Presentation 2012-06-29
Statistical natural language processing with continuous models
Daichi MOCHIHASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Statistical natural language processing so far mainly focused on combinatorial structure of languages, such as syntactic parsing and dependency analysis, based on a view of symbolic processing from an artificial intelligence research. However, when we delve into semantic space behind an enormous size of lexicon in natural languages, we are stuck to face their inherent continuous property that is originally inevitable for inducing discrete structures. With the latest advances in statistical natural language processing, we are now able to consider a joint model of languages with images, videos, sounds and even spatial and geographical locations and time-series data. They will eventually lay as a foundation of probabilistic robotics and cognitive studies of languages. This talk will first introduce some types of latest statistical topic models and overview the approaches to tackle with continuous data described above. The objective is to think over how we can combine the two different fields, natural language processing with pattern recognition and machine learning.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Statistical topic models / Gaussian processes / Boltzmann machine / MCMC
Paper # NLC2012-5,PRMU2012-25
Date of Issue

Conference Information
Committee NLC
Conference Date 2012/6/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Statistical natural language processing with continuous models
Sub Title (in English)
Keyword(1) Statistical topic models
Keyword(2) Gaussian processes
Keyword(3) Boltzmann machine
Keyword(4) MCMC
1st Author's Name Daichi MOCHIHASHI
1st Author's Affiliation Department of Mathematical Statistics, The Institute of Statistical Mathematics()
Date 2012-06-29
Paper # NLC2012-5,PRMU2012-25
Volume (vol) vol.112
Number (no) 110
Page pp.pp.-
#Pages 20
Date of Issue