Adaptive Retrieval Of Semi-Structured Data

The rapidly growing amount of heterogeneous semi-structured data available on the Web is creating a need for simple and universal access methods. For this purpose, we propose exploiting the notion of unspecified Ontology (UNSO), where the data objects are described using a list of attributes and their values. To facilitate efficient management of UNSO data objects, we use LoudVoice, a multi-agent channeled multicast communication platform, where each attribute is assigned a designated communication channel. This allows efficient searches to be performed by querying only the relevant channels, and aggregating the partial results. We implemented a prototype system and experimented with a corpus of real-life E-Commerce advertisements. The results demonstrate that the proposed approach yields a high level of accuracy and scalability.