Sign in to view source links and access this dataset
Description
209,543 TikTok videos from 1,872 creators posted between June and November 2024. The dataset includes daily engagement and follower statistics, with derived content fields such as video summaries, topic labels, and emotion scores generated from audio transcripts, screenshots, and music metadata using machine learning models. It was authored by lingbow and last updated on the Hugging Face platform in April 2026.
Use Cases
Modeling video popularity based on daily engagement and follower statistics mentioned in the description
Analyzing content trends using derived topic labels from video summaries
Studying creator-audience interaction patterns from the time-series engagement data
Training multimodal classifiers for emotion detection based on derived emotion scores
Strengths
Contains 209,543 video records, providing a substantial sample size for analysis
Covers a specific time range from 2024-06-24 to 2024-11-09, enabling temporal studies
Includes derived content features like topic labels and emotion scores, adding analytical depth
Limitations
Column-level documentation is absent; field semantics must be inferred after download
The dataset's geographic and demographic scope is unspecified, which may limit generalizability
Last updated 2026-04-29 20:29:44; freshness should be verified
Provenance
Source
Hugging Face dataset authored by lingbow
Collection Method
Data collection from TikTok, with derived fields generated from audio transcripts, screenshots, and music metadata using ML models
Time Range
2024-06-24 to 2024-11-09
Freshness
Last updated 2026-04-29 20:29:44
License information is unknown; users should verify permissions before use.