Where does PeARSearch make use of the Fruit Fly Optimization Algorithm

[Poslovitch]

While exploring the repos, I stumbled on this one: https://github.com/PeARSearch/PeARS-fruit-fly. After documenting myself on what the Fruit Fly Optimization Algorithm is (which was an interesting read btw!), I couldn't find where this algorithm was used in the various PeARS products. Was it just a proof of concept? (Considering the README has a few "todos" in it)

[minimalparts]

Aaaah... yes, you stumbled upon one of our pet projects (no pun intended). We have been trying for quite a while to generate Web document representations that would fit into a small binary hash and nicely preserve the semantics of the document. Achieving this would have all sorts of advantages, including obvious gains in speed and storage needed. So for a while, we generated thousands of virtual fruit flies with different properties and even made them breed with genetic algorithms.

Unfortunately, so far, we have not managed to find a version of the fruit fly that would work well across a variety of languages. So for the minute, the project is on pause. If you want to read more about our attempts, you can check out that published paper: https://aclanthology.org/2022.sustainlp-1.4/.