As a new intelligent computing method, artificial immune networks have been widely used in pattern recognition and data classification. Most of the existing artificial immune network classification algorithms have two defects: one is that the network scale is large and the calculation is complex; the other is that a single presentation of the antigen cannot guarantee the acquisition of the global optimal classifier. This paper proposes a new artificial immune network classification algorithm, which uses the strategy of each category corresponding to a single B cell, simplifies the network scale and reduces the inhibition operation between B cells of the same category. At the same time, a new affinity evaluation function based on the correct recognition rate of training samples is introduced to realize the selection strategy based on the priority of the antigen. The performance of the algorithm was fully tested using 5 groups of UCI linear data, 4 groups of mixed feature data and 1 SAR image. The results show that compared with the fuzzy C-means algorithm, the multi-valued immune (MVIN) algorithm and the clonal selection algorithm (CSA) based on classification problems, the new algorithm has certain advantages in classification accuracy and better robustness.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore