প্রতিক্ষণ কি ধরনের সমাস
Answers
Compound verbs (henceforth CV) are special type of complex predicates consisting of a
sequence of two or more verbs acting as a single verb and express a single expression of
meaning. However, not all verb sequences are considered as compound verbs. A compound verb
consists of a sequence of two verbs, V1 and V2 such that V1 is a common verb with /-e/ [non-
finite] inflection marker and V2 is a finite verb that indicates orientation or manner of the action
or process expressed by V1 (Dasgupta, 1977). The verb V1 is known as pole and V2 is called as
vector. For example, in the sentence রুটিগুলরো খেলে খপলরো (/ruTigulo kheYe phela/) “bread-plural-the
eat and drop-pres. Imp” “Eat the breads”, the verb sequence “kheYe phela” is an example of CV.
Identification of compound verbs from sentences is useful in many NLP applications including
Wordnet development, Information Retrieval, and Machine Translation. However, automatic
identification of compound verbs from a given text document is not a trivial task. As mentioned
in (Dasgupta, 1977), a sequence of two or more verbs does not always guarantees to be a
compound verb. Depending on the context a verb sequence may or may not act as CV. Thus,
automatic identification of compound verbs is extremely important and a challenging task.
This paper deals with the rule-based automatic identification of these types of Bangla CV,
where V1 is a pole and V2 is a vector. In our work (a) we shall propose rules through which a
system could automatically identify Bangla CVs from texts and these rules will be established on
the basis of syntactic interpretation of sentences, (b) we shall explain problems of CV
identification subject to the semantics and pragmatics of Bangla language, (c) finally we shall
make a statistical evaluation of our rules.
The rest of the paper is organized as follows: In section 2 we first perform the linguistic study
and the related concepts of the compound verb and different issues related to the automatic
extraction of CVs. Section 3 briefly discuss about the different related works done in this area.
Section 4 discuss about the different linguistic rules that can be applied to extract CVs from text
corpuses. Section 5 presents the experimentations, evaluations and results of our work.