Pat Inc
Published in

Pat Inc

Data is not Meaning: Wikidata Examples

Data may look good from a distance, but unlike data, the science of meaning is still beautiful when you get close. (Image: Adobe Stock)

At my company, Pat Inc (PAT), we need to represent (language-independent) meaning, and its associated language (words and phrases) to use in conversation. Representation needs to be unambiguous internally, so its generation is always correct in the target language selected.

This is a much higher bar than is needed for most applications, but by focussing on human-like accuracy, a lot of benefits…

--

--

--

A scientific breakthrough in #ConversationalAI. Meaning-based NLU vs. Deep Learning Intent NLU. Sign up for early access: https://pat.ai/

Recommended from Medium

Tips for Maximizing the Benefits of Process Analytics

Minimizing the cost function: Gradient descent

What are we optimizing for?

Ensemble Learning: When everybody takes a guess…I guess!

EFUN makes it possible for the surveyors to learn about the opinions of the mass, for the public to…

Accelerate through Matched Data

How can we calculate the statistical power to test a correlation difference?

Making an Insurance Claims Prediction model with CatBoost in R

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
John Ball

John Ball

I'm a cognitive scientist working on NLU (Natural Language Understanding) systems based on RRG (Role and Reference Grammar). A mouthful, I know!

More from Medium

We Need to Talk About Synthetic Data

How Semantics Enables Super Knowledge Graphs, Part 1

How Neural Search is Being Used in Production

Transformer-based Sequential Denoising Auto-Encoder: Job2Vec