Thousands of languages around the world are endangered, placing unique histories, traditions, and cultural identities at risk of disappearing. In this project, I explored whether natural language processing techniques could help reconstruct missing vocabulary in low-resource languages. Inspired by challenges facing endangered Dravidian languages, I investigated how machine translation and computational linguistics can leverage related languages to generate new words and support language revitalization efforts. This work highlights how artificial intelligence can contribute to preserving cultural heritage in the digital age.

Each year, nearly 800,000 people in the US suffer strokes. Of these, about 38%, or 300,000, experience some degree of aphasia, or loss of linguistic abilities … Regardless of symptoms, many aphasic patients show marked improvement over time, with some studies reporting up to 40% of patients recovering completely within a year of the stroke … In this project, I analyzed the differences in word use between well-recovered aphasic patients (those that perform well on the Western Aphasia Battery) and normal control subjects on a discourse task. Although the aphasic patients exhibited near-normal performance on the word and sentence levels, they produced different patterns of text structure and word use than normal subjects. This project introduced a methodology for statistically analyzing these differences in word use. In the future, a similar approach could be used to develop a diagnostic tool to identify patients with discourse impairments, based on analysis of the words used in a short transcript of speech …

Both French and Spanish have masculine v. feminine grammatical gender, while English does not have grammatical gender. For example, the phrase ’the table’ in English has the gender-neutral ’the’ preceding the word ’table’, while in French and Spanish, the feminine article ’la’ precedes both ‘mesa’ in Spanish and ’table’ in French . . . These grammatical differences between the languages led me to wonder whether there is a difference in [gender] attitudes among countries were Spanish, French, and English are predominantly spoken (Spain, France, US).