Machine Learning-Based Intelligent Scoring System for English Essays under the Background of Modern Information Technology
Table 1
The production process of an article and the calculation steps of the LDA model.
Order
Step
Calculation process
1
To determine the distribution of topics and words
LDA is used to calculate the polynomial distribution of feature words and describe the distribution with parameters
2
To determine the distribution of articles and topics
According to PD (Poisson distribution), the scale of feature words is calculated
3
To randomly determine the number N of words in the article
LDA is used to calculate the probability vector of topic distribution
4
If the number of currently generated terms is less than N, go to step 5; otherwise, step 6 is performed
From the mth essay (m = 1, 2, …, M; and M is the total number of essays), a feature word w of a topic is extracted. Then, the expectation maximization method is used to estimate the maximum likelihood of parameters, so as to establish the LDA three-layer model.
5
A topic is generated randomly according to article and topic distribution, and then a word is generated randomly based on topic and word distribution. Next, proceed to step 4.
