Principal Data Scientist @ Productsup
Sometime back I wrote an article about detecting clickbaits (https://www.linkedin.com/pulse/identifying-clickbaits-using-machine-learning-abhishek-thakur).
The article received a very good response and also a lot of criticism.
Some people mentioned that I should have taken care of website content,
some wanted more samples from different sources and some wanted me to
try deep learning methods.
In this article, I will address all those issues and take clickbait detection to the next level.
No Free
After going through my own Facebook feed, I found out that clickbaits
cannot be classified just by looking at the title of the post.
Clickbait also depends on the content. If the content of the website is
very relevant to the title, it shouldn’t be classified as a clickbait.
However, it is very difficult to define what an actual clickbait is.
Let’s look at some of real posts on Facebook
In this article, I will address all those issues and take clickbait detection to the next level.
No Free Lunch Information
After going through my own Facebook feed, I found out that clickbaits
cannot be classified just by looking at the title of the post.
Clickbait also depends on the content. If the content of the website is
very relevant to the title, it shouldn’t be classified as a clickbait.
However, it is very difficult to define what an actual clickbait is.Let’s look at some of real posts on Facebook
No comments:
Post a Comment