Immune Mechanism Based Ant Colony Algorithm And Its Application In Pattern Classification

In order to develop the productive forces, we have to explore and cognize the external world ceaselessly. The process of modeling can help us to gain insight into the original real world situation. However, in many fields, such as chemistry and chemical engineering, materials sciences, etc., whose internal mechanisms are not understood by human, it is of great difficulties to build an accurate mechanism model, which reflects the objective law of real world. Therefore, an alternative way is to construct a model based on observation data, which can predict the quantitative relationships between the dependent and independent variables, and it has been attracted lots of attention. Commonly, observation data in chemistry and chemical engineering are so intricate that it is difficult to mining the key elements that determine the reliability of model directly. In this situation, data mining technology can be applied to extract useful information from these complex data. As one of the most basic important tasks in data mining, classification of sample data is of extraordinary significance.In this thesis, the merits and disadvantages of ant colony optimization algorithm and other optimization algorithms are studied firstly. Then a novel algorithm, immune mechanism based ant colony optimization (IMACO), is proposed to solve combinatorial optimization problems. To demonstrate the performance of this algorithm, we conduct several TSP benchmark problems and some real pattern classification problems. Obtained results demonstrate that IMACO has well global optimization ability in combinatorial optimization problems. The major contributions of this work are summarized as follows.1. In order to overcome the phenomenon of premature convergence and complexity of solution construction in ant colony optimization, the immune mechanism, which can maintain the diversity of population and prevent the degradation of population, was introduced to ant colony optimization. It demonstrated that IMACO has well global optimization ability in combinatorial optimization problems.2. According to selection of appropriate rule expressions and rule evaluation function, rule induction problem was translated into corresponding combinatorial optimization problem, which was solved by IMACO.3. Through analysis of the interrelationship among rules expression, evaluation function and training sample data, a method was proposed to extract candidate points from every continuous attribute, which can lead to scale reduction of rule learning problem and improvement of optimization algorithm's performance. Finally, IMACO was applied to extract classification rules, and classifier constructed by these rules showed well performance in practical pattern classification problems.