Clustering long sentence vector of dimension [768,]

I want to cluster vector sentences of dim [768,] and I don’t know whether should I use K-Mean or DBSCAN or something else approach. Or somehow I can use cosine similarity or WMD to cluster these vectors.
Please suggest me regarding this.
Thank you.

1 Like

It’s a matter of trial and error to choose what helps you.
Try all those clustering methods you said, visualize them and decide for yourself which gives you satisfactory results.

These days it’s just a matter of a single function call to do those visualizations, so you can just experiment it out quickly :slight_smile: