Research and Design of Automatic Scoring Algorithm for English Composition Based on Machine Learning

Table 1

Text feature description.

Overview of lexical features	Overview of syntactic features
The proportion of word list size to composition Length (not repeated)	Average sentence length and variance
Number of sentences whose length is greater than a fixed value (e.g., 4, 8, 12)	Number of sentences whose length is greater than a fixed value (e.g., 4, 8, 12)
Statistical characteristics such as average character length (mean word length, median, standard deviation)	Average number of verbs, nouns, modal verbs, prepositions in sentences
The proportion of nouns, adjectives, verbs and prepositions	Average number of punctuation marks in a sentence
Number of high-frequency words	The number of sentences that fully express the meaning
The size of the word list after removing the stop word	The number of clauses and the average length of clauses