Data Processing Via Associative Memory

  • Nihat Adar
  • Atila Barkana
  • Sabin Kocak
Part of the International Centre for Mechanical Sciences book series (CISM, volume 307)


In this work about data processing using associative memory methods, the main principles of the operation of the human brain are taken as basis. Using the mathematical models developed earlier, a computer simulation of the associative memory is accomplished.

Reference patterns are first given to the system. Then any of these patterns, either partially erased or partially or completely deformed, is given as the key pattern to be recollected. The recollection is accomplished by obtaining a linear combination of all the patterns in the memory. In this method, the pattern which has the most similarity to the key pattern has the largest contribution in the reconstruction.

It is also possible to use a key pattern which does not exist in the memory at all. In this case, the system reconstructs a pattern which is some combination of the reference patterns, and which may or may not be similar to any of the patterns. This is just like the human memory, recognizing some object if he has seen one similar to it or not being able to tell what it is because he has never seen anything like it. It is possible to store any new patterns as reference patterns imitating the human brain function of memorizing. Thus, it is possible to reconstruct or recognize a group of data among various groups.


Associative Memory Reference Vector Reference Pattern Similarity Component Human Brain Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Fu, K.S., Keidel, W.D., Levelt, W.J.M. and Wolter, H.: Associative memory, A system-theoretical approach, Springer-Verlag Berlin, 1978, 176–230.Google Scholar
  2. 2.
    Adar, N.: Cagnsimli Hafiza Yöntemiyle Veri Isleme, M.S. Thesis, Anadolu University, Eskisehir, Turkey, 1988.Google Scholar
  3. 3.
    Lang, S., Linear algebra, Addison-Wesley Publishing Company, Amsterdam, 1973, 394–400.Google Scholar
  4. 4.
    Ince, F., Ana bilesenler, siniflandirma ve kümeleme teknikleri, TUBtTAK, Istanbul, 1979.Google Scholar
  5. 5.
    Andrews, H.C.: Introduction to Mathematical Techniques in Pattern Recognition, Wiley-Interscience, New York, 1972.zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Wien 1989

Authors and Affiliations

  • Nihat Adar
    • 1
  • Atila Barkana
    • 1
  • Sabin Kocak
    • 1
  1. 1.Anadolu UniversityEskişehirTurkey

Personalised recommendations