Pete Rivett

#Ontologist, @EKG_Foundation #EKGF #ontology #SemanticWeb #Semtech #LinkedData #FIBO #RDF #OWL #SPARQL
Also @rivettp.bsky.social, @[email protected]

[Pete Rivett @rivettp]

@yoheinakajima Your first sentence seems misstated when you say " when I'm listening". If you're already formulating a reply, which is indeed a common habit, you're clearly not truly listening. And listening is more than words, but how they're said, facial expressions, hand gestures and more

[Pete Rivett @rivettp]

@donwinslow Merry Christmas Mr Lawrence

[Pete Rivett @rivettp]

@AUMichaelBrown Do you still own and run this site?

[Pete Rivett @rivettp]

@matchaman11 @langchain @hwchase17 Langchain

[Anil Chandra Naidu Matcha @matchaman11]

@langchain @hwchase17 If you are interested in gaining early access

1. Like
2. Retweet
3. Comment "Langchain" and I will send you a dm

[Pete Rivett @rivettp]

@carlvellotti 60

[Carl Vellotti 🥞 @carlvellotti]

Stop wasting time with random PM content.

I've spent >100 hrs curating:
🔹 60 all-time best resources
🔹 Sequencing them for impact
🔹 Writing q's for maximum learning

Soon, I'll make this a course for $149.

For the next 24h: FREE!

Follow + RT + comment "60" and I'll DM.

[Pete Rivett @rivettp]

@valexiev1 @ontotext @PlantUML @buildingSMARTIn My comments were aimed at depictions of ontologies generally. I see your start point was the more restricted SOML, which, having dug into it a bit, is closer to an object model than an ontology and hence for which UML makes more sense.

[Pete Rivett @rivettp]

@valexiev1 @ontotext @PlantUML @buildingSMARTIn Another issue with the UML notation is that in UML the lines represent constraints on associations, whereas in RDFS/OWL domain and range represent inferences. So the UML interpretation is misleading.

[Pete Rivett @rivettp]

@reachvaldo ✅

[Pete Rivett @rivettp]

@prithjourney SEND

[Pete Rivett @rivettp]

@WaqarSaaS AI

[Pete Rivett @rivettp]

Our Golden Retriever https://t.co/oU5gXQvbWJ, enhancing ChatGPT with the power of the open Golden Knowledge Graph, gets the same results as Bloomberg for these queries

[elvis @omarsar0]

Here are some examples involving financial question answering (specifically recalling names of CEOs of companies) and how it compares to other models like FLAN-T5-XXL and GPT-NeoX. (using 10-shot setting)

Overall, it seems BloombergGPT is beneficial for tasks in the financial domain while not sacrificing performance on general-purpose tasks.

Definitely worth a read for more technical details and results.

[Jude Gomila @judegomila]

1/ I’d like to share a fun little demo we have been hacking on @golden

We attempt to answer factual questions in a more accurate way using the Golden Knowledge Graph by providing a retrieval enhancement to GPT-3.

Golden Retriever: https://t.co/X6Xwa0QMfe

Thread👇

[Pete Rivett @rivettp]

@veronikaheim Most visualizations (and indeed editors) are OO or ER variations and don't sensibly depict object properties with super-properties and no domain or range.

[Pete Rivett @rivettp]

I’m excited to announce that I’ve joined @Golden as their #Ontologist.
Looking forward to bringing together ideas, technologies and communities from #Web3 and the original Web 3.0 #SemanticWeb into an open distributed graph of canonical knowledge https://t.co/wircBADcJ1

[Pete Rivett @rivettp]

@hhedden At the risk of being trite, looks like you need a taxonomy of taxonomies.

[Pete Rivett @rivettp]

@lechatpito Interesting, but what you asked for is IMHO more like a degraded relational model (which is what it gave you) than an ontology. Surprised though that it doesn't seem to understand location and address.