Research On Neighborhood Detector Optimization Based On DNA Genetic Algorithm

Anomaly detection based on intelligent technologies such as data mining,machine learning,and artificial immunity has been particularly active in research fields such as computer graphics and network security.Artificial immunity is an artificial intelligence technology that simulates biological immunity.Its technology is becoming more and more perfect,its functions are becoming more powerful,and its scope of application is also expanding.Applying it to intrusion detection systems is the frontier of dealing with network intrusion problems.In the current field of artificial immunity,real-valued morphological space is used more,and the real-valued morphological space has the problems of high overlap rate and insufficient coverage.The clonal selection algorithm based on the neighborhood morphological space can solve the problem of multi-dimensional attribute samples.The dimensional disaster problem of the value clone selection algorithm,but the detection efficiency of this algorithm is low,the optimization process is easy to premature and fall into the local optimal,and the detection accuracy is also unstable.To address these issues,this article will add DNA vaccines and adaptive mechanisms to the clonal selection algorithm in the neighborhood morphological space,and introduce a bone marrow model,and propose a DNA genetic algorithm-based neighborhood detector optimization algorithm and its adaptive optimization model.Various aspects propose solutions and solutions for the above problems.First,the algorithm replaces the real-valued morphological space with the neighboring morphological space,and optimizes the genetic algorithm in the neighboring morphological space by introducing a DNA mechanism,and extracts gene fragments from the DNA point of view and mimics the biochip microarray by means of vaccine extraction and adaptive adjustment.Array structure constructs neighborhood DNA vaccine sets,constructs candidate detectors basedon excellent vaccines,and constructs candidate detectors that incorporate relevant DNA vaccine fragments using operators such as high-frequency mutation;then,using neighborhood DNA genetic algorithms as the core,using Neighborhood Negative Selection Algorithm(NNSA)affinity calculation determines whether the candidate sample meets the tolerance training requirements,speeding up the iterative optimization rate of the detector.Finally,a biological immune adaptive mechanism is added,and a neighborhood detector adaptive optimization model is designed based on DNA genetic algorithm,so as to further dynamically optimize the detector performance,construct an artificial immune abnormality detection model that meets actual needs,and solve the detector under real-value morphological space There are problems of premature optimization and local optimization,so as to improve the detection performance of the artificial immune intrusion detection system.Experiments show that the optimized neighborhood detector shows a good level in terms of generation,optimization rate and detection rate.The algorithm and model in this paper can further improve the artificial immune theory and make it better applied in practical applications such as intrusion detection.