Connectionist, Statistical and Symbolic Approaches to Learning for Natural Language Processing

Lecture notes in computer sciencePublished 1 January 1996

Stefan Wermter

DOI10.1007/3-540-60925-3

Source URL

Citations117

SJR quartileQ2

SJR score0.35

SNIP0.55

TL;DR

Embedded machine learning systems for natural language processing: Acquiring and updating hierarchical knowledge for machine translation based on a clustering technique and applying an existing machine learning algorithm to text categorization.

Abstract

This book is based on the workshop on New Approaches to Learning for Natural Language Processing, held in conjunction with the International Joint Conference on Artificial Intelligence, IJCAI'95, in M

Keywords

Computer Science

Lecture notes in computer scienceApplying an existing machine learning algorithm to text categorization

39 Citations1996Isabelle Moulinier, Jean‐Gabriel Ganascia

This paper describes how an existing similarity-based learning algorithm, Charade, is applied to the text categorization problem and compares the results with those obtained using decision tree construction algorithms.

Lecture notes in computer scienceApplying machine learning to anaphora resolution

39 Citations1996Chinatsu Aone, Scott Bennett

This work describes one approach to build an automatically trainable anaphora resolution system that employs the C4.5 decision tree algorithm by Quinlan and evaluates and compares the results of several variants of the machine learning-based approach with those of the existing anaphoras resolution system which uses manually-designed knowledge sources.

Lecture notes in computer scienceNatural language grammatical inference: A comparison of recurrent neural networks and machine learning methods

18 Citations1996Steve Lawrence, Sandiway Fong +1 more

It is found that the Elman and Williams & Zipser recurrent neural networks are able to find a representation for the grammar which is more parsimonious, and these models exhibit the best performance.

Lecture notes in computer scienceLearning approaches for natural language processing

12 Citations1996Stefan Wermter, Ellen Riloff +1 more

The purpose of this chapter is to provide an introduction to the field of connectionist, statistical and symbolic approaches to learning for natural language processing, based on the contributions in this book.

Lecture notes in computer scienceA dynamic approach to paradigm-driven analogy

8 Citations1996Stefano Federici, Vito Pirrelli +1 more

The general point is made that analogy-based self-learning techniques are no longer just psycholinguistically-plausible models, but competitive tools, combining the advantages of using language-independent, self- learning, tractable algorithms, with the welcome bonus of being more reliable for applications than traditional text-to-speech systems.

Lecture notes in computer scienceAcquiring and updating hierarchical knowledge for machine translation based on a clustering technique

5 Citations1996Takefumi Yamazaki, Michael J. Pazzani +1 more

Experimental results showing that the semantic hierarchies generated by the method yield learned translation rules with higher average accuracy are reported.

Lecture notes in computer scienceExtracting rules for grammar recognition from Cascade-2 networks

5 Citations1996Ross Hayward, Alan Tickle +1 more

A decompositional rule extraction technique which generates rules governing the firing of individual nodes in a feedforward neural network is described which employs heuristics to reduce the complexity in searching for rules which explain the behaviour of a neural network trained with sparsely coded data.

Lecture notes in computer scienceCan punctuation help learning?

3 Citations1996Miles Osborne

A model capturing some of the properties of punctuation is presented and a series of experiments demonstrate that even a simple model of punctuating increases the plausibility of learnt grammars over Grammars learnt without the use of punctuations.

Lecture notes in computer scienceA revision learner to acquire verb selection rules from human-made rules and examples

2 Citations1996Shigeo Kaneda, Hussein Almuallim +3 more

A learning method that automatically acquires English verb selection rules for machine translation using a machine learning technique with the most notable feature that any attribute-type learning algorithm can be adopted as the internal learner.

Lecture notes in computer scienceUsing parsed corpora for circumventing parsing

1 Citations1996Aravind K. Joshi, B. Srinivas

Corpora tagged with part-of-speech and phrase structure information have been used for both exploratory data analysis as well as unsupervised learning of language models.