Handling Ambiguities in Sindhi Named Entity Recognition (SNER)

Abstract

Natural language processing is considered the least advanced field of artificial intelligence and the main reason is the challenges in form of differences of accent, variations in script and other problems. A working Sindhi Named Entity recognition impose a challenging task, ambiguity; where the word can be a name of person and the day of the week. It is very hard to understand the difference between two meanings whereas the word is same in writing as well as in pronunciation. This research work presents the handling of ambiguities in Sindhi Entity Recognition (SNER). A rule based approach with the help of indexing has been used to handle ambiguities in SNER. The Sindhi Entity Recognition System recognizes ambiguous words of Sindhi language and produces corresponding meaning successfully. The SNER system clearly understands from the sentence and produces the accurate meaning corresponding to the ambiguous word and its available meaning.