Automatic Intelligent Korean Character Semantic Recognition and Analysis Framework based on Machine Learning

Abstract

Automatic intelligent korean character semantic recognition and analysis framework based on machine learning is implemented in this manuscript. In the research process of this article, algorithm uses the idea of genetic system programming paradigm to derive characters in the initial system, while the initial L system operates a set of actual related parameter strings represented by letters to expand the grammar. There are core innovations as follows. (1) The random optimization algorithm is only required to know the unbiased estimation of the gradient of the objective function, especially for machine learning problem of finite samples. Hence, it is extended into higher complexity. (2) The random optimization method only needs to calculate the gradient of the objective function, it is therefore used to enhance the overall efficiency. The experiment compared with the latest methods have proven the performance.