Profile

Cover photo
John Pate
Works at Macquarie University
Attended University of Edinburgh
Lived in Middletown, OH
79 followers|11,909 views
AboutPostsPhotosVideos

Stream

 
I shook the dust off the old blog to talk about why there is a conflict between brevity and incrementality in natural language, and give some thoughts on grammaticalization

http://jkpate.net/random_words/2014/10/13/why-does-linguistic-structure-exist/
1
Add a comment...

John Pate

Shared publicly  - 
 
In which I have read some neuroscience: http://jkpate.net/random_words/2013/06/22/neural-legos/
1
Add a comment...

John Pate

Shared publicly  - 
 
interesting...
 

"The krumpets gnorked the koof with a shlap"

While this sentence may not make much sense, we bet you could infer quite a lot from its structure.  For example, perhaps you would be able to guess that group of something called a “krumpet” did something called "gnorking" to something called a "koof", and that they did so with a "shlap".

This is because sentences in languages such as English have structure. This structure is called syntax, and knowing the syntax of a sentence is a step towards understanding its meaning. The process of taking a sentence and transforming it into a syntactic structure is called parsing. At Google, we parse a lot of text every day, in order to better understand it and be able to provide better results and services in many of our products.

There are many kinds of syntactic representations (such as sentence diagramming, http://goo.gl/UxnsS), and at Google, we've been focused on a certain type of syntactic representation called "dependency trees". Dependency-trees representation is centered around words and the relations between them. Each word in a sentence can either modify or be modified by other words. The various modifications can be represented as a tree, in which each node is a word.

This property by which you could infer the structure of the sentence based on various hints, without knowing the actual meaning of the words, is very useful. For one, it suggests that a even computer could do a reasonable job at such an analysis, and indeed it can! While still not perfect, parsing algorithms these days can analyze sentences with impressive speed and accuracy. For instance, our parser correctly analyzes the made-up sentence at the beginning of this post.

Today, Google announces the release of a very large dataset of counted dependency tree fragments from the English Books Corpus. This resource will help researchers, among other things, to model the meaning of English words over time and create better natural-language analysis tools. The resource is based on information derived from a syntactic analysis of the text of millions of English books. 

To learn more, visit the Google Research Blog, linked below. 
4 comments on original post
1
Add a comment...

John Pate

Shared publicly  - 
1
Add a comment...

John Pate

Shared publicly  - 
 
Just like the previous post, but with pictures! jkpate.net/random_words/?p=103
1
Aciel Eshky's profile photo
 
John Pate's on fire, y'all.
Add a comment...

John Pate

Shared publicly  - 
 
Can I regularize my Bayesian posterior? Or is that like bringing a lightsaber to a Star Trek opening? #MLFauxPas
1
Brent Biglin's profile photoChris Brew's profile photo
3 comments
Add a comment...
In his circles
100 people
Have him in circles
79 people
Uzy Roo's profile photo
Zhunchen Luo's profile photo
Antonio Kabeluchi's profile photo
David Huby's profile photo
Hubert Wagner's profile photo
Ben Allison's profile photo
Kelley McClain's profile photo
Jan Avende's profile photo
Richard P-Man's profile photo

John Pate

Shared publicly  - 
 
I just got my VPN set up with Macquarie, and thought I'd share some networking stuff that has been useful to me in the recent past: http://jkpate.net/random_words/2013/06/23/useful-networking-stuff/
1
Add a comment...

John Pate

Shared publicly  - 
 
"Principles and Parameters and Manifolds, oh my!": In which I accuse Generativism of not caring about grammar. http://jkpate.net/random_words/2013/02/12/principles-and-parameters-and-manifolds-oh-my/
1
Add a comment...

John Pate

Shared publicly  - 
 
John K Pate. (2013) Predictability effects in language acquisition. PhD dissertation. Submitted, defense pending. 
9
Brent Biglin's profile photoSasa Petrovic's profile photo
2 comments
 
Congrats!
Add a comment...

John Pate

Shared publicly  - 
 
Bayesian modelling as the "right" way to do a computational-level statistical cognitive model
1
Add a comment...
People
In his circles
100 people
Have him in circles
79 people
Uzy Roo's profile photo
Zhunchen Luo's profile photo
Antonio Kabeluchi's profile photo
David Huby's profile photo
Hubert Wagner's profile photo
Ben Allison's profile photo
Kelley McClain's profile photo
Jan Avende's profile photo
Richard P-Man's profile photo
Work
Occupation
Computational psycholinguist
Employment
  • Macquarie University
    Postdoc, 2013 - present
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Previously
Middletown, OH - Columbus, OH - Edinburgh, UK
Story
Tagline
Computational Psycholinguist
Education
  • University of Edinburgh
    Informatics, 2009 - 2013
  • Ohio State University
    Linguistics, 2005 - 2009
Basic Information
Gender
Male