Also none of the references in my comment are true. The explosion in Akira is not a nuclear one. Without spoiling too much, in the manga it was Akira the caused it. None of what I said has any relevance to anything at all it's just my outrage expressed in very crassly worded sarcasm.
@EroSɘnnin it's bad encoding of characters not intended for web use (due to bad character mapping, most likely). They probably copied and pasted it from some document that has some archaic font type and we not technical enough to figure out
If you want to prove me wrong, go ahead. But it is in my opinion that such questions are beyond the reasonable scope of an average user of this site. Such questions need to be reigned in, in order not to innudate this site, unless the OP volunteers to take a crack at it themselves first
The users of this site has different goals and motivations than ones from code golf. Not to mention different interests in skillsets. If this was a site about data analytics or informatic I would agree, but its not.
My suggestion is having a list of criteria to allow general user a shot at answering or disproving the viability of such a question. Too much of the question is based upon unproven assumptions. This is not code golf this is anime and manga.
I believe that the scope of work to answer this question is unreasonable on top of the underlying reason not sufficiently provable without narrowing the scope.
I disagree. Why not try submit an abstract to demonstrate the feasibility of you scope. I believe the scope is too generalized as the OP does not specify any acceptable criteria to "most connected." Voice actor oftentimes give ties have times outside of productions such as part of idol groups of the sam production company. Sometimes these voice actors never work on a project together, so merely cross referencing the roles in MAL introduce bias. Futhermore MAL's DB and API isn't as easy to query and reference as you may thing. On paper it may be feasible, but in practice is another thing...
I am skeptical that it can be answered with a reasonable amount of effort or inferred by most people as you claim. But you're free to try to convince me otherwise. One has to establish an abstract of who should be mapped and why they fit the criteria. No reasonable bound are give besides they have to be Japanese voice actors. But realize there are Japanese voice actors that don't do anime and maybe exclusive to things like games. Ambiguously scoped questions can be as much trouble as identification requests if not kept to a reasonable bound.
Sure you can query the many various DB sites, but that alone is not an accurate gauge of how connected they are. They might know some ppl from voice acting classes, went to the same school, or just met by coincidence and hang out. These all may affect how connected they are. How does one reasonable expect someone to collect all of this info? It's a subject worth of a college level informatics thesis imo.
If it was how closely connected to Kevin Bacon, it's fine. But how does one start to find the Kevin Bacon of Japanese voice actors and go about probing that he is the Kevin Bacon of Japanese voice actors. I'm saying the amount of work this question requires in unreasonable.