
Kevin Boyack
216 posts

Kevin Boyack
@KevinBoyack
Dad, grandpa, bibliometrics researcher, LDS
Katılım Mayıs 2020
98 Takip Edilen436 Takipçiler

@U_4GotHow2Play @BecomingCritter Yes, unpublished, the art that doesn't make it into the paper, stuff purposely held back, trade secret, etc. Still, in some cases there are clues, particularly in intelligence applications - authors that go dark, mismatch between topic and institution ...
English

@KevinBoyack @BecomingCritter Missing the the most important part of science... the unseen realm
English

@Jaydo2_0 @BecomingCritter It's not that there's no overlap, it's just not a primary overlap for either field.
English

@KevinBoyack @BecomingCritter Surprised about the lack of overlap between geology and chemistry
English

@sanadAlrashidi @BecomingCritter Short answer: citations. But, discipline-level maps are the wrong way to do it imo. With our paper-level maps (100M docs, 100k clusters), the cluster-cluster citation characteristics show the import/export patterns between topics. This is actionable.
English

@KevinBoyack @BecomingCritter I read this as a practical path to synthesis: a reference map that keeps disciplines distinct but comparable. If you had to pick one signal that best predicts real cross field transfer, what is it (citations, authorship, text similarity, something else)?
English

@bi02247255 @BecomingCritter Yes, indeed. I hear that particular phrase quite often!
English

@KevinBoyack @BecomingCritter All truth can be circumscribed into one great whole.
English

@CarlosL31742879 @BecomingCritter On the whole, from today's perspective, I'd say there is no 'central' science. Historically, one could argue philosophy. From a current enabling sense, math and chemistry are widely used in nearly all other fields, but that doesn't mean central imo.
English

@8BOXdev @BecomingCritter CE, ME, ChE, Mat are all related to each other through citation more than they are to other disciplines. So, if we carve all of science into 12 large categories, they'll be together. At the multihundred level, they are all separate.
English

@KevinBoyack @BecomingCritter My question is why are Mechanical, Civil, and Chemical Engineering all the same category? Is it because they're all connected by Materials Science?
English

@KuviacM @BecomingCritter Wasn't meant to be a binary answer. Yes, there are some citations, but equipment tends to be mentioned in the text rather than cited in references. And since we don't have full text of everything, we don't track that type of usage. Wish we could!
English

@KevinBoyack @BecomingCritter Confused. What do you mean neurological journals don't cite things in electrical engineering, like MRI machines, surgical robots, etc?
English

@alltopology @prismaticflow @AnnHertzz @BecomingCritter Yeah, funding is the dominant driver. Regarding career opps, it's ok. But for grad student choices, it doesn't work because (at least in the US), it's most often the professor that chooses the student, not the other way around.
English

@KevinBoyack @prismaticflow @AnnHertzz @BecomingCritter This is great for mapping career opportunities.
But to neo actual science, you must reconsider the weights on the nodes that are funded by mafias.
English

@jfarrow @BecomingCritter The overlay boxes are an interesting take, but I'd disagree. Most topics in the center box are concentrated in the CS part of paper-level maps. I've seen reasonable historical arguments that put philosophy at the center. Of course, this is a static map, so ...
English



@latterdaylaura Yes, wonderful! But I increasingly find myself going back to 3 Ne 9:13-22, and what he teaches them while it's still dark, and to the effect that this had on them. This is why they were at the temple for 3 Ne 11.
English

@jonnnnn234 @is_an_alt_man @BecomingCritter The undirected edge is actually the sum of the two directed edges. Also, it's not just a relationship, but the strength of the relationship because it's aggregated up from individual papers. So, it can be ranked/sorted. For structure, use of undirected edges is pretty common.
English

@is_an_alt_man @KevinBoyack @BecomingCritter the graph is not a directed one so I think it's impossible to do the sorting. The existence of the edge (undirected in this case ) only denotes there's a relation between 2 subject. what kind of relation do u want to build by creating directed edge in that kind of graph ?
English

@AnnHertzz @BecomingCritter This is an example of where journal-based analysis falls short. We abandoned journal-based maps after creating the one that started this thread. Instead, we do analysis based on papers, and those maps (90M papers, 100k clusters) show intersections such as those you suggest.
English

@KevinBoyack @BecomingCritter I'm surprised physics and chemical engineering are so far from molecular biology.
And brain studies from computer science.
I've noticed those trends in 2015 and they were new then but for some reason still surprising to me that there was so little citation overlap in 2005.
English

@Wendy45569083 I don't think it's a loss of context. Once a paper/method/etc. becomes standard knowledge, most people don't cite it any longer even though they use it. So, while knowledge is incorporated, citations tend to replace older works with newer ones.
English

True! One can trace the historical development of disciplines as you say. However, citation patterns today don't honor that history and they show a very different organization of the sciences. Not better, just different, answering different questions. Epistemology vs practice.
Stuart Clarkson@stuartfclarkson
@KevinBoyack @BecomingCritter It is not a perception. “Old academic notions” are actually based on the historical reality of the development of disciplines, including the emergence of science from the broader category of philosophy, through a web of decision making. You cannot simply disregard that reality.
English

@AnnHertzz @BecomingCritter It's not so much that it's not linked. Rather, it's incorporated. There are more math-related papers in non-math journals than in math journals. Math is an enabling science. You'll see this much more in a paper-based map than in a journal-based map.
English

@KevinBoyack @BecomingCritter This is so cool.
I have a question, what do you think are the implications of math being linked to so few others?
I'd naively expect math to be linked to everything as newer studies take advantage of math advancements...
English

And ... to answer the original question about the central science ... if the map is based on citations, then NO, there isn't one. But, if someone wants to use a different logic for ordering/linking disciplines, then MAYBE. Try it and share your findings.
critter@BecomingCritter
creator chimes in x.com/KevinBoyack/st…
English





