I created two clustering using k-means clustering. However, two silhouette coefficients are judged to be low. The average silhouette coefficient was about 0.2, the silhouette coefficient for Group A was 0.4, and the silhouette coefficient for Group B was 0.04. Is there any way to improve this or any part of the data that needs to be modified? I have less than 2,000 data.
From what I have searched, it seems to be a matter of data preprocessing. In my case, I used TfidfVector and applied it to Kmeans algorithm, but may it lack classification criteria or processing power?I need some advice.