An Immune Approach to Recognition of Handwritten Words

This paper presents an immune-based approach to recognition of handwritten words, or more generally, to binary classification. It is inspired by the negative selection mechanism attending the discrimination between self and nonself molecules in living organisms. Our approach incorporates two types of detectors: binary and real-valued. Relatively short binary detectors are used for primary detection, while the real valued detectors are used to resolve eventual doubts. Such a hybrid solution is much more economical in comparison with "pure" approaches. The binary detectors are faster than real-valued ones with respect to the number of operations needed to compare two items. Thus, we gain simplicity and speed of comparisons.