As of June 2022, more than *** hours of video were uploaded to YouTube every minute. This equates to approximately ****** hours of newly uploaded content per hour. The amount of content on YouTube has increased dramatically as consumer’s appetites for online video has grown. In fact, the number of video content hours uploaded every 60 seconds grew by around ** percent between 2014 and 2020. YouTube global users Online video is one of the most popular digital activities worldwide, with ** percent of internet users worldwide watching more than ** hours of online videos on a weekly basis in 2023. It was estimated that in 2023 YouTube would reach approximately *** million users worldwide. In 2022, the video platform was one of the leading media and entertainment brands worldwide, with a value of more than ** billion U.S. dollars. YouTube video content consumption The most viewed YouTube channels of all time have racked up billions of viewers, millions of subscribers and cover a wide variety of topics ranging from music to cosmetics. The YouTube channel owner with the most video views is Indian music label T-Series, which counted ****** billion lifetime views. Other popular YouTubers are gaming personalities such as PewDiePie, DanTDM and Markiplier.

Use our YouTube Videos dataset to extract detailed information from public videos and filter by video title, views, upload date, or likes. Data points include video URL, title, description, thumbnail, upload date, view count, like count, comment count, tags, and more. You can purchase the entire dataset or a customized subset, tailored to your needs. Popular use cases for this dataset include trend analysis, content performance tracking, brand monitoring, and influencer campaign optimization.

Throughout 2024, the majority of copyright claims received by YouTube were spotted by the platform's Content ID tool, which cross-checks uploaded videos against a larger file database. Over 2.2 billion claims were submitted via Copyright Match Tool, while approximately three million claims were submitted to the platform via webforms.

In 2024, users engaged more with the videos they watched on YouTube compared to the previous year. The number of average interactions on YouTube grew to 2.36 in the last measured year. This is an increase compared to 2023, when the number of comments, likes, and share on pieces of content hosted on YouTube was of approximately 2.1 interactions on average.

This dataset is designed to explore multistreaming social media video as a research method used to collect semi-structured interview data. The data are provided by Dr Karen E. Sutherland and Ms Krisztina Morris from the School of Business and Creative Industries at the University of the Sunshine Coast in Queensland, Australia. The dataset is drawn from the publicly available video recording of an interview undertaken as part of the research project called: ‘Like, Share, Follow’, a multistreaming show, featuring Dr Sutherland interviewing university graduates about their career journeys, that is broadcast across Facebook, LinkedIn, and Twitter and later uploaded to YouTube. This dataset examines how multistreaming video interview data can be used to answer research questions and the benefits and challenges this specific method of data collection can pose in the process of data analysis. The video example is accompanied by a teaching guide and a student guide.

Overview

This dataset contains engagement analytics from two prominent tech YouTube channels:

• WsCube Tech: A Hindi-language technology education channel covering web development, programming, and digital marketing.
• Google Developers: The official YouTube channel for Google’s developer ecosystem, featuring tech talks, product updates, and developer events.

Purpose

The purpose of this dataset is to analyze and compare the performance, engagement, and growth trends of both channels using metrics such as:

• VideoID
• Title
• UploadDate
• Views
• Likes
• Dislikes (Note: Not available via API since 2021)
• Comments

Possible Analysis

• 📈 Views Over Time: Growth and consistency of each channel
• 📊 Top Performing Videos: Views per video
• 🥧 Engagement Ratio: Pie chart showing likes vs comments distribution
• 📉 Upload Frequency Impact: How upload consistency affects views

Sources

Data collected using the YouTube Data API v3 between July 25–28, 2025. Only public video data is included.

Columns

Column	Description
VideoID	Unique ID of the video
Title	Title of the video
UploadDate	ISO format date of upload
Views	Total views (at time of extraction)
Likes	Number of likes
Dislikes	Not available (deprecated in YouTube API)
Comments	Number of comments

Licensing

Data is collected from publicly available sources (YouTube API). No copyrighted content is included.

Suggested Tags

• youtube
• YtDataSet
• Dashboard
• data-visualization
• wsCubeTech
• google-developers
• engagement-metrics
• social-media-analytics

In 2024, the engagement rate on YouTube content experienced a small decrease compared to the previous year. The average engagement rate on YouTube was of 3.87 percent in the last examined period, down from the 3.97 percent recorded in 2023.

About the Dataset : Youtube Views Prediction

This dataset contains information about trending YouTube videos from multiple countries, providing valuable insights for predicting video popularity based on various attributes. The dataset includes both numerical and categorical features that are essential for analyzing viewer behavior, engagement, and trends in content creation. The original source of this dataset can be found at : https://www.kaggle.com/datasets/datasnaek/youtube-new/data

Columns and Their Descriptions:

title: The title of the YouTube video.
channel_title: Name of the channel that published the video.
trending_date: The date the video started trending.
publish_date: The original upload date of the video.
publish_time: The exact time the video was published.
views: The total number of views the video received.
likes: The number of likes the video received.
dislikes: The number of dislikes the video received.
comment_count: The total number of comments on the video.
tags: Keywords or tags associated with the video, helping discoverability.
description: A detailed text description provided by the uploader.
category_id: The category assigned to the video (e.g., Music, Gaming, News).

Business Goals :

Predicting the number of views on youtube videos based on video attributes. The goal is to develop a model that can accurately predict the number of views a video will receive, using various video attributes such as likes, shares, comments, video duration, and more.

Business Metrics :

1. RMSE (Root Mean Squared Error) RMSE is a metric that measures the magnitude of the error between the values predicted by the model (Predicted Views) and the actual values (Actual Views). The lower the RMSE value, the more accurate the model's predictions.

2. R² (Coefficient of Determination) R² measures the extent to which the model can explain the variation in the data. R² values range from 0 to 1, where 1 means the model can explain all the variation in the number of views based on the given attributes, and 0 means the model cannot explain the variation. The higher the R², the better the model is at predicting views and the more relevant the features used in the model.

Model & Evaluation

The machine learning model was evaluated using several approaches, including different pre-processing techniques and multiple ML models. Ultimately, the chosen model for this analysis is the Random Forest Regressor. The final evaluation results show an RMSE of 630.741, indicating an average prediction error of approximately 630.741 units. Additionally, the R² score is 0.9623, meaning that the model explains 96.23% of the variance in the data (number of views). These results were deemed satisfactory and were selected as the final modeling approach for the system and its potential future applications.

📢 Published on : My LinkedIn

During the first quarter of 2024, Huge YouTube accounts, which had over 50,000 followers, reported an engagement rate of approximately 6.2 percent on their short-format content. In comparison, engagement was sensibly lower on long-format videos, which reported an engagement rate of 1.72 percent for Huge accounts. Medium YouTube accounts, which had a following between 2,001 and 10,000 users, reported engagement ratings of almost three percent on their Shorts, while long videos had an engagement of around 0.15 percent.

YouTube is an American online video sharing and social media platform headquartered in San Bruno, California. It was launched on February 14, 2005, by Steve Chen, Chad Hurley, and Jawed Karim. It is owned by Google, and is the second most visited website, after Google Search. YouTube has more than 2.5 billion monthly users who collectively watch more than one billion hours of videos each day. As of May 2019, videos were being uploaded at a rate of more than 500 hours of content per minute.

Youtube is very much used to influence, educate, free university (for me also) people (the users followers) in a particular way for a specific issue - which can impact the order in some ways.

In 2023, all the analyzed channels with an audience between 50,000 and 55 million subscribers had over 418,000 disliked on YouTube, against the approximately 17 million likes recorded in 2023. In comparison, all the tiny accounts analyzed - which had up to 500 subscribers - managed to accumulate a total of one million likes, as well as 53,600 dislikes and 41,430 comments.

Facebook received 73,390 user data requests from federal agencies and courts in the United States during the second half of 2023. The social network produced some user data in 88.84 percent of requests from U.S. federal authorities. The United States accounts for the largest share of Facebook user data requests worldwide.

Creative Commons YouTube

  Description

YouTube is large-scale video-sharing platform where users have the option of uploading content under a CC BY license. To collect high-quality speech-based textual content and combat the rampant license laundering on YouTube, we manually curated a set of over 2,000 YouTube channels that consistently release original openly licensed content containing speech. The resulting collection spans a wide range of genres, including lectures… See the full description on the dataset page: https://huggingface.co/datasets/common-pile/youtube.

Dataset with the most relevant data when evaluating the popularity/success of audiovisual content uploaded to the world-renowned Youtube platform.

• Information without pre-preprocessing and/or transforming.
• Info extracted through WebScraping techniques, obtaining a sample of the top 50 videos of "Youtube Trends".

Link: https://www.youtube.com/feed/trending?bp=6gQJRkVleHBsb3Jl

DataScience UOC Project

This dataset and supplementary tables are released in conjunction with the TPDL 2025 paper titled “How Retracted Research Persists on YouTube: Retraction Severity, Visibility, and Disclosure.” They provide detailed information used in the analysis to promote transparency, ensure reproducibility, and facilitate future studies on scholarly communication and retractions.

The dataset contains the following files:

Filename	Data Format	Description
01_dataset_scholarly_references_on_YouTube.json.gz	JSON Lines	An integrated dataset of scholarly references in YouTube video descriptions, covering videos posted up to the end of December 2023. This dataset combines the Altmetric dataset and the YA Domain Dataset and is the basis for identifying references to retracted articles. This dataset contains 743,529 scholarly references (386,628 unique DOIs) found in 322,521 YouTube videos uploaded by 77,974 channels.
02_dataset_references_to_retracted_articles_on_YouTube.json.gz	JSON Lines	A dataset of retracted articles referenced in YouTube videos, used as the primary source for analysis in this paper. The dataset was created by cross-referencing the integrated reference dataset with the Retraction Watch database. It includes metadata such as DOI, article title, retraction reason, and severity classification (Severe, Moderate, or Minor) based on Woo and Walsh (2024), along with video- and channel-level statistics (e.g., view counts and subscriber counts) retrieved via the YouTube Data API v3 as of April 22, 2025. This dataset contains 1,002 retracted articles (360 unique DOIs) found in 956 YouTube videos uploaded by 714 channels.
03_full_list_table3_sorted_by_reference_count_retracted_articles_on_YouTube.json.gz	JSON Lines	Complete list corresponding to Table 3, "Top 7 retracted articles ranked by the number of YouTube videos in which they are referenced." in the paper.
04_full_list_table5_top10_most-viewed_video.json.gz	JSON Lines	Complete list corresponding to Table 5, "Top 10 most-viewed YouTube videos that reference retracted articles, sorted by video view count." in the paper.
05_detailed_manual_coding_40_sampled_retracted_articles.xlsx	XLSX	This file provides detailed annotations for a manually coded sample of 40 YouTube videos referencing retracted scholarly articles. The sample includes 10 randomly selected videos from each of the four analytical groups categorized by publication timing (before/after retraction) and retraction severity (Moderate/Severe). The file includes reference stance for each video, visual/verbal mention of the article, and relevant timestamps when applicable. This dataset supplements the manual analysis results presented in Tables 6 and 7 in paper.

Due to concerns over potential misuse (e.g., identification or harassment of individual content creators), this dataset is not made publicly available.
Researchers who wish to use this dataset for scholarly purposes may contact the authors to request access.

References

• Woo, S., Walsh, J.P.: On the shoulders of fallen giants: What do references to retracted research tell us about citation behaviors? Quantitative Science Studies 5(1), 1–30 (2024). https://doi.org/10.1162/qss_a_00303
• Kikkawa, J., Takaku, M.: How Retracted Article Persists on YouTube: Retraction Severity, Visibility, and Disclosure. Accepted for publication in the Proceedings of the 29th International Conference on Theory and Practice of Digital Libraries (TPDL 2025).
• Accepted Papers (TPDL2025) - https://tpdl2025.github.io/Program/accepted_papers.html

Fundings

JSPS KAKENHI Grant Numbers JP22K18147 and JP23K11761.

English Audio Dataset from YouTube

This dataset contains English audio segments and creator uploaded transcripts (likely higher quality) extracted from various YouTube channels, along with corresponding transcript metadata. The data is intended for training automatic speech recognition (ASR) models.

  Data Source and Processing

The data was obtained through the following process:

Download: Audio (.m4a) and available English subtitles (.srt for en, en.j3PyPqV-e1s) were… See the full description on the dataset page: https://huggingface.co/datasets/OrcinusOrca/YouTube-English.

In December 2024, global traffic to YouTube.com came mainly from mobile devices. In comparison, 29.71 percent of the global traffic to YouTube.com came from desktop devices in the examined period.

This dataset was complied as a resource for analyzing viewer engagement, sentiment, and discussion trends on the Ben Shapiro YouTube channel over the specified period. It comprises user-generated comments extracted from the Ben Shapiro YouTube channel. The collection process involved first cataloging a comprehensive list of all videos published on the channel. Subsequently, these videos were categorized into three distinct time frames. From each time frame, the ten videos that garnered the highest number of comments were identified for detailed comment extraction. The extraction of videos and their associated comments was conducted utilizing YouTube Data Tools (Rieder, 2015). The dataset was finalized on September 12, 2022, and encompasses 711,909 comments ranging from September 1, 2020, to September 12, 2022. This dataset was uploaded and analyzed in the 4CAT: Capture & Anlysis Toolkit (Peeters & Hagen, 2022).

References:

Peeters, S., & Hagen, S. (2022). The 4CAT Capture and Analysis Toolkit: A Modular Tool for Transparent and Traceable Social Media Research. Computational Communication Research, 4(2), 571–589. https://doi.org/10.5117/CCR2022.2.007.HAGE

Rieder, B. (2015). YouTube Data Tools (1.11) [Computer software].

This provides a brief overview of the database. Further details are provided at: https://peterhcharlton.github.io/pwdb/ppwdb.html

Background: The shape of the arterial pulse wave (PW) is a rich source of information on cardiovascular (CV) health, since it is influenced by both the heart and the vasculature. Consequently, many algorithms have been proposed to estimate clinical parameters from PWs. However, it is difficult and costly to acquire comprehensive datasets with which to assess their performance. We are aiming to address this difficulty by creating a database of simulated PWs under a range of CV conditions, representative of a healthy population. The database provided here is an initial version which has already been used to gain some novel insights into haemodynamics.

Methods: Baseline PWs were simulated using 1D computational modelling. CV model parameters were varied across normal healthy ranges to simulate a sample of subjects for each age decade from 25 to 75 years. The model was extended to simulate photoplethysmographic (PPG) PWs at common measurement sites, in addition to the pressure (ABP), flow rate (Q), flow velocity (U) and diameter (D) PWs produced by the model.

Validation: The database was verified by comparing simulated PWs with in vivo PWs. Good agreement was observed, with age-related changes in blood pressure and wave morphology well reproduced.

Conclusion: This database is a valuable resource for development and pre-clinical assessment of PW analysis algorithms. It is particularly useful because it contains several types of PWs at multiple measurement sites, and the exact CV conditions which generated each PW are known.

Future work: However, there are two limitations: (i) the database does not exhibit the wide variation in cardiovascular properties observed across a population sample; and (ii) the methods used to model changes with age have been improved since creating this initial version. Therefore, we are currently creating a more comprehensive database which addresses these limitations.

Accompanying Presentation: This database was originally presented at the BioMedEng18 Conference. The presentation describing the methods for creating the database, and providing an introduction to the database, is available at: https://www.youtube.com/watch?v=X8aPZFs8c08 . The accompanying abstract is available here.

Accompanying Manual: Further information on how to use the PWDB datasets, including this preliminary dataset, are provided in the user manual. Further details on the contents of the dataset files are available here.

Citation: When using this dataset please cite this publication:

Charlton P.H. et al. Modelling arterial pulse wave propagation during healthy ageing, In World Congress of Biomechanics 2018, Dublin, Ireland, 2018.

Version History:

• v.1.0: Originally uploaded to PhysioNet. This is the version which was used in the accompanying presentation.

• v.2.0: The initial upload to this DOI. The database was curated using the PWDB Algorithms v.0.1.1. It differs slightly from the originally reported version in that: (i) the augmentation pressure and index were calculated at the aortic root rather than the carotid artery.

Text adapted from: Charlton P.H. et al., 'A database for the development of pulse wave analysis algorithms', BioMedEng18, London, 2018.

Paper: Barari, Soubhik, and Tyler Simko. "LocalView, a database of public meetings for the study of local politics and policy-making in the United States." Nature: Scientific Data 10.1 (2023): 135. Abstract: Despite the fundamental importance of American local governments for service provision in areas like education and public health, local policy-making remains difficult and expensive to study at scale due to a lack of centralized data. This article introduces LocalView , the largest existing dataset of real-time local government public meetings – the central policy-making process in local government. In sum, the dataset currently covers 139,616 videos and their corresponding textual and audio transcripts of local government meetings publicly uploaded to YouTube – the world’s largest public video-sharing website – from 1,012 places and 2,861 distinct governments across the United States between 2006-2022. The data are processed, downloaded, cleaned, and publicly disseminated (at localview.net) for analysis across places and over time. We validate this dataset using a variety of methods and demonstrate how it can be used to map local governments’ attention to policy areas of interest. Finally, we discuss how LocalView may be used by journalists, academics, and other users for understanding how local communities deliberate crucial policy questions on topics including climate change, public health, and immigration.