Skip to content

Latest commit

 

History

History
105 lines (72 loc) · 5.04 KB

README.md

File metadata and controls

105 lines (72 loc) · 5.04 KB
Raw

TagsAI

Flask (Python) website for Machine Learning Youtube #tags.

Info

This model is pretty slow. Need probably more than an hour per query to generate a raport.

I know ChatGPT, can generate tags much faster and have much bigger database, but even of small accuracy of this project, from my expirience this project generates a little bit better result in views for me than from ChatGPT. Accuracy is poor, because i dont have enaught information to work with. My data is really limited. Youtube studio have options to load clickrate, impressions etc. but this data is hidden from public.

Because of that I don't think that, there are bad results, considering that the "tags" is not most important indicator of video popularity.

I prefere to consider both models during choosing tags, because i dont know of how exactly Youtube alghoritm works. It is a pretty good tool to inspire yourself when selecting tags and learning from it.

Steps

  1. Program asks user for search query. Program will looks for youtube channels with similar topics. (Will add to channels list all thats uploaders info from videos that will appear after searching that text into youtube search in given pages number)
  2. Download video info (uploader name, views and list of tags). For each channel will be downloaded given number of videos info (or less, if channel doesnt have that many videos). This is taking the most of loading time, because for better analysis we recommend to download at least 1000 videos infos.
  3. Select channels. Select which channels will be considered in a training set.
  4. Preparing data.
  • Generate a boolean table (videos in rows and tags (with channel name in considiration because every channel have some sort of regular audiences) as columns).
  • Delete rows (videos) with to small amount of tags (rows with too small information to predict results)
  • Delete columns bearly used. (To small amout of videos uses this tag, so predicting result from this tags are uncertain)
  • Removing outliners points.
  1. Creating model In this project i use two models (For now. I planning to add more models later)
  • Standard Logistic Regression
  • Random Forest Classifier
  1. Cross-Validation (Accuracy) Program uses parallel multithreading LOOCV (Leave-One-Out Cross-Validation) and calculates MSE and accuracy (succesfully predicted views, by used tags).
  2. Suggested tags These models gives us the coefficient (or feature importances) numbers that we can use to predict tags that will give us the most of views. Sorted from the higher impact.
  3. Save your favorite tags Option for export your result for later.

Installing

Pyinstaller

I am building application for my releases, so all you need to do is to download them and run exe.

But if you want to build it on your own, then run following commands:

cd www
pyinstaller StandaloneVersion.spec

Via docker

Run following line

docker build -t tagsai:mestuq .
docker run -p 5000:5000 --name TagsAI tagsai:mestuq

Website is avalible in http://127.0.0.1:5000/ .

Via python script

Install required

pip install -r requirements.txt

Or use lock files avalible in "Lock files" folder.

pipenv shell
pipenv install --ignore-pipfile

Then run

python "www/main.py"

Website is avalible in http://127.0.0.1:5000/ .

Future plans

  • Currently im not considering "date of uploading" as a predictor.
  • Looking for faster method for downloading Youtube Data. (Currently using yt_dlp, because it is free), but planning Youtube API in the future.

Disclaimer:

This application is intended for personal and educational use only. By using this application, you agree to the following:

  1. Respect Copyright and Fair Use: You are responsible for ensuring that your use of this application complies with all applicable copyright laws and regulations. This application should not be used for infringing upon the rights of content creators or distributing copyrighted material without authorization.
  2. Compliance with YouTube's Terms of Service: You must adhere to YouTube's Terms of Service and any other policies or guidelines set forth by YouTube. This application is not affiliated with or endorsed by YouTube, and any use of this application to violate YouTube's terms is strictly prohibited.
  3. Responsible Use: Use this application responsibly and considerately. Avoid excessive or abusive downloading or scraping of content from YouTube, as it may lead to IP bans or other restrictions on your access to YouTube.
  4. No Warranty: This application is provided "as is," without any warranties or guarantees. The developers of this application are not responsible for any consequences resulting from your use of this application.
  5. Legal Consequences: Violation of copyright laws, YouTube's terms of service, or any applicable laws and regulations may have legal consequences. You are solely responsible for any legal actions that may arise from your use of this application.