Smartlogic Semaphore

We recently presented a webinar with Microsoft and Earley & Associates to look at how Smartlogic Semaphore can improve the search capabilities of Microsoft Sharepoint and FAST ESP, and how Semaphore makes accurate tagging of enterprise data easy, leading to a better search experience for users.

Presenting in the recent webinar, which you can watch online now, are our own Toby Conrad, Nate Treolar from Microsoft and Seth Earley from Earley & Associates, and we discussed how Semaphore enables enterprise content authors to accurately, easily and consistently tag their content, which means faster, more accurate search results for users.

Both Nate and Seth discuss the messy, ambiguous nature of searching, and how refinements are often needed to original search queries to extract the context of the user’s search. As well as drawing on the basic meta data for documents, Semaphore enables entity extraction, for example, a person’s name, or other instances of a given phrase within a document, to enable refinement of the search – delivering clearer, more defined results from ambiguous searches.

A topic that came up a few times was the idea of thinking of search “beyond the white box”. That is, equating search with an entire user experience, not just the function of typing a query into a text box.

In the webinar we saw how Semaphore’s enhancements for Shareopint and FAST ESP led to a more interactive search results page, with features such as document thumbnails and “quick previews” to give users visual cues that help them find the document they want.

It also provides contextual search results, and the ability to explore broader or narrower topics, which help searchers locate their information quicker, and feel more confident in the results they have been presented with.

Search is increasingly looking like navigation. “Users are more confident in the relevance of a document or content when they navigate to it than if they conduct a search for it using relevant terms,“ says Toby, Smartlogic Semaphore can create richer cataloguing of content to improve accuracy of search results, and help users to navigate to information rather than searching for it.”

You can hear the whole recorded webinar here.

Join Smartlogic's Toby Conrad and Lucid Imagination's Eran Yaniv for a detailed debrief on how combining Semaphore with Lucene/Solr turns a good search engine into the platform for compelling, intuitive search applications.

The webinar will be held on Tuesday, July 27 2010.

North America: 11:00 PDT / 14:00 EDT - Sign up

Europe: 14:00 BST/ 15:00 CET - Sign up

Are your search results cutting it? The daunting diversity and rapid growth in information that search applications must slice through puts constant pressure on the quality of search results. There are many individual approaches that help ensure users can find actionable information quickly — and the best bet is attacking the problem from more than one direction.

Now, Smartlogic have partnered with Lucid Imagination, the commercial company for Apache Lucene/Solr, to cut through chaos for better findability. Lucid Imagination helps you harness the power of Lucene/Solr search and the compelling economics of open source.

Smartlogic Semaphore adds a layer of findability on top of native Lucene/Solr's search, along with a rich set of APIs, enhancing content metadata using automated classification for dramatically better search results and improved navigation. Together, they deliver a semantic search experience that provides a flexible, scalable alternative for enterprise search.

In this free webinar, the presenters will discuss:

• Lucene/Solr search foundations and semantic extension for ontology and classification
• Understanding the Semaphore/Solr integration
• Using metadata to drive faceted search and intuitive, dynamic content navigation
• Using topic pages to drive more traffic to your site and to keep people for longer

About Toby Conrad and Smartlogic

Toby Conrad has been a Senior Consultant with Smartlogic for seven years and has advised organisations such as NASA, Bank of America, Royal Bank of Scotland, BNP Paribas, GSK and the University of North Carolina on enterprise search and semantic search.

Smartlogic’s Semaphore product adds Semantic Intelligence to Lucene/Solr and to content systems to enhance search results and provide powerful navigation of content. Semaphore delivers this by uniquely capturing an organisation’s topics, language and context into an ontology.

Smartlogic has hundreds of clients around the world, including Bupa, BNP Paribas, City of London, Unison, NASA, Bank of America, ABN Amro, Telegraph.co.uk, Ofsted, The NHS, and Yell.com.

Follow us on Twitter: twitter.com/smartlogic_com

Existing subscribers to McClatchy-Tribune’s news and image database already have access to a wide-range of comprehensive news, information on enterprises and images but they now want to improve the level of content classification and deliver more focused information. Integrating Smartlogic’s Open Semantic Platform to their database will help subscribers having to wade through thousands of results from a search and do one with the confidence that every story returned will be on the relevant subject.

For example, if a client wanted to read stories about Marriott Hotels, a general search might return a story about a girls’ lacrosse team that stayed in a Marriott on Maple Avenue. That story is about a girls’ lacrosse team, not about Marriott. Semaphore Classification Server classifies stories so that you could discover stories about Marriott – and indeed about any topic you’re interested in.

MCT will roll this platform out to their sales team to help them focus on more targeted sales leads, specific to location, business sector, and model of a given company. Smartlogic’s Semaphore helped define MCTs news content and returned only those news stories which are relevant. It searches precise topics, dates and organisations.

How it helped McClatchy-Tribune classify news content searches

Semaphore’s semantic search model uses ontology as a basis for content classification; taking a set of words and concepts that are related to one another, and defining those relationships to determine the exact topic of each story.

Smartlogic’s Semaphore Management software will help MCT’s sales team highlight more targeted prospects in sales situations. It will point to topic areas that are relevant to that specific sales lead. After the internal roll-out, it will be extended to MCTs subscribers as a value-added application that will save their clients considerable time in their online news and enterprise searches.

Watch our short video introducing the features of the Semaphore plug-in for OpenText's CMS (formerly RedDot CMS). This integration of Semaphore’s open semantic platform to the RedDot CMS makes it easy for editors to add complex, taxonomy-driven metadata to their pages.

Users don’t need to navigate complex tree structures to find appropriate meta data tags for content. Semaphore makes accessing extensive, complex taxonomies intuitive and straightforward. At the heart of the Semaphore plugin for RedDot are three simple ways to find or add metadata tags to OpenText/RedDot CMS:

• auto suggest – which reads the page and automatically suggests appropriate terms from the taxonomy
• search – where the user can enter a term and matching terms are returned from the taxonomy, and
• user suggest - where the user can feed back possible new terms to the information scientist who owns the taxonomy, via a note or through RedDot’s email function.

Once the user has chosen their tags, the RedDot element is updated and the screen refreshes with the new meta data tags. This integration of Smartlogic Semaphore’s open semantic platform to the RedDot CMS makes it easy for editors to add complex, taxonomy-driven metadata to their pages.

We invite you to a free online seminar that will cover the nuts and bolts of how to develop and tune FAST ESP or Microsoft Sharepoint search capabilities, and how to enable Sharepoint- or FAST-based portals to offer specific contextual navigation mechanisms that enhance the search/navigation experience.

The webinar will be held on June 29 2010.

Session 1: 6:00am US PDT / 9:00am US EDT / 2:00pm UK BST / 3:00pm EU CET

Session 2: 10:00am US PDT / 1:00pm US EDT / 6:00pm UK BST / 7:00pm EU CET

Register at www.earley.com/webinars/technology-showcase/fast.

With the insights and approaches covered in the webinar you will extend the value of assets to your users through unique experiences that expose relevant information quickly, drive innovation in critical business functions, consolidate multiple views of the customer, increase user adoption, tap into the knowledge of the community and facilitate collaboration, and “push” new content through alerts and notifications.

Learn more and sign up for the webinar.

Employees are coming to expect the same level of functionality from enterprise systems as they are getting from-public facing internet systems so there is internal pressure to invest. This degree of information classification and retrieval is driving enterprise adoption of semantic technologies.

Read the full article on Slashdot.org

Enterprise information management is trailing behind the World Wide Web – we’re only just starting to realise the importance of correctly tagging our documents. With automated, standardised and accurate meta tagging of enterprise information, the ability of staff to quickly and easily find and file the documents they need increases dramatically. Ontology-driven classification in semantic platforms like Semaphore means enterprise information can be grouped logically, and relationships between documents are clearly defined.

Smartlogic have today announced the general availability of the Semaphore 3.1.3 (point) release, delivering additional reporting and increased reliability.

The platform provides advanced semantic search capabilities when integrated with the Google Search Appliance, Sharepoint 2007, FAST ESP, other search engine and information management systems.

Improvements and additions for this point release of the enterprise semantic search platform include:

• Additional reporting and increased reliability in the Ontology Manager/Server
• Performance increases for Rulebase Generator
• Better UTF-8 support for Baseline Service and Search Enhancement Server
• “Https” support for topic maps in the sample GSA implementation we provide
• Increased reliability of Classification Server
• Updated documentation in both PDF and HTML versions

To arrange for a demonstration of Semaphore’s semantic search platform and information management capabilities, or for further information please contact support@smartlogic.com or call (UK) +44 (0)1223 451 046 / (US) +1 202 657 4483.

Expose enterprise ontologies and taxonomies to any application using Semaphore's search and navigation enhancement XML API.

In our short video, find out how Semaphore’s Search Enhancement Server can be used to improve search functionality on your website or enterprise information management system.

It exposes an organisation’s ontology via an XML API to enhance the ability of your website, CMS, document management system or intranet to query and retrieve content.

The browse service exposes the ontology structure – typically a hierarchy of terms. The output of the search query can be used in many ways. For example, a user can add taxonomy metadata to an item in Sharepoint. Semaphore calls the browse service to render a selection tree, so the user can browse the tree and find relevant tags. But Semaphore can also automate this process by automatically classifying content if required.

The term information service allows us to find all the information about a specific term. In this NASA example taxonomy we hold details of astronauts including basic profile info and a URL to an image.

In our GSA integration, the service is used to drive a pop-up panel as the user hovers over the astronaut’s name.

The concept mapping service executes a fuzzy search of the taxonomy or ontology contents. In our example, a health ontology shows several concepts that relate to healthy eating and diet. A query of “diet” will match several categories, either directly or via a synonym, or using stemmed variants, such as “dieting”.

The servers help users find matching content to an ambiguous query, such as “diet”. The number of results is shown against each returned category – the search enhancement server maintains a count of content items matching each taxonomy term - so only categories that will ultimately return search results can be displayed.

If the user selects “healthy eating”, a new content search is executed and a concept explorer is displayed that show the parent/child categories, all the ontology relationships, descriptions, synonyms, all in real time, from a call to the search enhancement server.

Although these terms are often used interchangeably, generally speaking, a taxonomy is a bit like a “tree” of structured information, whereas an “ontology” is a like a “forest”, and might be made of up of many taxonomies. Ontologies are also more carefully defined, and usually include a vocabulary of keywords that relate to the information in the ontology.