Datasets:
metadata
license: cc0-1.0
task_categories:
- text-classification
- text-generation
language:
- en
tags:
- movies
- letterboxd
- web-scraping
- recommendation-systems
- cinema
- film-analysis
size_categories:
- 10K<n<100K
Letterboxd Movies Dataset
Dataset Description
A comprehensive dataset of movies scraped from Letterboxd, including genres, ratings, runtime, countries, and detailed movie characteristics.
This dataset contains 16246 movies with 28 features each, scraped from Letterboxd. It's perfect for:
- 🎬 Movie recommendation systems
- 📊 Film industry analysis
- 🤖 Machine learning projects
- 📈 Rating prediction models
- 🔍 Movie discovery algorithms
Dataset Structure
Features
| Column | Type | Description |
|---|---|---|
| title | object | Movie title |
| year | float64 | Movie year |
| decade | int64 | Movie decade |
| decade_category | object | Movie decade category |
| movie_era | object | Movie movie era |
| age_years | float64 | Movie age years |
| age_category | object | Movie age category |
| runtime | float64 | Movie runtime |
| runtime_category | object | Movie runtime category |
| runtime_percentile | object | Movie runtime percentile |
| is_feature_length | bool | Movie is feature length |
| genres | object | Movie genres |
| primary_genre | object | Movie primary genre |
| genre_count | float64 | Movie genre count |
| genre_diversity | object | Movie genre diversity |
| is_single_genre | bool | Movie is single genre |
| country | object | Movie country |
| country_category | object | Movie country category |
| language | object | Movie language |
| is_english | bool | Movie is english |
| title_length | int64 | Movie title length |
| title_complexity | object | Movie title complexity |
| word_count | int64 | Movie word count |
| title_runtime_ratio | float64 | Movie title runtime ratio |
| runtime_efficiency | object | Movie runtime efficiency |
| production_scale | object | Movie production scale |
| is_recent | bool | Movie is recent |
| is_classic | bool | Movie is classic |
Statistics
- Total Records: 16,246
- File Size: 3.41 MB
- Data Quality: 99.1% complete
- Year Range: 1888 - 2029
- Average Runtime: 108.5 minutes
Usage
from datasets import load_dataset
# Load the dataset
dataset = load_dataset("codealchemist01/letterboxd-movies")
# Access the data
df = dataset['train'].to_pandas()
print(df.head())
Data Collection
The data was collected through web scraping of Letterboxd.com using ethical scraping practices:
- Respectful rate limiting
- Robots.txt compliance
- No personal user data collected
Citation
If you use this dataset in your research, please cite:
@dataset{letterboxd_movies_2025,
title={Letterboxd Movies Dataset},
author={Kutay Ahin},
year={2025},
url={https://huggingface.co/datasets/codealchemist01/letterboxd-movies}
}
License
This dataset is released under the CC0 1.0 Universal License.