sumya123/students-subject-preferences

Students' Subject Preferences

A small survey-style dataset recording which school subjects five students like and dislike. Each row is one student: their ID, the subjects they named as favorites, and the subjects they named as least favorites. Subject names are in Mongolian Cyrillic.

Files

File	Rows	Description
data/train.jsonl	5	One JSON object per student

Schema

Column	Type	Description
student_id	int	Student identifier, 1–5
favorite_subjects	list[string]	Subjects the student likes (up to 3, in the order given)
least_favorite_subjects	list[string]	Subjects the student dislikes (up to 3, in the order given)

Example row

{
  "student_id": 2,
  "favorite_subjects": ["Тамир", "Эрүүл мэнд", "Нийгэм"],
  "least_favorite_subjects": ["Монгол хэл", "Математик", "Уран зохиол"]
}

Loading

from datasets import load_dataset

ds = load_dataset("sumya123/students-subject-preferences")
print(ds["train"][0])

Notes on collection and normalisation

• The raw responses were written in Latin-script Mongolian; subject names here are normalised to standard Mongolian Cyrillic spellings. The mapping applied was: Angli hel → Англи хэл, niigem → Нийгэм, mat → Математик, pizik → Физик, gazarzui → Газарзүй, tamir → Тамир, Eruul mend → Эрүүл мэнд, Mongol hel → Монгол хэл, uran zoihol → Уран зохиол, tvvh → Түүх, himi → Хими, biologi → Биологи, undesni bichig → Үндэсний бичиг, hogjim → Хөгжим, oros hel → Орос хэл.
• ibo is kept verbatim: the respondent confirmed it is a subject name but did not give its canonical spelling, so it was not guessed.
• Every student named exactly 3 favorites and 3 least favorites in the source list; the order within each list is the order in which they were reported and is not a ranking.

Intended use

Reference data for school-subject preference questions, and a starting point for a larger survey collecting the same fields. With 5 rows it is illustrative, not statistically meaningful — no model should be trained or evaluated on it as if it were a real sample.

License

Released under CC BY 4.0. Change this if the underlying survey was subject to different terms.