it's pretty hard to tell signal from noise when it comes to social media trends when trying to make short to mid-term inferences about anything.
We recently built semantic search around a (semi)real-time trends dataset we've been curating for some time now (~2 years). tldr, search queries are embedded and we perform a similarity search against our vectorized trends dataset, returning X most related trends.
There's been some interest in accessing the data directly so we built an API and some ui around this feature - we welcome you to give it a look or try and let us know what you think!
I think there could be many interesting applications in leveraging and applying this data - financial/market analysis factoring in real-time trends, AI personas that are "up-to-date" (I'm currently working on something like this), curating information/news feeds for your niche, etc. would love to hear more ideas if anyone would like to share.
you can try it out at https://trndgtr.com
leave a comment or feel free to email me directly at fred@trndgtr.com for any questions or thoughts :)