Title: Fractal feature extraction of English language based on semantic analysis
Authors: Zhifen Yao
Addresses: Department of Foreign Language, Huanghuai University, ZhuMadian 463000, China
Abstract: In order to solve the problems of low extraction accuracy and long extraction time cost in traditional English language fractal feature extraction methods, an English language fractal feature extraction method based on semantic analysis is proposed. The similarity matrix is constructed to determine the fractal similarity of English language. The weight of English language fractal similarity data is determined by analytic hierarchy process (AHP). According to its weight in the whole set, the information entropy and conditional entropy of English language fractal data are determined to realise the fusion of English language fractal similarity data. Singular value decomposition is used to improve semantic analysis, reduce the dimension of English language fractal, and extract the fractal features of English language. The experimental results show that the highest accuracy of the proposed method is about 98%, and the shortest extraction time is about 1.3 s.
Keywords: semantic analysis; linguistic fractal features; AHP; information entropy; conditional entropy; singular value decomposition.
International Journal of Reasoning-based Intelligent Systems, 2022 Vol.14 No.4, pp.215 - 220
Accepted: 30 May 2022
Published online: 31 Oct 2022 *