Services
Purchase an archived recording with accompanying slides- $50 USD
Description:
In this call we heard about ways to integrate and apply taxonomies to content management systems from Seth Earley, Ted Sullivan of Raritan Technologies, and Meredith Lavine of the MathWorks.
Seth Earley
Seth started off the call by introducing the major issues in Content Management Systems (CMS) integration. Since Information Architecture is vague and big, a taxonomy can be applied to improve IA. After presenting some definitions for IA, Seth discussed various ways that CMS Information Architecture can be leveraged with a taxonomy. These include navigation, metadata design, user experience, CMS template development, workflow processes, wire frames, and search integration.
A guided navigation/faceted search/faceted browse leverages metadata applied to documents. It can also use stored searches or faceted search mechanisms. This allows for multiple navigation schemes based on taxonomy. Users think they are browsing, but they are actually searching.
A series of screenshots provided examples of fact/nodes as implemented on a site for navigation. In this case, three nodes of taxonomy (product, market, and application) are used to guide the search, which appears to the user as merely navigation.
A taxonomy may also support “term expansion” (mapping multiple terms with the equivalent meaning) by integrating a search tool with thesaurus entries.
Finally, issues in a large-scale taxonomy were presented.
Taxonomies and Content Management
Ted Sullivan, Raritan Technologies
Ted Sullivan’s presentation aimed to address CMS integration challenges, large taxonomies leveraging, search integration, multiple taxonomies, and changes and updates to taxonomies.
Ted began by explaining that, although a taxonomy organizes concepts, it does not (by itself) organize content. What is needed is a way of relating or "binding" the content (documents, web pages, spreadsheets, presentations, etc.) to the nodes in the taxonomy.
Categorization is the linking of concepts to content. This linking requires efficiency, accuracy, and completeness. The approach chosen should consider the number of documents and the numbers of taxonomy nodes. There are various approaches to taxonomy categorization: manual assignment of content to taxonomy nodes, matching of keywords (metadata), and search engine software to match query rules to documents.
Ted explained the different degrees automation in categorization and the different types of technologies for workflow approaches to fully-automated methods of categorization.
Screenshots of the Verity Intelligent Classifier tool for automatic classification and automated concept extraction were shown. Another site that creates a taxonomy on the fly by analyzing relationships between terms and documents is shown. The software uses algorithms, concept analysis, and statistical analysis.
Next, Ted discussed when to use multiple taxonomies rather than just one and then presents an example of multiple taxonomies in use through a series of three screenshots of the PharmaCI database, a repository of pharmaceutical documents, which uses a faceted search. It has three taxonomies linked: diseases from Clinical Trials, topics from Medical Subject Headings (MeSH), and companies. Ted also demonstrated term expansion (synonyms displayed) with screenshots from the CRISP biomedical thesaurus.
Uses of taxonomy and content management were explained. Clustering technology can be used to deal with changing terminology.
Taxonomy and Content Management at The MathWorks
Meredith Lavine, the MathWorks
Meredith Lavine’s presentation was of an example of a corporate taxonomy. The MathWorks taxonomy started out as a market segmentation exercise, whereby the taxonomy was proposed as a value for marketing. It has increased leads and prospects, registrations for marketing events, and cross-selling opportunities for current customers. The sales force able to find information quickly from multiple sources and turn it into potential revenue.
The taxonomy’s core nodes (referred to as “dimensions”) are Product, Application and Industry (PAI), Physical Type (documents), and Content Type (web content). Complementary taxonomies for certain content types reside in the CMS. A topical taxonomy is under development.
Meredith explained how the taxonomy has evolved and supports content management.
She presented a schematic diagram of how suggestions for taxonomy changes are approved in the organization. She also presented a diagram of how the taxonomy is managed through the Logistics system, the Electronic Document Management system (EDM), the Content Management System (CMS), and the Customer Relationship Management system (CRM).
The process is illustrated by an example of categorizing and promoting an event seminar. A series of web site screen shots demonstrate how the taxonomy is used to promote and display a specific event seminar. There are multiple web-interface access points based on the CMS taxonomy.
The taxonomy also drives the Product Technical Literature pages, which categorizes pdf documents. Meredith showed screenshots of how technical literature, book descriptions, and user stories (case studies), are each categorized and displayed in the interface. Finally, she showed a screenshot of how the taxonomy drives the Selling pages of the intranet and extranet.
Meredith concluded her presentation by listing various issues in the evolution of the taxonomy.
About your presenters:
Seth Earley is founder and senior consultant for Earley & Associates, Inc., a consulting firm specializing in knowledge, content and document management systems with a focus on taxonomy development and workflow design. With 20 plus years in the technology field, he has been involved in knowledge and content management processes for the past 12 years.
Meredith Lavine is business analyst in the eMarketing group at The MathWorks, where she is responsible for content and document management. She has ten years of experience with the design and implementation of knowledge repositories, content management systems, portals, communities and other organizational knowledge sharing strategies and processes. Previously, she held KM positions at IBM, Novell and The Boston Consulting Group. Meredith holds advanced degrees in music and library & information science.
Ted Sullivan has over 20 years of experience in a diverse range of information technologies. His background includes biomedical research, mathematics instruction, developing military planning systems, and numerous eBusiness solutions. Ted received a BSc from Cornell University and a PhD from Washington University. He was formerly on the faculty of Princeton University.