Dimensionality reduction applied to logical judgments

Purpose: To study the effect of the application of the dimensionality reduction in logical judgment (or logical reasoning, logical inference) programs. Methods: Use enumeration and dimensionality reduction methods to solve logical judgment problems.The effect of the two methods is illustrated in the form of a case study. Results: For logical judgmentproblems, using enumeration method to find the best answer is a comprehensive and fundamental method, but the disadvantage is that it is computationally intensive and computationally inefficient. Compared with the ideas of parallel treatment of known conditions by enumeration method, the application of dimensionality reduction thinking was built on the basis of fully mining information for feature extraction and feature selection. Conclusions: The dimensionality reduction method was applied to the logical judgment problems, and on the basis of fully mining information, the dimensionality reduction principle of statistics were applied to stratify and merge variables with the same or similar characteristics to achieve the purpose of streamlining variables, simplifying logical judgment steps, reducing computation and improving algorithm efficiency.