Grammatical Constraints on Intra-sentential Code-Switching: From Theories to Working Models
About
We make one of the first attempts to build working models for intra-sentential code-switching based on the Equivalence-Constraint (Poplack 1980) and Matrix-Language (Myers-Scotton 1993) theories. We conduct a detailed theoretical analysis, and a small-scale empirical study of the two models for Hindi-English CS. Our analyses show that the models are neither sound nor complete. Taking insights from the errors made by the models, we propose a new model that combines features of both the theories.
Gayatri Bhat, Monojit Choudhury, Kalika Bali• 2016
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Code-switching Real vs Fake Classification | All-CS (test) | Accuracy (Fake Class)97.83 | 10 | |
| Code-Switching Text Generation | All-CS Mono filter (test) | BERTScore63.3 | 10 | |
| Language Modeling | OpSub (test) | Test Perplexity55.56 | 8 | |
| Language Modeling | All-CS (test) | Test Perplexity276.6 | 8 | |
| Code-Switching Text Generation | All-CS (test) | Gzip (D)33.17 | 5 | |
| Human Evaluation of Code-Switching Generation | Human Evaluation Dataset 150 samples from 5 sources | Syntactic Score3.57 | 5 | |
| Code-Switching Text Generation | All-CS UNK filter (test) | BERTScore63.6 | 5 |
Showing 7 of 7 rows