Skip to content

README.md

Latest commit

 

History

History
24 lines (13 loc) · 1.23 KB

README.md

File metadata and controls

24 lines (13 loc) · 1.23 KB

Overview and Citation

The herein-presented functions and software are described in the below-mentioned publication. Please cite accordingly:

Tobler, S. (2024) Smart Grading: A Generative AI-Based Tool for Knowledge-Grounded Answer Evaluation in Educational Assessments. MethodsX, 12. https://doi.org/10.1016/j.mex.2023.102531

Software and Code

No Maintenance Intended

Please note that the functions are no longer being updated. To implement the application, we recommend adjusting the GPT-related part of the code by using the GPT functions from the ellmer package. Link

⌨️ Functions

  • eval_single: This function allows running the evaluation function for an individual answer.
  • eval_combined: This function allows the evaluation function to run in a loop over all participants.

👾 Software

  • SmartGrading_Application: This script allows running the above-described functions in a web application environment. The RShiny application can be directly launched from R Studio.

📊 Code

  • Krippendorff.R: Includes the code for analyzing inter-rater reliabilities, for example, between SmartGrading and own results.