- Syntagmatic relationships in phonology: rhythm, weight, tone, harmony, phonotactics
- Phonological interfaces: particularly, with morphosyntax
- Quantitative and computational modeling, statistical machine learning
- Corpus methods: particularly, with under-resourced languages
- Psycholinguistics, cognitive science
- Poetic and musical forms
- Scientometrics for issues of diversity and representation
My research--and my lab's research--deploys computational approaches towards understanding the cognitive system of language. In particular, I focus on the place of phonology within the linguistic system: what is its structure, and what is the nature of its interfaces with other components of language (and beyond)? Computation and natural language data allow us to break open these questions by making language (and its use) a tractable and quantifiable object for scientific study.
Some of my current projects include
- Phonological influences on word and constituent order:
e.g., in Tagalog word order
- Learning of lexically-specific phonological behaviours:
e.g., unsupervised feature discovery for (morphological) class-conditioned phonology
- Phonotactic modeling of variable surface lexical tone patterns:
e.g., in Mende, Dioula, Hausa
- Phonological correspondences between music and language cross-linguistically:
e.g., in Japanese and English song text-setting
I am also interested in using scientometrics (quantative meta-study of science as a field) to turn a critical eye on (gender) diversity, representation, and the consequences of collaboration for scientific development in the field of linguistics.
- Shih & Zuraw. Under review. Phonological conditions on variable adjective-noun word order in Tagalog. Download PDF. Also available on lingbuzz/002796
- Shih. 2016. Super additive similarity in Dioula tone harmony. WCCFL Proceedings. Download PDF.
- Shih & Inkelas. 2016. Morphologically-conditioned tonotactics in multilevel Maximum Entropy grammar. AMP Proceedings. Download PDF.
- Shih. To appear. Phonological influences in syntactic choice. In The morphosyntax-phonology connection: locality and directionality at the interface. Download PDF.
- Shih & Inkelas. Under review. Autosegmental aims in surface optimizing phonology. Download PDF. Also available on lingbuzz/002520.
- Random Forests for Classification Trees and Categorical Dependent Variables: An informal quick start R guidelast updated: 2 February 2011
- Random Forests, for Model (and Predictor) SelectionThis is a more up-to-date introduction to random forests, from a guest lecture in May 2013. The accompanying files for the data can be found at Bruce Hayes' course website here.
- Summary of Basic Statistical Tests in R
- Cascadilla Proceedings Project Template for Microsoft Word (2010+)
- Agreement by Correspondence Conference Archive & Bibliography.
- Scientometrics of linguistics.
- A primer to navigating the academic job market.