ArinRoths/movie-data

24-679 (Fall 2026): Movie Genre Descriptions

ArinRoths/movie-data

Short movie descriptions written for this 24-679 assignment at Carnegie Mellon University. Each original sample contains a movie description and its corresponding genre: Thriller, Action, Romance, Comedy, or Animation. This dataset supports a classroom comparison of fixed embeddings, fine-tuning, and few-shot prompting.

Source and task

The preparation notebook reads movie_descriptions.csv, retains the movie title, description, and genre, and converts each movie into a separate record. Labels come from the assigned movie genre: 0 = Thriller; 1 = Action; 2 = Romance; 3 = Comedy; 4 = Animation. The descriptions were written for this assignment based on the selected movies.

Preparation source: 24-679 Text Data notebook. Course: 24-679, Fall 2026, Carnegie Mellon University. Repository maintainer: the account shown above.

Fields

Field	Meaning and modeling role
movie	Movie title; identifying information, not intended as a model input.
description	Input movie-description text.
label	Classification target: 0 = Thriller; 1 = Action; 2 = Romance; 3 = Comedy; 4 = Animation.
genre, label_name	Readable target descriptions; exclude both from model inputs.
response_row	Original CSV row number; provenance only.
source_id, parent_id	Unique example key and its original description key; never input features.
augmentation, is_augmented	Method identifier and synthetic-row flag; provenance only.

Splits and original-source counts

These counts are computed from the packaged splits for this run.

Split	Original rows	Synthetic rows	Total rows
train	70	1120	1190
validation	15	0	15
test	15	0	15

Class counts include synthetic rows in training only.

Split	Label	Rows
train	0	238
train	1	238
train	3	238
train	4	238
train	5	238
validation	0	3
validation	1	3
validation	3	3
validation	4	3
validation	5	3
test	0	3
test	1	3
test	3	3
test	4	3
test	5	3

Requested holdout fraction: 30%; test receives 50% of that holdout. The first split uses seed 24679, and the holdout split uses seed 24680.

Split the original movie descriptions first, stratified by genre, before creating any augmented variants. Every generated variant stays in training with its parent. Validation and test contain only untouched original descriptions. Reuse these partitions for all model comparisons. Do not randomly re-split the augmented rows.

Augmentation and preprocessing

Character rate in this run: 30% of eligible words, capped at 3 words per description. The corpus contains 100 original movies. Each method generates up to 4 independently seeded variants per original training description.

Each method independently starts from an original training description; transformations are not stacked.

• Internal character swaps: swap adjacent, distinct interior letters in selected words.
• Internal character deletions: remove one interior letter from each selected word.
• Within-description word swap: exchange two distinct words within one movie description while preserving its word inventory.
• WordNet synonym replacement: use NLTK's English POS tagger and the first matching WordNet sense to replace one eligible base-form noun, verb, adjective, or adverb with a single-word alternative. Proper nouns, inflected forms, pronouns, and negation are protected or excluded.

The character methods apply the configured rate to eligible words of at least four letters, round up, and respect the per-description cap; first and last letters remain intact. A description with no eligible change is skipped. The WordNet first-sense rule is a heuristic, not contextual sense disambiguation.

Training method	Stored rows
internal_character_deletions	280
internal_character_swaps	280
none	70
within_description_word_swap	280
wordnet_synonym_replacement	280

Intended use and limitations

Use for teaching data preparation, augmentation audits, and movie genre classification. This is a small collection of selected movies and descriptions, not a representative sample of all movies or genres. Genre-specific words or plot details may make some labels easier to identify. English-oriented matching and WordNet may produce awkward or contextually incorrect replacements. Augmentation can change meaning or grammar even when its genre label is copied correctly. Inspect the variants; structural assertions do not prove semantic label preservation. Report test counts alongside accuracy/F1 and compare against original-only training.

Privacy and licensing

The prepared schema contains movie titles and descriptions but no dedicated personal information fields. The movie descriptions were written for this course assignment. Review applicable course permissions and copyright considerations before reuse. No license is assigned by this card.

Load and compare

from datasets import load_dataset
ds = load_dataset("ArinRoths/movie-data")
# Train with ds["train"], choose settings with ds["validation"], then score ds["test"].

AI Usage Disclosure

I used ChatGPT to help understand the assignment reqirements and how the code from last week would need to changed based on the data set I collected. This included learned some new functions and replacing others slightly. I also used it to debug my code.

Use an account with access if repository visibility changes. For reproducible comparisons, record the dataset commit and model/environment versions. Regenerate this card with the preparation notebook after changing the data; its counts are calculated from the actual packaged splits. The YAML schema and split configuration are preserved from the upload.