Part-Of-Speech Tagger (POS Tagger) is a tool that scans the text in specific language and allocates chunks of speech to individual word (and another token), such as verb, adjective, nown etc., as more fine-grained POS tags are used in computational applications like 'noun-plural'. Basically, the goal of a POS tagger is to allocate linguistic (mostly grammatical) information to sub-sentential units, called tokens as well as to words and symbols (e.g. punctuation). This paper presents a survey of POSĀ Tagger used for code-Mixed Indian and Foreign languages. Various methods, procedures, and features required to device POS Tagger for code-mixed foreign languages especially for Indian are studied and observations related to it are reported.
Copyrights © 2019