Our company processes unstructured CV data to make it structured. Obvious if you're in the HR industry. We were sued by IBM for patent infringement on processing "unstructured CV data" 10 years ago, when we had prior art years before their patent. Just ridiculous and left a disdain towards IBM. With the advent of AI and machine learning, hopefully this type of trolling doesn't occur as often.
If anyone is interested in visualising graphs, I wrote an application called Graphia for this purpose. Its main use is in biological sciences, but it's designed to be general purpose.
This was very interesting to learn about the data structure itself, but is there a good source (preferably video, but the format doesn't matter too much) to learn about how a knowledge graph is populated in the first place? That's a pretty vital bit of information that I don't know.
About 30 years ago I used to setup a database system called Vineyard that came from Finland I think. It was all about building relationships between objects graphically. It let you discover interesting relationships and connections you couldn’t otherwise see.
This method seem to suck when you look something fringe stuff that has more popular "synonyms". In those cases it's not uncommon that Google completely ignores a word or two from the search to steer it toward something more popular.
I'm not quite sure this explanation works well enough without explaining a bit more about ontologies in the first place. I've been working with RDF and related stuff for quite a bit in an enterprise PLM context, and I wouldn't have seen a knowledge graph as an AI building block - but maybe my thinking is a bit off... I might also have mentioned a few of the upsides and downsides of practical work with a knowledge graph, but I guess this is a bit too specific for an intro video.
Just want to put in a plug for Obsidian Notes if anyone wants to play with building their own knowledge graph! Just started using it and it's really cool
How do companies like Google and others that have huge amount of this data make sure that they don't get two or more edges with different labels that reference the same data? For example, let's the underlying relationship is "location", but for LSE the edge is labelled "located in" and for KCL "location"? If that were the case they couldn't effectively query it anymore right.
I work in medical AI, but I’m not particularly familiar with knowledge graphs, and this explanation didn’t really clarify anything about what they are or how they work, or the difference between data vs. “knowledge” in this context. Just me? Anyone else just feel like we talked about Bushhouse for twelve minutes, and a very surface level concept of mathematical graphs?
just started playing around with Dgraph/GraphQL. Dgraph has an entirely free cloud starter if anybody wants to tinker. (Not for those unfamiliar with querying databases.)
I'm making a hobby app called humoredly that uses knowledge graphs to help write jokes and this video would've been a great help when I was starting out. My main beef against knowledge graphs though is that a lot of implementations use a limited number of relations. I've been meaning to look at models like REBEL that supposedly don't have this limitation I'm but not quite there yet
Although some Wikipedia Infoboxes draw data from Wikidata, most infobox content is not. In the case of Bush House none of it is from Wikidata (at the time of checking)
When you infer information by following a transitive relation would you decrease the certainty of the information at each step? Or limit the length of the inference chain to reduce the risk of it containing an untrue claim if you don't model uncertain information?
My issue with those systems has always been: what if things change over time? Most things do. - I wish to say X is a fact at time T, and then something may replace it at a later time. (the T is part of the record, as I always insert facts about past, not some arbitrary now) - And I would like to keep history. I have found terminusdb promising to do that, but it doesn't quite seem like what I was looking for... - Does anyone know good tools for that?
take a shot everytime she says Bush house
Our company processes unstructured CV data to make it structured. Obvious if you're in the HR industry. We were sued by IBM for patent infringement on processing "unstructured CV data" 10 years ago, when we had prior art years before their patent. Just ridiculous and left a disdain towards IBM. With the advent of AI and machine learning, hopefully this type of trolling doesn't occur as often.
If anyone is interested in visualising graphs, I wrote an application called Graphia for this purpose. Its main use is in biological sciences, but it's designed to be general purpose.
This was very interesting to learn about the data structure itself, but is there a good source (preferably video, but the format doesn't matter too much) to learn about how a knowledge graph is populated in the first place? That's a pretty vital bit of information that I don't know.
About 30 years ago I used to setup a database system called Vineyard that came from Finland I think. It was all about building relationships between objects graphically. It let you discover interesting relationships and connections you couldn’t otherwise see.
tell me more, I am interested, I have been thinking of such a system myself.
Great to see that kind of paper she drew on, still exists! Haven't seen it in decades. 😀
My god, I am in love with the way she says ‘knowledge graphs’.
This method seem to suck when you look something fringe stuff that has more popular "synonyms". In those cases it's not uncommon that Google completely ignores a word or two from the search to steer it toward something more popular.
I'm not quite sure this explanation works well enough without explaining a bit more about ontologies in the first place. I've been working with RDF and related stuff for quite a bit in an enterprise PLM context, and I wouldn't have seen a knowledge graph as an AI building block - but maybe my thinking is a bit off... I might also have mentioned a few of the upsides and downsides of practical work with a knowledge graph, but I guess this is a bit too specific for an intro video.
👍❗
Is there an open source code that creates knowledge graph 📉 out of a given data? No matter whether a text, pics or video or tables?
Just want to put in a plug for Obsidian Notes if anyone wants to play with building their own knowledge graph! Just started using it and it's really cool
as person who works with Tigergraph and Neo4j i like this video
How do companies like Google and others that have huge amount of this data make sure that they don't get two or more edges with different labels that reference the same data? For example, let's the underlying relationship is "location", but for LSE the edge is labelled "located in" and for KCL "location"? If that were the case they couldn't effectively query it anymore right.
would like to know whether this is also called a data model. if not what are the differences between a data model and a Knowledge Graph?
I work in medical AI, but I’m not particularly familiar with knowledge graphs, and this explanation didn’t really clarify anything about what they are or how they work, or the difference between data vs. “knowledge” in this context.
Just me? Anyone else just feel like we talked about Bushhouse for twelve minutes, and a very surface level concept of mathematical graphs?
At around 8 minutes onwards, she is drawing a knowledge graph. That seems pretty concrete to me.
data is unstructured, knowledge is a story being told by connected data. She specifically said structured information
I think there should be a follow up video which has more of an implementation side in it
EXACTLY WHAT I NEEDED!
Had been looking into ontology and KGs for the past few days for an NLP project and I am going for it :)
just started playing around with Dgraph/GraphQL. Dgraph has an entirely free cloud starter if anybody wants to tinker. (Not for those unfamiliar with querying databases.)
where do you still get those dot-matrix printer papers?
I'm making a hobby app called humoredly that uses knowledge graphs to help write jokes and this video would've been a great help when I was starting out. My main beef against knowledge graphs though is that a lot of implementations use a limited number of relations. I've been meaning to look at models like REBEL that supposedly don't have this limitation I'm but not quite there yet
Nicely explained and pleasant to listen to. Can you take this further for us? ♡
It's been always interesting to listen intellegence people talking about computer science
wikidata might be the most reliable source of data we have these days
Only for Hard Sciences
Love the subtle Python reference. Tip of the cap.
"This video gonna be viral"
- according to knowledge graph
This feels like it's an almost purely linguistic structure. I wonder if other languages have no concept of a "relationship"
Working with RDF right now!
that's correct 🍰🍰🍰🍰🍰
What do you think about RDF vs property graphs?
Although some Wikipedia Infoboxes draw data from Wikidata, most infobox content is not. In the case of Bush House none of it is from Wikidata (at the time of checking)
i think this is more like an ER-Diagram rather than a knowledge graph? are they the same ?
Thanks. How do reccomendation systems (Spotify yt) work?
I immediately wondered if my orgroam database is itself some kind of knowledge graph.
I'm doing research in semantic SLAM, this seems like really relevant concept
When you infer information by following a transitive relation would you decrease the certainty of the information at each step? Or limit the length of the inference chain to reduce the risk of it containing an untrue claim if you don't model uncertain information?
That should not be the case. If you say 1 < 2, 2 < 3 and 3 < 4, that doesn't mean you are less sure that 1 < 4.
My issue with those systems has always been: what if things change over time? Most things do.
- I wish to say X is a fact at time T, and then something may replace it at a later time. (the T is part of the record, as I always insert facts about past, not some arbitrary now)
- And I would like to keep history. I have found terminusdb promising to do that, but it doesn't quite seem like what I was looking for...
- Does anyone know good tools for that?
RDF 4ever!
Excellent explaination.
So the birth place of graphql! I think this is a great explanation. Thank you!
This is very intereesting stuff
thank you so much 🥰
This video doesn't provide much information... it just sort of talks about obvious surface stuff
Is her accent a mix of British with a touch of German? It’s so interesting. Not pure British. There’s a little something else in there.
how about knowledge gaps
"If A causes B & B causes C then A causes C" seems to be what we had on 8 bit confusers 40 years ago.
in English something like tag association
I tried listen for 50% of this video, and it provides absolutely zero knowlegde lol
Ten thousand years of social evolution and all I got was an AI recommendation.
Ontology
Star citizen
Yummy food for algorithm ;-)
First
But at what cost?
second
second