Feature Description

This feature vets through the list of common mistakes provided and generates corresponding messages to describe the difference between the responses and the answers.

It is driven mainly by AST pattern matching of the Sympy-parsed representations of the responses and answers. The function and its dependencies are hosted as a docker image on ECR and is deployed on AWS lambda.

Usage

The input to the function is in this JSON format:

{
  "commonMistakes": [
    {
      "submission": "4",
      "latexSubmission": "5",
      "simplifiedSubmission": "5",
      "frequency": 6,
      "feedback": "Incorrect",
      "mark": 0,
      "color": "",
      "responseAreaId": "655f95f1-c606-41a0-8905-a8baaa4a8a9f",
      "universalResponseAreaId": "da886822-618a-4c32-a332-18b0783e33dc",
      "params": {
        "strict_syntax": false,
        "symbols": {}
      },
      "answer": "-4"
    },
    {
      "submission": "3",
      "latexSubmission": "3",
      "simplifiedSubmission": "3",
      "frequency": 5,
      "feedback": "Incorrect",
      "mark": 0,
      "color": "",
      "responseAreaId": "655f95f1-c606-41a0-8905-a8baaa4a8a9f",
      "universalResponseAreaId": "da886822-618a-4c32-a332-18b0783e33dc",
      "params": {
        "strict_syntax": false,
        "symbols": {}
      },
      "answer": "4"
    }]
}

The function returns data in the following format:

{
  "statusCode": 200,
  "body": [
    {
      "submission": "4",
      "latexSubmission": "5",
      "simplifiedSubmission": "5",
      "frequency": 6,
      "feedback": "Incorrect",
      "mark": 0,
      "color": "",
      "responseAreaId": "655f95f1-c606-41a0-8905-a8baaa4a8a9f",
      "universalResponseAreaId": "da886822-618a-4c32-a332-18b0783e33dc",
      "params": {
        "strict_syntax": false,
        "symbols": {
          "lamda": {
            "latex": "\\lambda",
            "aliases": ["lambda", "lambda"]
          }
        }
      },
      "answer": "-4",
      "recommendedFeedback": "(1) The student's response differs by the term -."
    },
    {
      "submission": "3",
      "latexSubmission": "3",
      "simplifiedSubmission": "3",
      "frequency": 5,
      "feedback": "Incorrect",
      "mark": 0,
      "color": "",
      "responseAreaId": "655f95f1-c606-41a0-8905-a8baaa4a8a9f",
      "universalResponseAreaId": "da886822-618a-4c32-a332-18b0783e33dc",
      "params": {
        "strict_syntax": false,
        "symbols": {
          "lamda": {
            "latex": "\\lambda",
            "aliases": ["lambda", "lambda"]
          }
        }
      },
      "answer": "4"
    }]
}

Dependencies

• sympy==1.12.1
• ppython-Levenshtein==0.23.0
• numpy==1.26.1
• zss==1.2.0

Additional dependencies included in the docker file includes Karl's evaluationFunction and also a modified implementation of the compare.py from the zss library.