IBM Launches Customisable Semantic Email Search Tool
Last week a semantic email search application was made available by IBM for free download, IBM OmniFind Personal Email Search http://www.alphaworks.ibm.com/tech/emailsearch/download
The application addresses one of the challenges faced by email users unaware of the Semantic Web: it performs metadata tagging automatically without the user requiring knowledge of semantic annotation.
The core technology underlying OmniFind, UIMA (Unstructured Information Management), is not new; it uses pattern recognition to extract instances of concepts such as phone numbers and addresses, and labels them with semantic XML tags, creating relationships between them. Users then enter keywords to represent their search queries. The advantage of the application is that it is usable by both ordinary email users and developers. It enables experienced users to customise their search facility by editing the default tags, or creating their own, using regular expressions to extract concepts. It is suggested by IBM that this tool could be extremely useful in a corporate setting, as tags can be customised to an organisation’s specific information requirements.
The components include an e-mail crawler available as a plugin for Lotus Notes or Microsoft Outlook, a document-processing, concept-tagging pipeline, an indexing component, a run-time semantic search component, and a tagger utility. The image below provides a link to an example of the output of the tool.
Tags: email, IBM, pattern recognition, regular expressions, semantic search, XML