This is the code of the paper Analytic Automated Essay Scoring based on Deep Neural Networks Integrating Multidimensional Item Response Theory published in COLIG2022.
- python = 3.8.10
- tensorflow = 2.9.2
- numpy = 1.21.6
- pandas = 1.3.5
- nltk = 3.7
If you want to use word series as an input, you should download glove.6B.50d.txt from https://nlp.stanford.edu/projects/glove/ and place in the embeddings folder.
The following code executes 5-fold cross validation.
bash CV_CTS.sh
bash CV_Proposed.sh
Note that the bash file implements each of the five different seeds for each of the word and pos inputs.
Codebase from cross-prompt-trait-scoring