A Fast, Compact, Accurate Model for Language Identification of Codemixed Text
This addresses the problem of accurately and quickly identifying languages in mixed-language text for applications like social media and online documents, representing a strong specific gain.
The paper tackled fine-grained multilingual language identification for codemixed text, achieving a 19.5% averaged absolute gain in accuracy and an 800x speed-up compared to previous methods.
We address fine-grained multilingual language identification: providing a language code for every token in a sentence, including codemixed text containing multiple languages. Such text is prevalent online, in documents, social media, and message boards. We show that a feed-forward network with a simple globally constrained decoder can accurately and rapidly label both codemixed and monolingual text in 100 languages and 100 language pairs. This model outperforms previously published multilingual approaches in terms of both accuracy and speed, yielding an 800x speed-up and a 19.5% averaged absolute gain on three codemixed datasets. It furthermore outperforms several benchmark systems on monolingual language identification.