Publicação

Schema evolution and change coupling in MediaWiki

Ver documento

Detalhes bibliográficos
Resumo:Software developers face many adversities while working on projects, one of the most important ones being schema evolution. This is an inevitable procedure that, when not given the proper attention, can render an whole application unusable. The best option to study the impact of schema evolution on software development was the mining of data in a large database application. The chosen case study was MediaWiki. Using existing tools to mine the les of the MediaWiki project, the aim was to nd which les caused and which were a ected by schema evolution, and to extract frequent patterns. Building on existing work on mining data schema updates in each revision of MediaWiki, it was possible to extract over a hundred di erent association rules. These rules allowed the development of a recommendation system. This system allows to provide two les working on a database schema and receive as an output possible missing updates that should be applied to the newer schema. This will help software developers to keep their database schemas coherent and concise.
Autores principais:Silva, Ricardo Jorge Ferreira da
Assunto:Schema Evolution Databases Data Mining Frequent Patterns Association Rules
Ano:2019
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Aveiro
Idioma:inglês
Origem:RIA - Repositório Institucional da Universidade de Aveiro
Descrição
Resumo:Software developers face many adversities while working on projects, one of the most important ones being schema evolution. This is an inevitable procedure that, when not given the proper attention, can render an whole application unusable. The best option to study the impact of schema evolution on software development was the mining of data in a large database application. The chosen case study was MediaWiki. Using existing tools to mine the les of the MediaWiki project, the aim was to nd which les caused and which were a ected by schema evolution, and to extract frequent patterns. Building on existing work on mining data schema updates in each revision of MediaWiki, it was possible to extract over a hundred di erent association rules. These rules allowed the development of a recommendation system. This system allows to provide two les working on a database schema and receive as an output possible missing updates that should be applied to the newer schema. This will help software developers to keep their database schemas coherent and concise.