Merge branch 'main' of https://github.com/epfl-ada/ada-2025-project-datahunt3rs

Author: advaithsriram

src/website/src/content/content.mdx

@@ -209,8 +209,10 @@ We moved beyond anecdotes and ran the numbers, performing a linear regression on
 
 The scale of this shift is staggering. In 2005, only 10.8% of videos leaned on sensational titles. By 2019, that number had climbed to 33.3%. That is more than just a slight uptick; it’s a 207.4% relative increase. To put it simply, a video uploaded in 2019 was three times more likely to use a clickbait title than one from the platform's infancy. Every year, the baseline for "normal" shifts higher, as creators fight to stay visible in an increasingly crowded room. The "quiet library" we mentioned earlier is slowly disappearing.
 
+Global trends only tell half the story. YouTube is actually a collection of distinct neighborhoods, each with its own unwritten rules. **Does a chemistry teacher really face the same pressure to bait as a Minecraft streamer?** We broke the ecosystem down into 15 categories to see exactly where the battle for attention is fiercest.
+
 <FadeInSection>
-## Categorical Divergence in Clickbait Adoption and Effectiveness
+## Not All Categories Behave the Same
 
 If the entire platform is screaming louder, is the screaming actually working? Or are some genres more immune to the "noise" than others? To find out, we looked into the YouNiverse dataset's 15 distinct categories: from Gaming and Comedy all the way to Education and Science. The following graph shows the various categories that are present and the number of videos that can be found in each of the categories: