This paper demonstrates how a proposed parser identifies part of speech (POS) of Bengali lexicons in Bengali (or Bangla) sentences along with annotating semantics information. A parser is used in machine translation (MT) for identifying the part of speech in a sentence. We call it an intelligent parser as it handles semantics as well as POS identification. The aim of such rule-based intelligent Bangla parser is to ease the task of handling semantic issues in the subsequent stages in machine translation.