Algorithmic Classification of Five Characteristic Types of Paraphasias Purpose This study was intended to evaluate a series of algorithms developed to perform automatic classification of paraphasic errors (formal, semantic, mixed, neologistic, and unrelated errors). Method We analyzed 7,111 paraphasias from the Moss Aphasia Psycholinguistics Project Database (Mirman et al., 2010) and evaluated the classification accuracy of ... Research Article
Research Article  |   December 01, 2016
Algorithmic Classification of Five Characteristic Types of Paraphasias
 
Author Affiliations & Notes
  • Gerasimos Fergadiotis
    Portland State University, OR
  • Kyle Gorman
    Oregon Health and Sciences University, Portland
  • Steven Bedrick
    Oregon Health and Sciences University, Portland
  • Disclosure: The authors have declared that no competing interests existed at the time of publication.
    Disclosure: The authors have declared that no competing interests existed at the time of publication. ×
  • Correspondence to Gerasimos Fergadiotis: gfergadiotis@pdx.edu
  • Editor: Anastasia Raymer
    Editor: Anastasia Raymer×
  • Associate Editor: Neila Donovan
    Associate Editor: Neila Donovan×
Article Information
Language Disorders / Aphasia / Speech, Voice & Prosody / Supplement: Select Papers From the 45th Clinical Aphasiology Conference / Research Articles
Research Article   |   December 01, 2016
Algorithmic Classification of Five Characteristic Types of Paraphasias
American Journal of Speech-Language Pathology, December 2016, Vol. 25, S776-S787. doi:10.1044/2016_AJSLP-15-0147
History: Received September 16, 2015 , Revised March 9, 2016 , Accepted June 20, 2016
 
American Journal of Speech-Language Pathology, December 2016, Vol. 25, S776-S787. doi:10.1044/2016_AJSLP-15-0147
History: Received September 16, 2015; Revised March 9, 2016; Accepted June 20, 2016

Purpose This study was intended to evaluate a series of algorithms developed to perform automatic classification of paraphasic errors (formal, semantic, mixed, neologistic, and unrelated errors).

Method We analyzed 7,111 paraphasias from the Moss Aphasia Psycholinguistics Project Database (Mirman et al., 2010) and evaluated the classification accuracy of 3 automated tools. First, we used frequency norms from the SUBTLEXus database (Brysbaert & New, 2009) to differentiate nonword errors and real-word productions. Then we implemented a phonological-similarity algorithm to identify phonologically related real-word errors. Last, we assessed the performance of a semantic-similarity criterion that was based on word2vec (Mikolov, Yih, & Zweig, 2013).

Results Overall, the algorithmic classification replicated human scoring for the major categories of paraphasias studied with high accuracy. The tool that was based on the SUBTLEXus frequency norms was more than 97% accurate in making lexicality judgments. The phonological-similarity criterion was approximately 91% accurate, and the overall classification accuracy of the semantic classifier ranged from 86% to 90%.

Conclusion Overall, the results highlight the potential of tools from the field of natural language processing for the development of highly reliable, cost-effective diagnostic tools suitable for collecting high-quality measurement data for research and clinical purposes.

Order a Subscription
Pay Per View
Entire American Journal of Speech-Language Pathology content & archive
24-hour access
This Article
24-hour access