzzy and Possibilistic Optimization . In: Lodwick, W.A. , Kacprzyk , J . (eds) Fuzzy Optimization . Studies in Fuzziness and Soft Computing , Vol. 254 . Springer, Berlin, Heidelberg, pp. 33 - 62 ( 2010 ). doi: 10 .1007/978-3- 642 -13935- 2 _ 2

16. Kondratenko , Y.P. , Kozlov , O.V. , Gerasin , O.S. , Zaporozhets , Y.M. Synthesis and research of neuro-fuzzy observer of clamping force for mobile robot automatic control system . In: 1st IEEE International Conference on Data Stream Mining and Processing , DSMP, pp. 90 - 95 . Lviv, Ukraine ( 2016 ). doi: 10 .1109/DSMP. 2016 .7583514

17. Kondratenko , Y.P. , Klymenko , L.P. , Kondratenko , V.Y., Kondratenko , G.V. , Shvets , E.A. : Slip displacement sensors for intelligent robots: Solutions and models . In: Proceedings of the IEEE 7th International Conference on Intelligent Data Acquisition and Advanced Computing Systems, IDAACS , Vol. 2 , pp. 861 - 866 . Berlin, Germany ( 2013 ). doi: 10 .1109/IDAACS. 2013 .6663050

18. Kondratenko , Y.P. , Kondratenko , G.V. , Sidenko , I.V. : Knowledge-Based Decision Support System with Reconfiguration of Fuzzy Rule Base for Model-Oriented AcademicIndustry Interaction . In: Gil-Lafuente , A. , Merigó , J. , Dass , B. , Verma , R . (eds) Applied Mathematics and Computational Intelligence . FIM 2015. Advances in Intelligent Systems and Computing , Vol. 730 . Springer, Cham, pp. 101 - 112 ( 2018 ). doi: 10 .1007/978-3- 319 - 75792- 6 _ 9

19. Kondratenko , Y. , Kondratenko , G. , Sidenko , I. : Two-stage method of fuzzy rule base correction for variable structure of input vector . In: IEEE 1st Ukraine Conference on Electrical and Computer Engineering , UKRCON, pp. 1043 - 1049 . Kyiv, Ukraine ( 2017 ). doi: 10 .1109/UKRCON. 2017 .8100409

20. Opanasenko , V. , Kryvyi , S. : Synthesis of multilevel structures with multiple outputs . In: 10th International Conference of Programming, UkrPROG , pp. 32 - 37 . Kyiv, Ukraine ( 2016 ).

21. Mendel , J.M. : Uncertain Rule-Based Fuzzy Logic Systems: Introduction and New Directions . Prentice Hall PTR , NJ ( 2001 ).

22. Merigo , J.M. , Gil-Lafuente , A.M. , Yager , R.R.: An Overview of Fuzzy Research with Bibliometric Indicators . In: Applied Soft Computing, No. 27 , pp. 420 - 433 ( 2015 ).

23. Messarovich , M.D. , Macko , D. , Takahara , Y. : Theory of Hierarchical Multilevel Systems . Academic Press, N.Y. ( 1970 ).

24. Piegat , A. : Fuzzy Modeling and Control . Springer, Heidelberg ( 2001 ).

25. Rotshtein , A.P. : Intelligent Technologies of Identification: Fuzzy Logic, Genetic Algorithms, Neural Networks . Universum Press, Vinnitsya ( 1999 ). (in Russian)

26. Setnes , M. , Babuska , R. , Kaymak , U., van Nauta Lemke, H.R. : Similarity measures in fuzzy rule base simplification . In: IEEE Transactions on Systems, Man, and Cybernetics , Part B (Cybernetics ) , Vol. 28 , Iss . 3, pp. 376 - 386 ( 1998 ). doi: 10 .1109/3477.678632

27. Setnes , M. : Simplification of Fuzzy Rule Bases . In: Proceedings of the International Conference, EUFIT , pp. 1115 - 1119 , Aachen, Germany ( 1996 ).

28. Zadeh , L.A. : Fuzzy sets . In: Information and Control , Vol. 8 , Iss . 3, pp. 338 - 353 ( 1965 )

29. Zimmerman , H.J.: Fuzzy Set Theory . Kluwer, Boston ( 1991 ).

30. Gil-Lafuente , A.M. , Merigó , J.M. : Decision Making Techniques in Political Management . In: Lodwick, W.A. , Kacprzyk , J . (eds) Fuzzy Optimization . Studies in Fuzziness and Soft Computing , Vol. 254 . Springer, Berlin, Heidelberg, pp. 389 - 405 ( 2010 ). doi: 10 .1007/978- 3- 642 -13935-2_ 18

31. Król , D. , Lasota , T. , Trawiński , B. , Trawiński , K. : Comparison of Mamdani and TSK Fuzzy Models for Real Estate Appraisal . In: Apolloni, B. , Howlett , R.J. , Jain , L . ( eds) Knowledge-Based Intelligent Information and Engineering Systems . KES 2007. Lecture Notes in Computer Science , Vol. 4694 . Springer, Berlin, Heidelberg, pp. 1008 - 1015 ( 2007 ). doi: 10 .1007/978-3- 540 -74829-8_ 123

32. Verma , N.K. , Singh , V. , Rajurkar , S. , Aqib , M. : Fuzzy Inference Network with Mamdani Fuzzy Inference System . In: Verma, N. , Ghosh , A . (eds) Computational Intelligence: Theories, Applications and Future Directions - Volume I. Advances in Intelligent Systems and Computing , Vol. 798 . Springer, Singapore, pp. 375 - 388 ( 2019 ). doi: 10 .1007/ 978 -981-13- 1132-1_ 29

33. Aras , M.S.M. , Ali , F.A. , Azis , F.A. , Hamid , S.M.S.S.A. , Basar , M.F.H.M.: Performances evaluation and comparison of two algorithms for Fuzzy Logic rice cooking system (MATLAB Fuzzy Logic Toolbox and FuzzyTECH) . In: IEEE Conference on Open Systems , pp. 400 - 405 . Langkawi, Malaysia ( 2011 ). doi: 10 .1109/ICOS. 2011 .6079267

34. Kondratenko , Y. , Kondratenko , G. , Sidenko, I.: Multi-criteria Decision Making and Soft Computing for the Selection of Specialized IoT Platform . In: Chertov, O. , Mylovanov , T. , Kondratenko , Y. , Kacprzyk , J. , Kreinovich , V. , Stefanuk , V. (eds) Recent Developments in Data Science and Intelligent Analysis of Information. ICDSIAI 2018. Advances in Intelligent Systems and Computing , Vol. 836 . Springer, Cham, pp. 71 - 80 ( 2019 ). doi: 10 .1007/978-3- 319 -97885- 7 _ 8

35. MATLAB for Artificial Intelligence . Available on [https://www.mathworks.com/].

36. Zadeh , L.A. , Abbasov , A.M. , Shahbazova , S.N. : Fuzzy-Based Techniques in Human-Like Processing of Social Network Data . In: Int. J. of Uncertainty, Fuzziness and KnowledgeBased Systems , Vol. 23 , Iss . 1, pp. 1 - 14 ( 2015 ). doi: 10 .1142/S0218488515400012

37. Babuska , R. , Verbruggen , H.B.: A New Identification Method for Linguistic Fuzzy Models . In: International Conference on Fuzzy Systems . pp. 905 - 912 . Yokohama, Japan ( 1995 ). doi: 10 .1109/FUZZY. 1995 .409790

38. Gil-Aluja , J. : Investment in Uncertainty. Kluwer Academic Publishers, Dordrecht, Boston, London ( 1999 ).

39. Fahad , S. , Shah , A. : Intelligent Testing using Fuzzy Logic . In: Iskander, M. (eds) Innovations in E-learning , Instruction Technology, Assessment, and Engineering Education . Springer, Dordrecht, pp. 95 - 98 ( 2007 ). doi: 10 .1007/978-1- 4020 -6262-9_ 17

40. Atamanyuk , I.P. : Algorithm to determine the optimal parameters of a polynomial Wiener filter-extrapolator for nonstationary stochastic processes observed with errors . In: Cybernetics and Systems Analysis , Vol. 47 , Iss . 2, pp. 305 - 310 ( 2011 ). doi: 10.1007/s10559- 011-9312-8

41. Figueroa-García , J.C. , Ochoa-Rey , C. , Avellaneda-González , J.: A Self-Organized Fuzzy Neural Network Approach for Rule Generation of Fuzzy Logic Systems . In: Huang, D.S. , Gupta , P. , Wang , L. , Gromiha , M. (eds) Emerging Intelligent Computing Technology and Applications. ICIC 2013 . Communications in Computer and Information Science , Vol. 375 . Springer, Berlin, Heidelberg, pp. 25 - 30 ( 2013 ). doi: 10 .1007/978-3- 642 -39678- 6 _ 5