improve the resource matching

Potential approaches:

* [stable marriage problem](https://en.wikipedia.org/wiki/Stable_marriage_problem)
* Gale-Shapley algorithm - `O(n^2)`
* Ensures "there does not exist any match `(A, B)` by which both `A` and `B` would be individually better off than they are with the element to which they are currently matched"
* "not necessarily optimal from all individuals' points of view"
* https://rosettacode.org/wiki/Stable_marriage_problem#Go
* https://github.com/chasestarr/match
* [assignment problem](https://en.wikipedia.org/wiki/Assignment_problem)
* [Hungarian algorithm](https://en.wikipedia.org/wiki/Hungarian_algorithm) - `O(n^4)`
* Finds lowest overall cost
* https://godoc.org/github.com/cpmech/gosl/graph#Munkres
* [bilinear map](https://en.wikipedia.org/wiki/Bilinear_map)
* [Pearson correlation](https://en.wikipedia.org/wiki/Pearson_correlation_coefficient)
* Miller's Weil Pairing Algorithm
* all pairs similarity search
* https://twitter.com/aidanfeldman/status/932783372442701824