A Multidimensional Analysis Of The Language Differences Of University Engineering Textbooks Register

With the development of construction projects along the "Belt and Road",the demand for engineers continues to increase,and how to train them in academic writing has become an important topic of current discussion.However,the current research on the characteristics of engineering language is relatively scarce.This research used the multi-dimensional analysis model proposed by Douglas Biber(1988)to investigate the distribution of vocabulary and grammar items in engineering textbooks.First,we built a corpus of micro-engineering undergraduate textbooks,covering the civil engineering,mechanical engineering,materials science and engineering,with a total of 1,001 texts and approximately 1.8 million words.Subsequently,we used computer software for word segmentation and part-of-speech tagging,and used 111 vocabulary grammar items for factor analysis.Then,we discovered the co-occurrence patterns and interpreted them into seven dimensions:information description and action description,description and explanation of the development trend of things,description and explanation of experiments,reference,expression of uncertainty,expression of magnitude,logic and development.Furthermore,we compared the average scores of the various dimensions of subjects register,and found that the material texts are typical in the language characteristics of undergraduate engineering textbooks,the civil texts are not.Finally,with the help of multiple analysis of variance and multiple comparisons,we found the conclusions of this study are generalizable,and used the context analysis framework to make qualitative analysis of specific texts.We learned that language differences between registers are mainly caused by the settings,discussion topics,and subject attributes.This research studies language differences from the perspective of subject register,refines the field of language difference research,and enriches the results of register research and corpus linguistics research.The results are beneficial to academic Chinese writing and natural language processing.