Knowledge-based repository scheme for storing and retrieving business components: a theoretical design and an empirical analysis

Abstract

Component-based development (CDB) promises to reduce complexity and cost of software development and maintenance through reuse. For CBD to be successful, a vibrant market for commercial business components is essential. One of the key requirements of an active market for business components is an effective scheme for classifying and describing them at various levels of detail, as well as a corresponding repository for storing and retrieving these components. Such a scheme needs to support various constituents such as business users, managers, and application assemblers. The scheme and repository should help users and managers to select components that match their requirements and aid application assemblers in identifying components most compatible with their deployment environment (such as the platform) and system inputs (such as data types). Drawing from the concepts of group technology and software reuse paradigm, this paper proposes a scheme for classifying and describing business components and the design of a knowledge-based repository for their storage and retrieval. The proposed scheme is implemented in a prototype repository. The effectiveness of the prototype and the underlying classification and coding scheme is assessed empirically through controlled experiments. Results support the assertion that the scheme is effective in enhancing the users' and analysts' ability to find the needed business components.

An Experiment in Software Component Retrieval based on Metadata and Ontology Repository

“…Much domain knowledge is stored in a repository; knowledge is expressed by rules which show implicit relations [14], these rules are defined as follows: If A comes into existence, then B can be concluded, the confidence degree is CL, marked up A = (B, CL), hereinto, A is called antecedent which is combination of a series of conditions, i.e. A = A1 Λ A2 Λ A3……….…”

Open Source Collaboration for Fostering Off-The-Shelf Components Selection

“…The selection of OTS has been recognized as a critical process in the OTS-based development risk mitigation [5]. Regardless the specific properties of OSS and COTS components, we may consider that the high-level selection process (i.e.…”

Classificação semi-automática de componentes Java

“…Vitharana et al [91] propuseram um esquema efetivo de classificação para componentes de negócio considerando vários níveis de detalhe e abstração, além de um repositório para armazenar e recuperar tais componentes. Diferentes abordagens de classificação podem ter de ser feitas para os diferentes tipos de objetos no sistema.…”

“…Vitharana [91] considera o domínio de aplicação uma informação bem definida ou estruturada (um atributo dentro do modelo de classificação • Role: papéis que o componente pode (potencialmente) assumir uma variedade de aplicações.…”

“…De acordo com [91], uma linguagem de marcação como XML [9]é apropriada para codificar conhecimento baseado em facetas, pois possibilita a manutenção de esquemas para codificar uma base semi-estruturada de conhecimento do componente e dar suporteà busca de componentes.…”