Some nominalists dispense with substantive universals (see Platonism) in treating the one over many principle by saying that what unites a group of objects of the same kind is that they resemble one of their number taken as a standard.
Objections to this are that resemblance itself seems to be an eliminable liversal, and so does the respect in which objects resemble the standard one (that is, resemblance seems to presuppose rather than explain universals).
Ludwig Wittgenstein (1889-1951) in his Philosophical Investigations (1953) dispensed with standard instances by taking a family resemblance view, where objects in the group have nothing in common to all of them, but any one of them has much in common with a large number of the others (the vagueness is deliberate), even though members on opposite sides of the central area of the 'family' may have nothing relevant in common.
The cluster theory is somewhat similar.
M A Simon, 'When is a Resemblance a Family Resemblance?', Mind (1969);
D Gasking, 'Clusters', Australasian Journal of Philosophy (1969)