Readability Assessment of Chinese Linguistic Texts Based on Dependent Syntactic Networks

Based on Chinese linguistic text features, this paper classifies dependent syntactic networks into two types: supervised and unsupervised, and researches to illustrate the inter-conversion relationship and difference between the two. Based on traditional linguistic features, the orthogonal features...

Full description

Bibliographic Details
Main Author: Zhao Jun
Format: Article
Language:English
Published: Sciendo 2024-01-01
Series:Applied Mathematics and Nonlinear Sciences
Subjects:
Online Access:https://doi.org/10.2478/amns-2024-0210
Description
Summary:Based on Chinese linguistic text features, this paper classifies dependent syntactic networks into two types: supervised and unsupervised, and researches to illustrate the inter-conversion relationship and difference between the two. Based on traditional linguistic features, the orthogonal features in linguistic features are fused with the neural network features extracted from the pre-trained model utilizing feature projection to complete the construction of the Chinese linguistics text readability assessment model, and the Chinese linguistics text readability assessment model is empirically analyzed. The results show that the accuracy of the six classification algorithms takes the range of 0.379-0.648 when only the baseline model is used, which is much lower than the performance of the corresponding classification models on the feature set constructed in this study, confirming that the algorithms in this paper can better fulfill the task of automatically assessing the readability of Chinese linguistics texts. The research results in this paper can be applied to educational scenarios to help teachers select reading materials of appropriate difficulty for learners.
ISSN:2444-8656