Overview

This repository contains supplementary materials for the following conference paper:

Valdemar Švábenský, Ryan S. Baker, Andrés Zambrano, Yishan Zou, and Stefan Slater.
Towards Generalizable Detection of Urgency of Discussion Forum Posts.
In Proceedings of the 16th International Conference on Educational Data Mining (EDM 2023).
https://educationaldatamining.org/EDM2023/proceedings/2023.EDM-short-papers.29/2023.EDM-short-papers.29.pdf

Contents of the repository

Data

• data-per-course: A folder with the data set divided per the 9 courses and split into two parts for the two coders.
• All_Courses_REDACTED_CODED.csv: The final training and cross-validation data set used for this work, resulting from merging all files from data-per-course into a single file. For details, please refer to Sections 3.1–3.3 of the paper.
• Stanford.csv: The test data set used for this work. For details, please refer to Section 3.6 of the paper. Citation of this data set:

Agrawal, A. and Paepcke, A., 2014. The Stanford MOOC Posts Dataset.
URL: http://infolab.stanford.edu/~paepcke/stanfordMOOCForumPostsSet.tar.gz.
Original page available on URL: https://web.archive.org/web/20220908024430/https://datastage.stanford.edu/StanfordMoocPosts/.

This version of the data is slightly different than what was used in the paper due to fixing a minor redaction error. Therefore, the results of running the analysis on this data will differ from the results reported in the paper.

Code

• main.py: Python code responsible for data pre-processing and model training. For details, please refer to Sections 3.4–3.5 of the paper.
• process-Stanford.py: Python code responsible for computing the numbers in Table 1 of the paper for the test set.

Other documents and files

• Coding-Instructions.pdf: Protocol used for training the coders of the data set.
• LICENSE: MIT license for the Python code.
• README.md: This file.

How to cite

If you use or build upon the materials, please use the BibTeX entry below to cite the original paper (not only this web link).

@inproceedings{Svabensky2023towards,
    author    = {\v{S}v\'{a}bensk\'{y}, Valdemar and Baker, Ryan S. and Zambrano, Andr\'{e}s and Zou, Yishan and Slater, Stefan},
    title     = {{Towards Generalizable Detection of Urgency of Discussion Forum Posts}},
    booktitle = {Proceedings of the 16th International Conference on Educational Data Mining},
    series    = {EDM '23},
    editor    = {Mingyu Feng and Tanja Käser and Partha Talukdar},
    location  = {Bengaluru, India},
    publisher = {International Educational Data Mining Society},
    month     = {07},
    year      = {2023},
    pages     = {302--309},
    numpages  = {8},
    isbn      = {978-1-7336736-4-8},
    url       = {https://educationaldatamining.org/EDM2023/proceedings/2023.EDM-short-papers.29/2023.EDM-short-papers.29.pdf},
    doi       = {10.5281/zenodo.8115790},
}