Please use this identifier to cite or link to this item:
|Type:||Artigo de evento|
|Title:||Towards Query Model Integration: Topology-aware, Ir-inspired Metrics For Declarative Graph Querying|
|Author:||Gomes Jr. L.|
|Abstract:||Accompanying the growth of the internet and the consequent diversification of applications and data processing needs, there has been a rapid proliferation of data and query models. While graph models such as RDF have been successfully used to integrate data from diverse origins, interaction with the integrated data is still limited by inflexible query models that cannot express concepts from multiple paradigms. In this paper we analyze data and query models typical of modern data-driven applications. We then propose an integrated query model aimed at covering a broad range of applications, allowing expressive queries that capture elements from diverse data models and querying paradigms. We employ graphs models to integrate data from structured and unstructured sources. We also reinterpret as graph analysis tasks several ranking metrics typical of information retrieval (IR) systems. The metrics allow flexible correlation of data elements based on topological properties of the underlying graph. The new query model is materialized in a query language named in*(in star). We present experiments with real data that demonstrate the expressiveness and practicability of our approach. © 2013 ACM.|
|Appears in Collections:||Unicamp - Artigos e Outros Documentos|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.