@proceedings{DBLP:conf/propor/2024,
  editor       = {Igor Cataneo Silveira, André Barbosa and Denis Deratani Mauá},
  title        = {Computational Processing of the Portuguese Language - 16th International
                  Conference, {PROPOR} 2024, Galiza, March 13-15, 2024, Proceedings},
  series       = {Lecture Notes in Computer Science},
  volume       = {TODO},
  publisher    = {Springer},
  year         = {2024},
  url          = {TODO},
  doi          = {TODO},
  isbn         = {TODO},
  timestamp    = {TODO},
  biburl       = {TODO},
  bibsource    = {dblp computer science bibliography, https://dblp.org}
}

Data Structure

Features:
- id: id of scraped page. id_prompt+id should be unique
- id_prompt: Essay's theme
- essay_title: Essay title
- essay_text: Essay text
- grades: list of 6 elements containing the grade for each of the five concepts plus the sum of all grades
- essay_year: Essay's year
Number of Instances:
- sourceAOnly:
  - train: 227
  - validation: 68
  - test: 90
- sourceAWithGraders:
  - train: 744
  - validation: 195
  - test: 216
- sourceB:
  - full: 3219
Data Splits:
- sourceAOnly: sourceA data
- sourceAWithGraders: sourceA data augmented with Grader's review. In a nutshell, each row becomes three (the original grade plus two graders result)
- sourceB: sourceB data
- PROPOR2024: Same split used in PROPOR2024 paper. The others are updated and fix some tiny bugs (eg reproducilibity issues)