> any information which increases the likelihood of a claim being true

This sounds like "Bayesian evidence". However, I don't think "scientific evidence" has a precise agreed-upon definition. I would take "there is scientific evidence" to mean "some scientists say they have significant evidence that they themselves obtained directly and that they have published or presented".

And scientific evidence is also a very different bar depending on what sort of science you're doing. In medicine the gold standard is a double blind RCT but a geologist can't reasonably conduct RCTs with volcanic eruptions so they aren't required, but thankfully rocks are much simpler than biological systems. Or for psychologists p=.05 is enough for a study to be good evidence but particle physicists require much lower p values than that for something to be official scientific evidence.

It's clearly not good enough to pass a border, or to buy alcohol.

And here I have absolutely no idea if it is true or not. I don't know you, I can't see you. You could be a bot, a dog, a scammer, a teenager... Your evidence means nothing "evident".

If you could see me, could you be sure your eyes and ears were not lying to you? Could you be certain I was not a fragment of a paranoid delusion created by your subconscious?

Are you sure you're even awake right now? Are you sure that you've ever been awake? Are any of your senses evidently reliable?

Perceived reality can never be perfect evidence, and it's not possible to step outside of your own perceptions.

Evidence is just that. If you try really hard to achieve full epistemological rigour I think it's inevitable you'll wind up at the conclusion you don't know anything at all.