miket

10 ms·

31.
by miket 7y ago
Here is how a relation extraction system would represent these two concepts "live" (i.e. place of residence) and "visit" (a temporary location), as well as the temporality of the statements. http://relex.diffb
32.
by miket 7y ago
You can try a live demo of Diffbot's knowledge extraction from text here: http://relex.diffbot.com:8085
33.
by miket 7y ago
Obligatory XKCD: https://xkcd.com/2054/
34.
by miket 8y ago
We are trying to build a global knowledge graph from the web using AI at diffbot.com.
35.
by miket 8y ago
The main downfall of the Semantic web efforts are not technological, but due to a misalignment of incentives. Semantic web formats require content creators to annotate metadata for machines where webpages are intended for human readers. We
36.
by miket 8y ago
Interesting idea! Is there any way to customize what the user sees next in the video based upon how they answer a question?
37.
by miket 8y ago
Founder here. There are many differences in the result when you have an automated system building a Knowledge Graph vs. a human one. Obvious one is scale, Wikipedia has on order 10M entities and represents the work of thousands of humans wh
38.
by miket 8y ago
We've tried Neo4J, ArangoDB, as well as many others to store the triples. Neo4J locked up at around 100M entities, and also the loading/injection times weren't sufficient build the KG at a regular interval. However, we are cl
39.
by miket 8y ago
Founder here. The internet is full of contradictory statements, but that doesn't mean we can't attempt to form a useful model of reality. Humans do, just to get through daily life and AI can do the same as well. As you point out,
40.
by miket 8y ago
Founder here. OpenCyc (as well as Freebase) are human attempts to enter and curate a structured knowledge base. Likewise DBPedia is a set of scripts that extract Wikipedia infoboxes (semi-structured data which is also human crowd-sourced).
41.
by miket 8y ago
Interesting parallel between this result and the technique of "dropout" which is a key technique used today in training neural networks that has found to prevent the network from overfitting to a locally optimal solution by random
42.
A Universal Music Translation Network(research.fb.com)
2 points by miket 8y ago|0 comments
43.
Kremlin Cash Behind Billionaire’s Twitter and Facebook Investments(mobile.nytimes.com)
3 points by miket 9y ago|0 comments
44.
Sentiment aware input(github.com)
1 points by miket 9y ago|0 comments
45.
Burger King Just Launched Its Own Cryptocurrency, the Whoppercoin(adweek.com)
1 points by miket 9y ago|0 comments
46.
Putin: Leader in artificial intelligence will rule world(houstonchronicle.com)
13 points by miket 9y ago|2 comments
47.
Automatically Generating Commit Messages from Diffs Using NMT(arxiv.org)
2 points by miket 9y ago|0 comments
48.
Nvidia Tesla V100 GPU Architecture [pdf](images.nvidia.com)
1 points by miket 9y ago|0 comments
49.
by miket 9y ago
So, in other words, $25B worth of attention is transferred from iOS users to Google[1], which Google pays Apple $3B for this access. [1] https://www.recode.net/2017/7/24/16020330/google-digital-mob...
50.
Wikimedia Objective Revision Evaluation Service(mediawiki.org)
2 points by miket 9y ago|0 comments
51.
Realtime Multi-Person Pose Estimation(github.com)
2 points by miket 9y ago|0 comments
52.
Uber’s Ad-Toting Drones Are Heckling Drivers Stuck in Traffic(technologyreview.com)
6 points by miket 10y ago|3 comments
53.
Neural Network Zoo(asimovinstitute.org)
32 points by miket 10y ago|0 comments
54.
by miket 10y ago
A developer-friendly API to do this is Diffbot ( https://www.diffbot.com/ )
55.
by miket 10y ago
From the complaint: "Space Data started offering commercial wireless services in 2004 from a constellation of altitude controlled balloons which drift in the stratosphere. Since that time, it has extended the technology to service the
56.
by miket 10y ago
Check out http://sirius.clarity-lab.org/
57.
by miket 10y ago
That's a good question. One answer is that in each of those cases there are alternative mapping/news/music services. An open-source, privacy-first device could be an intermediary that wraps other services, and if it got sign
58.
by miket 10y ago
Was recently reported that Google receives 100K DMCA requests per hour: http://www.theverge.com/2016/3/7/11172516/google-takedown-re...
59.
Google Asked to Remove 100,000 “Pirate Links” Every Hour(torrentfreak.com)
1 points by miket 11y ago|0 comments
60.
by miket 11y ago
Exactly, you got it.

More ›