According to many experts, computers cannot process human languages. This is because understanding language means, knowing what concepts a word or phrase represents and knowing how to link those concepts together in a meaningful way. While natural language may be the easiest symbol system for people to learn and use, it is to be hard for a computer to understand and recognize the exact meaning of the sentence.
To address these problems, we are using a mix of knowledge-engineered and statistical/machine-learning techniques to disambiguate and respond to natural language input. The paper proposed by Mohd Ibrahim, Rodina Ahmad (Department of Software Engineering University of Malaya) on “Class diagram extraction from textual requirements using Natural language processing (NLP) techniques” [1] proposes a method and tool to facilitate requirements analysis process and class diagram extraction from textual requirements supporting natural language processing (NLP) technique. The algorithm “Requirement Analysis and Class Diagram Extraction (RACE)” analyze textual requirements, find core concepts and its relationships, and step by step extraction of the class diagram.
The Open NLP is an open-source and re-usable algorithm. It provides our system with lexical and syntactic parsers. Open NLP part of speech (POS) tagger (lexical) takes the English text as input and outputs the corresponding POS tags for each word; On the other hand, Open NLP Chunker (syntactic) chunks the sentence into phrases (Noun phrase, verb phrase, etc.) according to English language grammar. Open NLP uses lexical and syntactic annotations to denote to the part of speech of the terms. Stemming is a technique that abbreviates word by removing affixes and suffixes. Word Net is used to validate the semantic correctness of the sentences generated at the syntactic analysis.
RACE includes an interactive user interface (UI) that manages the tasks such as creating, printing, saving and analyzing requirements. It also handles the graphical representation of the class diagram. The system uses Open NLP parser, RACE stemming algorithm, and Word Net, to extract concepts related to the given requirements.
In addition to this we are generating deployment diagram using same concepts which are used to generate the class diagram.
Our system converts the requirement document given by client to class or deployment diagram which is helpful to Project Manager to get rough idea about the requirements of the client.
Technology:-
Java(netbeans)
0 comments:
Post a Comment