Hello scholars et al.
I am perusing the idea of doing a #wikipedia wikiversity #onphd on that subject, but things are still fuzzy, since I can not pick up a good research question. It is much more software engineering than research and to my mind doesn't qualify as a #phd.
I hope to meet like minded people around!
@amz3 What progress have you made?
@jackpark I have a prototype that passed manual tests. I have an implementation of a permutation algorithm of n-tuples that allows to generalize triple store to n-tuple store while making it possible to have the same algorithm to execute queries whatever the dimensions of tuples. This allow me to factorize some code because...
@jackpark ... because I need a 3-tuple store for storing repository history (DAG), branch references and tags. I also need a 6-tuple to store changes items (graph, subject, predicate, object, alive, change identifier). And maybe I will also need a 4-tuples to store snapshots.
@jackpark Regarding querying for the next step, I will rely on minikanren (similar to core.logic). Based on that minikanren I will need to adapt existing work to support SPARQL OPTIONAL and UNION. I am missing a SPARQL query parser.
@jackpark I figured benchmarks I can do to compare existing solutions to mine. That's what I will work on this afternoon.
@jackpark I know it works, I need to benchmark and optimize.
@amz3 what's "personal knowledge base software"? I read a little of NLP but my background is on bayesian inference to estimate parameters (i just know the basics tho).
@seba_arroyo I think wikipedia definition is accurate "A personal knowledge base (PKB) is an electronic tool used to express, capture, and later retrieve the personal knowledge of an individual.". Zotero and wikis are kinds of PKB.
PKB is not a NLP task.
I think one can get more advantage of a PKB by applying NLP techniques like keyword extraction, key phrase, summaries etc...
you can think of my PKB as the product of a wiki, a search engine and ipython notebooks
@amz3 I see, It's an ambitius goal. very relevant too!
Here's my two cents: Ways of doing topic segmentation already exist. How about generating a concept map (graph) that relates those topics? I see concept maps and the like becoming increasingly usefull.
@seba_arroyo learning structure is very difficult apparantly, see https://scikit-learn.org/stable/tutorial/machine_learning_map/index.html
@amz3 you may want to check out writings and work by @enkiv2 -- see http://www.lord-enki.net/, particularly on Xanadu / ZZstructure (which he worked on for a number of years), anything mentioning hypertext, and for general perspective on future of software (book!). #ShortIntroductionsBecauseItsWayPastSleepTimeAndZzzz #AnywaySoundsNeatGoodLuck
Thad Starner (formerly of MIT, now at Georgia Tech last I heard) did graduate work involving integrating a personal knowledge base into a wearable computer in the late 1990s (iirc, in the form of an extension to EMACS org-mode to support something similar to the LEAP navigation in Jef Raskin's Swyft on his personal notes).
If I recall, he was also involved in the MIT research on replacing normal alarm reminders with subliminal alarm reminders (for appointments, etc.) in order to avoid hard task switching in the middle of important operations. Later, at GT, he worked on implicit procedural learning (specifically, teaching how to play piano with vibrating gloves worn 8+ hours a day).
Scholar Social is a microblogging platform for researchers, grad students, librarians, archivists, undergrads, academically inclined high schoolers, educators of all levels, journal editors, research assistants, professors, administrators—anyone involved in academia who is willing to engage with others respectfully.
We strive to be a safe space for queer people and other minorities in academia, recognizing that there can only be academic freedom where the existence and validity of interlocutors' identities is taken as axiomatic.
"An academic microblog that you can be proud to put on the last slide of a presentation at a conference"
(Participation is, of course, optional)
Scholar Social features a monthly "official" journal club, in which we try to read and comment on a paper of interest.
Any user of Scholar Social can suggest an article by sending the DOI by direct message to @email@example.com and one will be chosen by random lottery on the last day of the month. We ask that you only submit articles that are from *outside* your own field of study to try to ensure that the papers we read are accessible and interesting to non-experts.