{"id":437,"date":"2017-01-04T16:36:40","date_gmt":"2017-01-04T16:36:40","guid":{"rendered":"http:\/\/www.wellformedness.com\/blog\/?p=437"},"modified":"2017-01-25T03:03:47","modified_gmt":"2017-01-25T03:03:47","slug":"classifying-paraphasias-with-nlp","status":"publish","type":"post","link":"https:\/\/www.wellformedness.com\/blog\/classifying-paraphasias-with-nlp\/","title":{"rendered":"Classifying paraphasias with NLP"},"content":{"rendered":"<p>I&#8217;m excited about <a href=\"http:\/\/ajslp.pubs.asha.org\/article.aspx?articleid=2594844\">our new article<\/a> in the American Journal of Speech-Language Pathology (with <a href=\"http:\/\/www.pdx.edu\/sphr\/gerasimos-fergadiotis-phd-ccc-slp-assistant-professor\">Gerasimos Fergadiotis<\/a>\u00a0and\u00a0<a href=\"http:\/\/www.bedrick.org\/\">Steven Bedrick<\/a>) on <a href=\"http:\/\/ajslp.pubs.asha.org\/article.aspx?articleid=2594844\">automatic classification of paraphasias\u00a0<\/a>using basic natural language processing techniques.<\/p>\n<p>Paraphasias are speech errors associated with aphasia. Roughly speaking, these errors may be phonologically similar\u00a0to the target (<em>dog<\/em>\u00a0for the target\u00a0<em>LOG<\/em>) or dissimilar.\u00a0They also may be semantically similar to the target (<em>dog<\/em>\u00a0for the target\u00a0<em>CAT<\/em>), or both (<em>rat<\/em> for the target\u00a0<em>CAT<\/em>). Finally, they may be neologisms (<em>tat<\/em> for the target\u00a0<em>CAT<\/em>). Finally, some paraphasias may be real words but neither phonologically nor semantically similar. The relative frequencies of these types of errors differ between people with aphasia. These can\u00a0be measured in a confrontation naming task and, with complex and time-consuming manual\u00a0error classification, used to create individualized profiles for treatment.<\/p>\n<p>In the paper, we take archival data from a confrontation naming task and attempt to automate the classification of paraphasias. To quantify phonological similarity, we automate a series of baroque rules. To quantify semantic similarity, we use a computational model of semantic similarity (namely cosine similarity with word2vec embeddings). And, to identify neologisms, we use frequency in the SUBTLEX-US corpus. The results suggest that test scoring can in fact be automated with performance close to that of human annotators. With advances in speech recognition, it may soon be possible to develop a fully-automated computer-adaptive confrontation naming task in the near future!<\/p>\n","protected":false},"excerpt":{"rendered":"<p>I&#8217;m excited about our new article in the American Journal of Speech-Language Pathology (with Gerasimos Fergadiotis\u00a0and\u00a0Steven Bedrick) on automatic classification of paraphasias\u00a0using basic natural language processing techniques. Paraphasias are speech errors associated with aphasia. Roughly speaking, these errors may be phonologically similar\u00a0to the target (dog\u00a0for the target\u00a0LOG) or dissimilar.\u00a0They also may be semantically similar to &hellip; <a href=\"https:\/\/www.wellformedness.com\/blog\/classifying-paraphasias-with-nlp\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;Classifying paraphasias with NLP&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"link","meta":{"_crdt_document":"","footnotes":""},"categories":[4],"tags":[],"class_list":["post-437","post","type-post","status-publish","format-link","hentry","category-language","post_format-post-format-link"],"_links":{"self":[{"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/posts\/437","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/comments?post=437"}],"version-history":[{"count":2,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/posts\/437\/revisions"}],"predecessor-version":[{"id":444,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/posts\/437\/revisions\/444"}],"wp:attachment":[{"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/media?parent=437"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/categories?post=437"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wellformedness.com\/blog\/wp-json\/wp\/v2\/tags?post=437"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}