Συγκριτική μελέτη γραμμικών μοντέλων και νευρωνικών δικτύων για την λήψη ορθολογικών αποφάσεων = Comparative study of linear models and neural networks for rational decision making.

Χρήστος Ι. Τζώτζης


Τα τεχνητά νευρωνικά δίκτυα αποτελούν, μέχρι σήμερα, μια πολύ δημοφιλή διαδικασία προσδιορισμού της βέλτιστης λύσης διάφορων προβλημάτων. Στην παρούσα εργασία κατασκευάζονται δώδεκα διαφορετικά τεχνητά νευρωνικά δίκτυα και τέσσερα γραμμικά μοντέλα με σκοπό τον έλεγχο των περιορισμών των ορθολογικών αποφάσεων που λαμβάνονται από αυτά. Τα μοντέλα αυτά εφαρμόζονται στην πρόβλεψη της μέσης τιμής ακινήτων στα προάστια της Βοστώνης και μελετώνται οι προβλέψεις που κάνουν σε σχέση με τις πραγματικές τιμές. Η αξιολόγηση των μοντέλων αυτών έγινε με τη μέθοδο της επαναλαμβανόμενης διασταυρωμένης επικύρωσης, ενώ για την σύγκρισή τους χρησιμοποιήθηκαν η ρίζα του μέσου τετραγωνικού σφάλματος και ο συντελεστής προσδιορισμού. Από τα αποτελέσματα που προκύπτουν, εξάγεται το συμπέρασμα πως τόσο τα γραμμικά μοντέλα, όσο και τα νευρωνικά δίκτυα, είναι σε θέση να λάβουν μία ικανοποιητική αλλά και μία ορθολογική απόφαση. Η σύγκριση των μοντέλων αναδεικνύει το τεχνητό νευρωνικό δίκτυο ως το καταλληλότερο μοντέλο πρόβλεψης για το πείραμα.

Artificial neural networks are, to date, a very popular process of identifying the best solution for various problems. In the present study twelve different artificial neural networks and four linear models are constructed to control the limitations of rational decisions taken by them. These models are applied to prediction of the average property price in Boston suburbs and their predictions are made relative to real prices. The evaluation of these models was done by the repeated cross-validation method, while for comparison, the root mean square error and the coefficient of determination were used. From the results obtained, it is concluded that both linear models and artificial neural networks are able to obtain a satisfactory and also a rational decision. The comparison of models highlights the artificial neural network as the most appropriate prediction model for the experiment.

Πλήρες Κείμενο:



ΠΑΝΟΣ AΡΓΥΡΑΚΗΣ, (2001), Νευρωνικά Δίκτυα και Εφαρμογές, ΕΛΛΗΝΙΚΟ ΑΝΟΙΚΤΟ ΠΑΝΕΠΙΣΤΗΜΙΟ, σελ. 192

D. Graupe, (2013), Principles of Artificial Neural Networks.3rd Edition, World Scientific Publishers,

Auer, Peter; Harald Burgsteiner; Wolfgang Maass (2008). "A learning rule for very simple universal approximators consisting of a single layer of perceptrons". Neural Networks. 21 (5): 786–795. doi:10.1016/j.neunet.2007.12.036.

Nielsen, Michael A. (2015). "Chapter 6". Neural Networks and Deep Learning.

A. W. Harley, (2015), An Interactive Node-Link Visualization of Convolutional Neural Networks, in ISVC, pages 867-877,

https://en.wikipedia.org/wiki/Feedforward_neural_network (date of access: 1/10/2018)

M. McCord Nelson and W. T. Illingworth, (1991), A practical guide to Neural Nets, Addison–Wesley (Reading, Mass)

B. Widrow and M. E. Hoff, Adaptive Switching Circuits, 1960 WESCON

Convention, Record Part 4, pp. 96–104; Human Neurobiology, 4,229(1985).

R.C. Johnson, (1989), Neural Nose to Sniff Out Explosives at JFK Airport,

Electronic Engineering Times 536,1.

T. J. Sejnowski and C. R. Rosenber, (1986), NETtalk: A Parallel networkthat learns to read aloud J. Hopkins University Electrical Engineering and Compuetr

Science Technical Report, JHU/EECS–86/01,.

ΖΩΡΗΣ ΝΙΚΟΛΑΟΣ, ΚΑΤΣΙΝΟΥΛΑΣ ΝΙΚΟΛΑΟΣ, (2014), ΝΕΥΡΩΝΙΚΑ ΔΙΚΤΥΑ ΚΑΙ ΕΦΑΡΜΟΓΕΣ ΑΥΤΩΝ, Διατριβή (apothetirio.teiep.gr/xmlui/bitstream/handle/123456789/109/tlp_000379.pdf?...1)

Imran Maqsood, Muhammad Riaz Khan, Ajith Abraham, (2004), An ensemble of neural networks for weather forecasting, Neural Computing & Applications Volume 13, Issue 2, pp 112–122

Holger R. Maier Graeme C. Dandy, (1996), The Use of Artificial Neural Networks for the Prediction of Water Quality Parameters, Water Resources Research, pp 1013-1022

A. T. Tzallas, M. G. Tsipouras, D. I. Fotiadis, (2007), Automatic Seizure Detection Based on Time-Frequency Analysis and Artificial Neural Networks, Computational Intelligence and Neuroscience, doi: 10.1155/2007/80510

Tom Auld ; Andrew W. Moore ; Stephen F. Gull, (2007), Bayesian Neural Networks for Internet Traffic Classification, IEEE Transactions on Neural Networks ( Volume: 18 , Issue: 1), pp 223 – 239

Geoffrey Hinton, Li Deng, Dong Yu, George E. Dahl, Abdel-rahman Mohamed, Navdeep Jaitly, Andrew Senior, Vincent Vanhoucke, Patrick Nguyen, Tara N. Sainath, Brian Kingsbury, (2012), Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups, IEEE Signal Processing Magazine ( Volume: 29 , Issue: 6), pp 82 – 97

X.Wu, J.Ghaboussi, J.H.GarrettJr, (1992), Use of neural networks in detection of structural damage, Elsevier: Computers & Structures (Volume 42, Issue 4), pp 649-659

https://en.wikipedia.org/wiki/Decision-making (date of access: 1/10/2018)

George C. Homans, (1958), Social Behavior as Exchange, American Journal of Sociology 63: pp 597-606.

Kelley, H. H., (1973), The processes of causal attribution, American Psychologist, 28(2), 107-128.

Simon, Herbert A., (1956), Rational Choice and the Structure of the Environment

Simon, Herbert A., (1947), Administrative Behavior: a Study of Decision-Making Processes in Administrative Organization (1st ed.). New York: Macmillan.

Amos Tversky, Daniel Kahneman, (1974), Judgment under Uncertainty: Heuristics and Biases, Science, New Series, Vol. 185, No. 4157., pp. 1124-1131

Simon, Herbert A. (1997). Administrative Behavior: a Study of Decision-Making Processes in Administrative Organizations (4th ed.). New York: Free Press.

David Rettinger, Reid Hastie, (2001), Content Effects on Decision Making, Organizational Behavior and Human Decision Processes 85(2), pp 336-359

Simon, Herbert A. (1976). Administrative Behavior: a Study of Decision-Making Processes in Administrative Organizations (3rd ed.). New York: Free Press. pp 81

James March, Herbert Simon, (1993), Organizations, 2nd edn Oxford

H. A. Simon (1973), “The Structure of Ill-Structured Problems”, Artificial Intelligence, 4: 181-201.

March, James G. (2006). "Rationality, Foolishness, and Adaptive Intelligence". Strategic Management Journal. 27: pp 201–214

Brown, Reva (2004). "Consideration of the Origin of Herbert Simon's Theory of 'Satisficing' (1933-1947)". Management Decision. 42 (10): pp 1240–1256.

https://cran.r-project.org/web/packages/MASS/index.html(date of access 1/10/2018)

https://cran.r-project.org/web/packages/caret/caret.pdf (date of access 1/10/2018)

https://cran.r-project.org/web/packages/glmnet/glmnet.pdf (date of access 1/10/2018)

https://cran.r-project.org/web/packages/mlbench/mlbench.pdf (date of access 1/10/2018)

https://cran.r-project.org/web/packages/psych/psych.pdf (date of access 1/10/2018)

https://cran.r-project.org/web/packages/neuralnet/index.html (date of access 1/10/2018)

Lehmann, E. L.; Casella, George (1998). Theory of Point Estimation (2nd ed.). New York: Springer. ISBN 0-387-98502-6. MR 1639875.

Hyndman, Rob J.; Koehler, Anne B. (2006). "Another look at measures of forecast accuracy". International Journal of Forecasting. 22 (4): 679–688. doi:10.1016/j.ijforecast.2006.03.001.

Pontius, Robert; Thontteh, Olufunmilayo; Chen, Hao (2008). "Components of information for multiple resolution comparison between maps that share a real variable". Environmental Ecological Statistics. 15: 111–142.

Willmott, Cort; Matsuura, Kenji (2006). "On the use of dimensioned measures of error to evaluate the performance of spatial interpolators". International Journal of Geographic Information Science. 20: 89–102.

Steel, R. G. D.; Torrie, J. H. (1960). Principles and Procedures of Statistics with Special Reference to the Biological Sciences. McGraw Hill.

Glantz, Stanton A.; Slinker, B. K. (1990). Primer of Applied Regression and Analysis of Variance. McGraw-Hill. ISBN 978-0-07-023407-9.

Draper, N. R.; Smith, H. (1998). Applied Regression Analysis. Wiley-Interscience. ISBN 978-0-471-17082-2.

Devore, Jay L. (2011). Probability and Statistics for Engineering and the Sciences (8th ed.). Boston, MA: Cengage Learning. pp. 508–510. ISBN 978-0-538-73352-6.

Colin Cameron, A.; Windmeijer, Frank A.G. (1997). "An R-squared measure of goodness of fit for some common nonlinear regression models". Journal of Econometrics. 77 (2): 1790–2. doi:10.1016/S0304-4076(96)01818-0.

Imdadullah, Muhammad. "Coefficient of Determination". itfeature.com.

Legates, D.R.; McCabe, G.J. (1999). "Evaluating the use of "goodness-of-fit" measures in hydrologic and hydroclimatic model validation". Water Resour. Res. 35 (1): 233–241. doi:10.1029/1998WR900018.

Ritter, A.; Muñoz-Carpena, R. (2013). "Performance evaluation of hydrological models: statistical significance for reducing subjectivity in goodness-of-fit assessments". Journal of Hydrology. 480 (1): 33–45. doi:10.1016/j.jhydrol.2012.12.004.

https://en.wikipedia.org/wiki/Overfitting (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Tikhonov_regularization (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Lasso_(statistics) (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Elastic_net_regularization (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Cross-validation_(statistics) (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Cook%27s_distance (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Box_plot (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Backpropagation (date of access 7/10/2018)

https://en.wikipedia.org/wiki/Types_of_artificial_neural_networks (date of access 7/10/2018)

https://www.r-bloggers.com/fitting-a-neural-network-in-r-neuralnet-package/ (date of access 15/3/2018)

Burnham, K. P.; Anderson, D. R. (2002), Model Selection and Multimodel Inference (2nd ed.), Springer-Verlag

https://papers.nips.cc/paper/1895-overfitting-in-neural-nets-backpropagation-conjugate-gradient-and-early-stopping.pdf (date of access 25-07-2018)

Tzotzis Christos, Kalampakas Argyrios, Makris C. Georgios, (2018), Examining the Limited Rational Decisions of Neural Networks, IOSR Journal of Mathematics (IOSR-JM), pp 50-57.

Εισερχόμενη Αναφορά

  • Δεν υπάρχουν προς το παρόν εισερχόμενες αναφορές.