Alan Patterson (EBay). They’ve been working on their Knowledge Graph for under a year. Biggest problem is identity (record linkage), depending on context.
#iswc2018 #iswc_conf
Jamie Taylor (Google). KG has 1 billion things, 70 billion facts. Not enough. Challenges: managing identity (Chicago Bull w MJ and w/o MJ are the same?) managing semantic stability (I.e. the meaning of sports has evolved)
#iswc2018 #iswc_conf
A reader with domain expertise and no engineering background will find only a subset of names self-evident. An experienced engineer with no domain knowledge will find a different subset to be self-evident.
Let's say you wanted to train a facial recognition algorithm on aging. What would do? Maybe start a meme like #10yearchallenge wired.com/story/facebook…
Numerical Linear Algebra for Programmers
aiprobook.com/numerical-line…
basically…
the only #AI book that walks the walk
100% executable code
step-by-step instructions
from theory to implementation
superfast implementation
#Nvidia #GPU #CUDA
#AMD #OpenCL
#Intel #JVM #MachineLearning
Bummed that Kindle on #iphone can’t render Arabic script in G. Willow Wilson book. Is this a problem with the format or the app? How can I help you support displaying the Arabic alphabet alongside other characters?
Today my company and I are excited to release qlkit, an open source clojure/clojurescript web development framework based on graph queries! Check out our walkthrough:
medium.com/@conrad_9565/l…
Proselint is a linter for prose—a command-line tool which analyzes text for redundancy, jargon, illogic, clichés, sexism, misspelling, inconsistency, misuse of symbols, malapropisms, oxymorons, security gaffes, hedging, apologizing, pretension, and more proselint.com