Code-switching is the linguistic phenomenon where a multilingual person alternates between two or more languages in a conversation, whether that be spoken or written. This thesis studies the automatic detection of code-switching occurring specifically between English and Spanish in two corpora.
Twitter and other social media sites have provided an abundance of linguistic data that is available to researchers to perform countless experiments. Collecting the data is fairly easy if a study is on monolingual text, but if a study requires code-switched data, this becomes a complication as APIs only accept one language as a parameter. This thesis focuses on identifying code-switching in both Twitter data and the Miami-Bangor corpus. This is done by conducting three different experiments. Our first experiment is a logistic regression model where we attempt to distinguish code-switched data from monolingual data. The second experiment is using a novel Word2Vec average nearest neighbor (WANN) classifier based on word embeddings to detect code-switching. The third experiment uses Doc2Vec, where the model uses the mean vector of each document to learn and distinguish between code-switched and monolingual data. Each of these experiments are performed twice, once with tweets and once with the Miami Bangor corpus. The results show that the WANN model performs best on Twitter data. The Doc2Vec model performs best on the Miami Bangor corpus. However, both approaches did well and the performances are comparable.
","abstract_html":"<p>Code-switching is the linguistic phenomenon where a multilingual person alternates between two or more languages in a conversation, whether that be spoken or written. This thesis studies the automatic detection of code-switching occurring specifically between English and Spanish in two corpora.</p> <p>Twitter and other social media sites have provided an abundance of linguistic data that is available to researchers to perform countless experiments. Collecting the data is fairly easy if a study is on monolingual text, but if a study requires code-switched data, this becomes a complication as APIs only accept one language as a parameter. This thesis focuses on identifying code-switching in both Twitter data and the Miami-Bangor corpus. This is done by conducting three different experiments. Our first experiment is a logistic regression model where we attempt to distinguish code-switched data from monolingual data. The second experiment is using a novel Word2Vec average nearest neighbor (WANN) classifier based on word embeddings to detect code-switching. The third experiment uses Doc2Vec, where the model uses the mean vector of each document to learn and distinguish between code-switched and monolingual data. Each of these experiments are performed twice, once with tweets and once with the Miami Bangor corpus. The results show that the WANN model performs best on Twitter data. The Doc2Vec model performs best on the Miami Bangor corpus. However, both approaches did well and the performances are comparable.</p>","abstract_has_math":false,"creators":["Polanco, Yohamy C"],"institution":"The Graduate School and University Center of The City University of New York","degree_name":"Master of Arts","degree_level":"Master","degree_discipline":"Linguistics","degree_department":null,"school":null,"contributors":[],"advisors":["Kyle Gorman"],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021-02-01T08:00:00Z","date_published":"2021-02-01T08:00:00Z","updated_at":"2026-07-24T01:59:14Z","subjects":["Computational Linguistics","Other Linguistics","Spanish Linguistics","translanguaging","computer mediated communication","Doc2Vec","Word2Vec","KNN"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://academicworks.cuny.edu/gc_etds/4195","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Kyle Gorman"]},{"key":"dc:creator","label":"Author","values":["Polanco, Yohamy C"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2021-01-29T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Linguistics"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Master"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Arts"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["The Graduate School and University Center of The City University of New York"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Computational Linguistics","Other Linguistics","Spanish Linguistics","translanguaging","computer mediated communication","Doc2Vec","Word2Vec","KNN"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://academicworks.cuny.edu/gc_etds/4195"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Code-switching is the linguistic phenomenon where a multilingual person alternates between two or more languages in a conversation, whether that be spoken or written. This thesis studies the automatic detection of code-switching occurring specifically between English and Spanish in two corpora.
Twitter and other social media sites have provided an abundance of linguistic data that is available to researchers to perform countless experiments. Collecting the data is fairly easy if a study is on monolingual text, but if a study requires code-switched data, this becomes a complication as APIs only accept one language as a parameter. This thesis focuses on identifying code-switching in both Twitter data and the Miami-Bangor corpus. This is done by conducting three different experiments. Our first experiment is a logistic regression model where we attempt to distinguish code-switched data from monolingual data. The second experiment is using a novel Word2Vec average nearest neighbor (WANN) classifier based on word embeddings to detect code-switching. The third experiment uses Doc2Vec, where the model uses the mean vector of each document to learn and distinguish between code-switched and monolingual data. Each of these experiments are performed twice, once with tweets and once with the Miami Bangor corpus. The results show that the WANN model performs best on Twitter data. The Doc2Vec model performs best on the Miami Bangor corpus. However, both approaches did well and the performances are comparable.
"]},{"key":"dc:title","label":"Title","values":["A Computational Study in the Detection of English–Spanish Code-Switches"]}]}],"canonical_facts":{"dc:contributor.advisor":["Kyle Gorman"],"dc:creator":["Polanco, Yohamy C"],"dc:date.available":["2021-01-29T08:00:00Z"],"dc:description.abstract":["Code-switching is the linguistic phenomenon where a multilingual person alternates between two or more languages in a conversation, whether that be spoken or written. This thesis studies the automatic detection of code-switching occurring specifically between English and Spanish in two corpora.
Twitter and other social media sites have provided an abundance of linguistic data that is available to researchers to perform countless experiments. Collecting the data is fairly easy if a study is on monolingual text, but if a study requires code-switched data, this becomes a complication as APIs only accept one language as a parameter. This thesis focuses on identifying code-switching in both Twitter data and the Miami-Bangor corpus. This is done by conducting three different experiments. Our first experiment is a logistic regression model where we attempt to distinguish code-switched data from monolingual data. The second experiment is using a novel Word2Vec average nearest neighbor (WANN) classifier based on word embeddings to detect code-switching. The third experiment uses Doc2Vec, where the model uses the mean vector of each document to learn and distinguish between code-switched and monolingual data. Each of these experiments are performed twice, once with tweets and once with the Miami Bangor corpus. The results show that the WANN model performs best on Twitter data. The Doc2Vec model performs best on the Miami Bangor corpus. However, both approaches did well and the performances are comparable.
"],"dc:identifier":["https://academicworks.cuny.edu/gc_etds/4195"],"dc:subject":["Computational Linguistics","Other Linguistics","Spanish Linguistics","translanguaging","computer mediated communication","Doc2Vec","Word2Vec","KNN"],"dc:title":["A Computational Study in the Detection of English–Spanish Code-Switches"],"thesis:degree_discipline":["Linguistics"],"thesis:degree_level":["Master"],"thesis:degree_name":["Master of Arts"],"thesis:institution_name":["The Graduate School and University Center of The City University of New York"]},"updated_at":"2026-07-24T01:59:14Z"}