Algorithmic changes in YouTube by Digital Dnyan Academy

Views:
 
Category: Education
     
 

Presentation Description

How YouTube search algorithm works ?. How to does the recommended videos list is generated ?. What factors does YouTube considers for your uploaded video? https://www.digitaldnyan.com

Comments

Presentation Transcript

slide 1:

Algorithmic changes in YouTube By Omkar

slide 2:

YouTube is the world’s largest platform for creating sharing and discovering video content. YouTube recommendations are helping more than a billion users discover the most useful and required content from an ever-growing corpus of videos. Website : www.digitaldnyan.com Contact : 9284693861

slide 3:

YouTube has one of the most sophisticated recommendation systems in world wide web. The system uses deep learning to provide us with these results. YouTube algorithms consistency to provide relevant and personalized videos to their users is according to these three major perspectives They are as follows --- Website : www.digitaldnyan.com Contact : 9284693861

slide 4:

Scale: Many existing recommendation algorithms will work flawlessly on on an average scale but there arent useful for the scale of YouTube. Highly functional distributed learning algorithms and efficiency serving systems are very important for handling YouTube’s massive user base and corpus. Website : www.digitaldnyan.com Contact : 9284693861

slide 5:

Freshness: YouTube has a ever changing corpus with many hours of video are uploaded per second. The recommendation system should be responsive enough to consider new uploaded content as well as the recent actions taken by the user. Balancing new content with well established videos can be understood from an exploration/exploitation perspective. Website : www.digitaldnyan.com Contact : 9284693861

slide 6:

Noise: User behavior on YouTube is difficult to predict due to sparsity and a variety of unobservable external factors. We rarely obtain the ground reality of user satisfaction and use model noisy implicit feedback signals. Furthermore metadata associated with content is poorly structured without a well defined ontology. Our algorithms need to be powerful to these particular characteristics of our training data Website : www.digitaldnyan.com Contact : 9284693861

slide 7:

Using these sections YouTube helps user to find the most relevant videos for their searched query and the videos they will most likely enjoy. Six major places that impacts YouTube’s algorithm: search results suggested videos home page trending subscriptions and notifications. Website : www.digitaldnyan.com Contact : 9284693861

slide 8:

SO what are these Algorithmic changes Website : www.digitaldnyan.com Contact : 9284693861

slide 9:

Before YouTube used to prefer no of views of the particular video and now they consider time spend watching that particular videofor conclusions like: if the bounce rate of the video is high then its most likely a click bait The algorithm depends much on user data ie. has the user liked the video have they shared the video did they subscribed the channel considering all these factors YouTube prioritizes its recommended videos list Website : www.digitaldnyan.com Contact : 9284693861

slide 10:

The YouTube search now doesnt work on only keywords decided by the uploader to what context user is watching the video even if a keyword is not specific to a particular video still it might show up in the sidebar cause thats what many previous users watched in continuity. YouTube algorithm changes 300 times a year so almost every day hence there is no perfect way to SEO videos for an uploader. Website : www.digitaldnyan.com Contact : 9284693861

slide 11:

Digital Dnyan Academy Address Website : www.digitaldnyan.com Contact : 9284693861 Shop No.4 Kumar Millenium Beside Rajrashri Shahu Bank Shiv Thirth Nagar Paud Raod Kothrud Pune. 411038 Email : Infodigitaldnyan.com Contact : 9284693861

slide 12:

THANKS FOR WATCHING Website : www.digitaldnyan.com Contact : 9284693861

authorStream Live Help