A quantitative categorization of phonemic dialect features in context

Authors:
Naomi Nagy;Xiaoli Zhang;George Nagy;Edgar W. Schneider
Affiliations:
English Department, University of New Hampshire, Durham, NH;DocLab, ECSE, Rensselaer Polytechnic Institute, Troy, NY;DocLab, ECSE, Rensselaer Polytechnic Institute, Troy, NY;Department of English Linguistics, Regensburg University, Regensburg, Germany
Venue:
CONTEXT'05 Proceedings of the 5th international conference on Modeling and Using Context
Year:
2005

Citing 7
Cited 1

Algorithms for clustering data

Algorithms for clustering data
Fundamentals of speech recognition

Fundamentals of speech recognition
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Landscape of Clustering Algorithms

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 1 - Volume 01
Adaptive Clustering Ensembles

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 1 - Volume 01
Style Context with Second-Order Statistics

IEEE Transactions on Pattern Analysis and Machine Intelligence
Style Consistent Classification of Isogenous Patterns

IEEE Transactions on Pattern Analysis and Machine Intelligence

Interactive, mobile, distributed pattern recognition

ICIAP'05 Proceedings of the 13th international conference on Image Analysis and Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We test a method of clustering dialects of English according to patterns of shared phonological features. Previous linguistic research has generally considered phonological features as independent of each other, but context is important: rather than considering each phonological feature individually, we compare the patterns of shared features, or Mutual Information (MI). The dependence of one phonological feature on the others is quantified and exploited. The results of this method of categorizing 59 dialect varieties by 168 binary internal (pronunciation) features are compared to traditional groupings based on external features (e.g., ethnic, geographic). The MI and size of the groups are calculated for taxonomies at various levels of granularity and these groups are compared to other analyses of geographic and ethnic distribution. Applications that could be improved by using MI methods are suggested.