When I would use a specific similarity coefficient over another?

Question

Like using Jaccards over Dice. I want real examples, of when I would prefer to use Jaccards, Dice, Cosine or any other similarity coefficient.

Answer 1

• Jaccard - measures similarity of assymetric, binary attributes. For example, if you have insurance claims with binary attributes ("poor driving record", "premium paid in cash") you can compare claims with those attributes.
• Cosine - measures similarity between vectors, like feature vectors. Could be used in a recommender system where a user asks to see items similar to some selected item.
• Dice - equivalent to F1-score. Often used in image segmentation, comparing a model's output with reference masks.