Font Size: a A A

Semantic-based Component Retrieval System Realization

Posted on:2011-07-17Degree:MasterType:Thesis
Country:ChinaCandidate:Z Q ZhangFull Text:PDF
GTID:2208330332977261Subject:Software engineering
Abstract/Summary:PDF Full Text Request
CBSD, i.e. Component-Based Software Development is a new technology for developing software application by means of assembling software components rather than writing all code from start scratch. This method reuses existing assets that works well in order to reduce efforts, also it raises productivity of software engineering. However, another problem raised, that is although there are many components in components base, it is difficult for programmer to retrieve. For the description and retrieve methods is too weak.Traditional components retrieval methods focus on keyword-based, signature method or facet-based, they often get low precision and recall due to domain knowledge always misses. In order to deal with this problem, many researches are proposed, ontology-based component retrieval method is the most promising one. This retrieval method use ontology to describe component formally, and take the semantic of user query and the knowledge of certain domain into account in all steps, which results in higher precision and recall. But, there is one major problem that is a huge discrepancy exists between nature language and formal semantic descriptions of components.In this paper, an interactive query method is proposed. This method allows users use natural language to describe what they want and system return components according to the description. It maps users'nature language to ontology by using concept identification process. Then extends queries by semantic constraints,and then query is revised according to retrieved results. Concept identification makes full use of user query context. The process of concept identification and query revision use conversational between system and users, which can extracts implicit semantics need of user, so that it leads to improve result of components retrieve, and reduce effort of retrieval process.
Keywords/Search Tags:software reuse, component retrieval, semantic, ontology, concept identification, conversation, interaction
PDF Full Text Request
Related items