A Digital Analysis on Textual Coherence in English Continuation Task
The integration of computational linguistics and corpus linguistics propels the empowerment of digital intelligent technology in text analysis.By employing statistical methods to analyze linguistic data,the characteristics of large-scale real texts can be effectively presented.This study,aimed at examining the alignment of English continuation task,uses Coh-Metrix 3.0 to analyze the discourse coherence features in three dimensions(connectives,latent semantic analysis,referential cohesion)and nineteen specific indicators,and analyzes the effects of alignment through paired sample t-tests.Also,a correlation analysis is conducted regarding discourse coherence features,discourse knowledge,and composition scores based on the results of a student discourse knowledge survey questionnaire.The findings reveal:(1)the discourse coherence features of students'writing is diversely distributed,showing performance in all three dimensions,but the effect of alignment is mainly reflected in latent semantic analysis,noun overlap,and metaphorical overlap in the referential cohesion;(2)the discourse coherence features of students'writing is significantly correlated with students'discourse knowledge in the connectives and noun overlap of adjacent sentences in the referential cohesion;(3)the use of connectives is a significant predictor of composition scores.Discourse coherence is one of the criteria for high-quality compositions,and in teaching,in addition to emphasizing the use of connectives,enriching discourse coherence by emphasizing latent semantic analysis and referential cohesion can improve the quality and scores of English continuation task.
discourse knowledgetextual coherenceeffect of alignmentcontinuation task
NETL
NSTL
万方数据
