An immune-inspired political boycotts action prediction paradigm

Abstract

The political boycotts action has been highly valued and studied by social scientists. So the research on political boycotts prediction has important theoretical research value and significance. In order to explore an efficacious method of political boycotts action prediction problem, and enhance the prediction accuracy, a novel immune-inspired political boycotts action prediction paradigm (referred as Aipo) is proposed. Aipo first initialize the pools of antibody cell and memory cells; then through the training of each antigen, antibody cells will evolve; after antibody cells' convergence, the best antibody cell will be selected to update the pool of memory cell; finally the political boycott prediction is accomplished by through memory cells with KNN classification of the test data. The proposed algorithm has the characteristics of non-linearity, and the characteristics of biological immune system, such as clonal selection, immune network and immune memory. The proposed paradigm is evaluated on World Values Survey datasets, and the experimental results show that Aipo is effective and efficient.