Associative matrices are considered a type of neural network topology used to recall and recognize previously known or unknown patterns. For example, the Hebbian linear associative matrices can be trained to recognize a particular DNA sequence into another specimen sequence that may help biologist to identify similarities and other characteristics important in the knowledge and recognition of a particular sequence in another specimen. This approach may result especially beneficial in mutated sequences where mutations or other changes in the sequence as deletions and insertions are present. Associative matrices have been used to recognize characters, shapes, or specific objects from an image. Such changes are considered noisy patterns that are one of the important features of using associative matrices in this field.

The project introduces students to the use of associative matrices concepts in learning and pattern recognition. Students will experiment the use of the matrices to recognize DNA sequences and its possible mutations.
The objectives of the project is that the students understand the use of associative matrices in learning and recognizing patterns, objects, and strings such as DNA sequences in bioinformatics. After the completion of this project the student should be able to:
  • Understand the use and implementation of associative matrices.
  • Recognize the capability of these type of networks to associate new matrices with previously learned patterns allowing to recognize patterns with noise or different from the original patterns.
  • Understand the use of associative networks and pattern recognition.
  • Understand the mathematical foundations of matrix or linear algebra and its applications.
  • Introduce the students to the applications of artificial intelligence, learning, and neural networks to the bioinformatics field.
The student should have basic knowledge of matrix or linear algebra. The student should have knowledge and experience of a programming language such as C++ or Matlab to implement the models and test the algorithms.
Theory and examples of the use of associative matrices and the Hebb rule, morphological associative matrices, and bioinformatics can be found in the:
  • Stuart Russell and Peter Norvig. Artificial Intelligence: A Modern Approach, 2nd edition. Prentice Hall, Upper Saddle River, NJ, USA, 2003. Chapter 20 and Appendix A.
  • Jean-Michel Claverie, Cedric Nothedame. Bionformatics for Dummies, 1st Edition. For Dummies, 2003. Chapters 1 and 2.
The detailed project description is available in the PDF file CompleteProject.pdf. You will need the free Adobe Acrobat Reader to view this file.
This project is customizable to accommodate different approaches to teaching and different implementations. Additional exercises are also included for students seeking more extended challenges.
A sample syllabus used at the University of Puerto Rico at Mayaguez when this project was assigned is available at:
Syllabus for AI Course at the University of Puerto Rico

Additional readings are included in the Background section above.