Supplementary data for the paper 'A multilingual metadata dataset of 90,189 explicitly labelled ASMR videos uploaded to YouTube between 2008 and 2026''
Wat erin zit
This article describes a dataset of public metadata for 90,189 videos uploaded to YouTube between 1 January 2008 and 31 August 2026 whose titles contain the term “ASMR” (autonomous sensory meridian response). The videos come from 18,126 channels and 81 language categories. Each record holds the video identifier, title, description, channel identifier and name, upload timestamp, duration, view and like counts, language label, and derived variables: views and likes per day relative to a fixed reference date, an engagement rate, a duration bucket, a growth category, title-style indicators and ten Boolean theme indicators from versioned, rule-based text patterns. Records were collected with the YouTube Data API v3 using consecutive three-month upload windows, supplemented by search through the pytubefix library, and were merged, filtered and enriched with a documented pipeline. The dataset is distributed as a gzip-compressed comma-separated table (51.8 MB), the complete source snapshot as JSON (40.1 MB compressed) with its schema, the list of video identifiers seen by the collector, a data dictionary, the theme rules, a reproducibility manifest with a SHA-256 digest of the source …
Bestanden en diensten
- DataCite REST API (JSON)api.datacite.orgOpenen