SDDNet: A semantic network for software design and development domain via graph database

Abstract

When building software systems, deciding implementation details manually for specific system requirements is time consuming and often leads to sub-optimal choices. An automated, knowledge-base driven approach to select from possible technology alternatives can replace or reduce manual effort in realizing these requirements. This motivated us to develop such a knowledge system which can ultimately help in opportune and cost-effective software development. We create a rich knowledgebase by automatic extraction of useful information from a large volume of existing documentation pertaining to software components and technologies that are used to build variety of business applications. We store the knowledge-base in a graph database. A semantic network depicting relations between concepts found in Software Design and Development (SDD) domain is constructed from the database. This knowledge-base can be queried for deducing additional facts about the concepts stored therein.