On the campus network of multiple data source information retrieval system design and implementation of.docVIP

On the campus network of multiple data source information retrieval system design and implementation of Paper Keywords: information integration of heterogeneous data ODI Nutch Lucene Abstract: campus network resources, a huge amount of information, the information release system, independent of each other and use a variety of heterogeneous data sources in the context of a unified campus network information retrieval obstacle, the system focus on solving the unstructured text Structured data and database data into a multi-data source integration and integration issues, based on the Nutch search engine interface to use Lucene to index a variety of data sources, build a multi-data source full-text information retrieval platform, so as to effectively implement the whole network of information retrieval and improve search speed and accuracy. I. Introduction As the campus continues to process information in depth, the number of online information resources on campus rapid expansion of independent information dissemination system to improve efficiency, but also within the scope of a unified campus network to set up barriers to information retrieval, information on campus resources consist mainly of two types of data: one is unstructured text data to Web files, text files, e-mail stored in multiple information systems are: the other is the structured data, data stored in different records heterogeneous database. As a result of independent information systems not interconnected channels, rapid retrieval of information within the campus network, there are big difficulties how to design a stable and efficient structure, able to integrate multiple data sources of information and integration, to achieve network-wide full-text information retrieval process as the campus information technology is an important research topic. Campus network information retrieval technology can be roughly divided into three stages: the first stage is based on the database query t