Generalized isotonic conditional random fields

Yi Mao

doi:10.1007/s10994-009-5139-1

2009-08-18

Yi Mao *

Guy Lebanon

DOI：10.1007/s10994-009-5139-1

原文链接

原文求助

摘要

En 中文

Conditional random fields are one of the most popular structured prediction models. Nevertheless, the problem of incorporating domain knowledge into the model is poorly understood and remains an open issue. We explore a new approach for incorporating a particular form of domain knowledge through generalized isotonic constraints on the model parameters. The resulting approach has a clear probabilistic interpretation and efficient training procedures. We demonstrate the applicability of our framework with an experimental study on sentiment prediction and information extraction tasks.

Keyword:

Conditional random fields

Isotonic constraints

Prior elicitation

Sentiment prediction

Information extraction

期刊

Machine Learning

IF:

2.9

论文数:

2.7K

被引数:

3.4W

机构

university system of georgia

学者数:

7.3W

论文数: 6.6W

被引数: 101

引用论文

A TUTORIAL ON HIDDEN MARKOV-MODELS AND SELECTED APPLICATIONS IN SPEECH RECOGNITION 关于语音识别中的隐马尔可夫模型和选定应用的教程

PROCEEDINGS OF THE IEEE

IF25.9

1989-01-01

1.6W

PREAI

RABINER, LR

IF0

PREAI

Search for new supernova remnant shells in the Galactic plane with H.E.S.S.

Proceedings of The 34th International Cosmic Ray Conference — PoS(ICRC2015)

IF0

2016-08-18

PREAI

Gerd Puehlhofer; Francois Brun; Massimo Capasso; Ryan Chaves; Christoph Deil; Arache Djannati-Ataï; Axel Donath; Peter Eger; Daniel Gottschall; Helene Laffon; Vincent Marandon; Louise Oakes; Matthieu Renaud; Manami Sasaki; Regis Terrier; Jacco Vink; Aya Bamba

Physical adsorption of vapors on ice. III. Argon, nitrogen, and carbon monoxide

The Journal of Physical Chemistry

IF0

2002-05-01

PREAI

N. K. Nair; Arthur W. Adamson

Segmental stabilization and muscular strengthening in chronic low back pain - a comparative study

Clinics

IF0

2010-01-01