Tuesday, November 4, 2014

Continental Breakfast
8:00 a.m. - 9:00 a.m.
WELCOME & KEYNOTE: From Cataloguers to Designers: A New Role for Taxonomists in Knowledge Graphs, Machine Classification, and Search Based Applications
9:00 a.m. - 10:00 a.m.
Patrick Lambe, Principal Consultant, Straits Knowledge Author, Principles of Knowledge Auditing

Professional taxonomists are still largely focused on organising information and content for the enterprise. However there are three big, interlocking trends in knowledge organisation beyond the enterprise that will drive the demand for taxonomists’ skills in the future: knowledge graphs and graph-based databases; machine-aided classification and its application to big data and statistical and monitoring systems; and the role of taxonomies in powering search based applications. In this shift, taxonomists will increasingly be called upon to act as designers of knowledge organisation systems, beyond their traditional role of cataloguing and mapping concepts and their relationships to each other. In this keynote, Patrick Lambe will illustrate the differences, in particular the importance of the shift from cataloguing what is, to designing what can be.

Coffee Break
10:00 a.m. - 10:15 a.m.
Track 1
Moderated by:
Aline Martinez, Taxonomist, News Product, The Washington Post
Taxonomy Fundamentals Workshop
10:15 a.m. - 11:45 a.m.
John Horodyski, Managing Director, Salt Flats
Meredith Brown, Associate, Optimity Advisors
This tutorial covers the fundamentals, strategies, standards, design methodology, and best practices of developing taxonomies for your digital asset management system, content management system, search, or metadata management system for your organization. It is based on best practices for establishing information structure, and managing and governing your digital assets. This tutorial addresses how to develop a taxonomy, improve existing metadata schemas, establish good governance using taxonomies and metadata, and calculate ROI for particular classes of users, digital assets, and media channels.

12:00 p.m. - 1:00 p.m.
Industry Insights
11:45 a.m. - 12:00 p.m.
Bryan Bell, Regional Vice President of Sales, Lucidworks
Hear from an industry leader about taxonomies, the management of information and knowledge, and insights on customers and trends.

1:00 p.m. - 2:00 p.m.
Taxonomy: Resources & Revelations
Eric Ziecker, Taxonomist, Editor, Information Consultant, Access Innovations, Inc. LightSource Information & Research Services, Ltd.
How does one stay current with taxonomy-related research and implementation? This presentation provides something for the novice taxonomist as well as something for the expert. In addition to plumbing the depths of some of the existing online taxonomy reservoirs, it introduces some new resources. Fresh research from the results of various surveys, polls and interviews of those working within the taxonomy field is presented.

Taxonomy: Taking a Team-Based Approach
Beth Maser, Director of Records and Information Systems, PPC
Robert Allen, Content Applications Developer, Production, F.A. Davis Company
A 125-year-old publishing company is still wedded to print. It has no taxonomy and is in the process of designing an enterprise taxonomy. The company utilized a team-based approach and added a taxonomy consultant to help it get through the initial design phase. This talk details the process from personality testing taken to set up the team to the final product. It takes an in-depth look at lessons learned and best practices in taxonomy design when working with a team of beginners.

Buy It or Build It? How Do You Make the Decision?
Marti Heyman, Executive Director, Metadata Standards and Services, Cengage Learning
When is it best to purchase a taxonomy rather than build it? Is it realistic to think you can ever buy a “plug-and-play” taxonomy—one which you can drop into your environment that will immediately meet your business needs? Probably not, but how do you know when it’s close enough and worth the investment versus when you should bite the bullet and build what you need from scratch? This talk discusses a process to use and the objective factors to consider (and quantify) to help drive some objectivity into what is potentially a very subjective decision.

Selecting Your Future: A Taxonomy Management Tool Selection Process
Ahren Lehnert, Principal Taxonomist, Nike Inc.
If taxonomy is in your future, then so too is a road map for implementation. When it comes to managing your taxonomy in a production environment, there are tool choices ranging from spreadsheets and text files to highly sophisticated taxonomy management software. How do you select the right taxonomy management tool for your organization? What process should you follow? This session describes the process used to select a taxonomy management tool for an oil and gas products and services provider, from requirements gathering to implementation.

2:00 p.m. - 3:00 p.m.
Getting Your Organization Ready for an Enterprise Taxonomy
Gary Carlson, Founder, Factor
Enterprise taxonomies cross organizational, system, and distribution boundaries. Crossing these boundaries presents real challenges to the taxonomist. Any omni-channel strategy or comprehensive portal project will need to address these challenges. Unfortunately, creating and managing the taxonomy itself is usually the least of the problems. Resourcing, organizational alignment, internal politics, and technology all impact the success of a taxonomy project. Identifying the risks and the organizational readiness for a taxonomy project can be a huge factor in the project success. Using real-world examples, this presentation highlights the types of problems that can torpedo a project and provides remedies that companies have used to make the taxonomy project (or more specifically, the omni-channel experience or intranet project) successful.

Promoting Taxonomy & Metadata Projects in Your Organization
Ben Licciardi, Manager, PwC
Suzanne Carroll, Product Director, Data Intelligence, XO Group (The Knot)
Whether you’re just getting started with taxonomies, building out a metadata schema, or working as a full-time taxonomist, you’ve likely found yourself in the unenviable position of explaining taxonomy and metadata to people who are totally unfamiliar with—and possibly even uninterested in!—the finer points of controlled vocabularies. Presenters provide tips and tricks on how to clearly explain what you do as a taxonomist to your coworkers, clients, and friends. Licciardi explores the many different—and sometimes conflicting—definitions of taxonomy and metadata and discusses what they mean in the context of information and content management. He provides useful advice on how to explain taxonomy and metadata to your colleagues, whether they’re content authors, IT specialists, database gurus, or C-level executives. Carroll shows how she’s borrowed techniques from data visualization and change management to quickly get her colleagues at XO Group engaged and on the same page.

Creating Visual Taxonomies in the Digital Age
Jocelyn Coverdale, Owner, Ballantrae Solutions
From the toolkit of LEAN Office, taxonomy builders can borrow an effective concept known as “Visual Controls,” a tool to manage processes through easy-to-understand, “at-a-glance” visuals that help teams and individuals quickly understand a process, and, perhaps as important, immediately spot deviations from standard practices with a quick check against a visual guide. The presentation provides a framework with examples, tools, and case studies of how to structure a taxonomy as a “visual control”—or process map—that makes filing and retrieval easy for end users, not to mention fun! The visual taxonomies demonstrated in this presentation offer participants creative approaches, methods, and examples to engage end users with the at-a-glance” findability a visual taxonomy provides. Demos include the features and benefits of various visual/digital tools that, when leveraged and incorporated into a visual taxonomy, can support the integration of paper and digital records/files in one system; the integration of local and cloud storage locations; metadata development and implementation; collaborative opportunities to build and use a centralized visual taxonomy; portal design and access for information assets; and the integration of social media data. Visual findability systems are not just the future of taxonomies, but a creative idea whose time has come now.

Taxonomy for Bartenders: Getting Stakeholder Buy-In
Angela Howze Pitts, Senior Analyst, Enterprise Knowledge
The best way to share a taxonomy is to break it down by parts to share with members of your team. The entire taxonomy as a spreadsheet is often too overwhelming, and spreadsheets are rarely the best tool to edit and proofread node names and categories. Here are tips for making the taxonomy accessible to stakeholders from whom you would like to elicit feedback, and gain approval or buy-in. Pitts demonstrates how to convert a .csv file in spreadsheet form into a readable and shareable Word document using fun examples of cocktails and recipes.

Coffee Break
3:00 p.m. - 3:15 p.m.
3:15 p.m. - 3:45 p.m.
Using Taxonomies to Leverage Value From Your Business Processes
Joshua Rattan, Director, PwC
Bethany Sehon, Manager, PriceWaterhouseCoopers
This session focuses on ways in which taxonomies can help companies locate, use and store information. Knowledge of how business units interact allow the taxonomy to provide a “window” into the vast interworkings of an enterprise and its business processes and can impact master data management, business intelligence, information security, and records management. It provides examples of how taxonomy and metadata models add value and insight, and how these models might be incorporated into an information management strategy.

The Economics of Taxonomies— Managing Assets & Liabilities
Justin Kollinger, MA Candidate in Communication, Culture & Technology, Graduate School of Arts and Sciences, Georgetown University
Although we have several decades of experience working with taxonomies, we don’t have a robust set of metrics that tells us whether what we’re doing is or isn’t working. Creating a taxonomy is the first step and has associated costs. From a management and an economic perspective, though, we need to know whether the taxonomies we’ve created are actually generating benefits and value, or whether they’re creating liabilities and costing us real dollars or other opportunities. Kollinger makes the case that taxonomies generate value when aligned with business goals, have clear outcomes, are well-designed and suited to their purpose, are managed in a sustainable way, and are continuously evaluated. When these factors are not in place, taxonomies can actually create liabilities and cost an organization in terms of reputation and future opportunities. Hear about metrics for different types of taxonomies, values you might expect and can promote to your management, and costs and liabilities you should try to avoid.

3:45 p.m. - 4:15 p.m.
Change Request Management With SharePoint or Lotus Notes— Lessons Learned
Ralph Tamlyn, Principal, Taxonomy and Classification Metadata Consulting IEEE, SLA, (ACBL as a hobby)
Managing taxonomy and ontology change requests is a cornerstone of taxonomy and ontology management. Change requests need to be managed from creation through assessment to resolution (approval or rejection). Assessment includes both assessment of the change within the taxonomy or ontology and assessment of the impact of the change on existing systems and business processes. Tamlyn discusses best practices for managing change requests and how to use a collaborative tool such as Lotus Notes or SharePoint to manage the change requests as items in a list. Among the topics covered is the use of views, as an aid in accessing, resolving, and implementing change requests.

Optimizing Taxonomy & Content Governance
Richard Iams, Information Architect, The Eliassen Group
Now that you have a taxonomy or content information architecture, how do you maintain it? This discussion leverages two contrasting, real-life scenarios as examples. The first depends on using best practices to build a taxonomy governance plan from scratch. The second considers existing, accepted business process workflows as inputs for developing governance policies and procedures. It’s all about finding the right balance for efficient decision-making.

One Taxonomy. Multiple Businesses. Unlimited Possibilities
Ann Donovan, Director, Content Strategy, User Experience Design, Fidelity Investments
Ashley O'Brien, Senior Publisher, Central Publishing, Fidelity Investments
Taxonomies are important because they allow us to create and maintain consistency when categorizing and organizing content. This is especially important for large and diverse companies such as Fidelity Investments, where there are multiple business units and multiple websites. In this talk, Donovan and O’Brien explain how they maintain and govern a single taxonomy that is used across multiple businesses and a number of websites. They also discuss how the taxonomy is used to not only organize information, but to share content across site sections, display content dynamically, gain useful web measurement statistics, and create a personalized web experience for Fidelity’s customers.

The Curious Lives of Full-Time Taxonomists
4:15 p.m. - 5:00 p.m.
Moderator: Zach Wahl, CEO, Enterprise Knowledge, LLC
Daniel O’Connor, Senior Analyst, Product Taxonomy Team, Target
Jami Ansell, Experience Analyst, Phase2
Sarah Barrett, Senior Information Architect, Factor
Barbara McGlamery, Director, Taxonomy & Metadata, Higher Education, Pearson
This popular session facilitates a conversation with a panel of full-time taxonomists from the public and private sectors, and the consulting world. The taxonomists discuss their career path, daily activities, and noted trends in the industry. The audience has the opportunity to ask questions, with answers and different perspectives provided from each panelist.

6:00 p.m. - 7:00 p.m.
Track 2
Moderated by:
Michael Crandall, University of Washington
10:15 a.m. - 11:45 a.m.
Moderator: Bob Kasenchak, Information Architect, Factor
This session is rich in short overviews of what you can do with your taxonomy: implementation in websites, production and publishing workflows, author submission, delivery in education and brand promotion. Come and hear what other things you can do now that you have a taxonomy.

Integrating Multimedia/Multilingual Content Through Taxonomy
Jami Ansell, Experience Analyst, Phase2
The State Department uses controlled vocabularies and auto-classification to provide appropriate, consistent, and up-to-date metadata for multimedia content produced by multiple sources, in multiple languages. This high-quality metadata helps connect State Department employees around the world with the resources they need to help them best communicate with their local audiences through a taxonomy-enhanced search application. The result is efficient access to diverse and geographically relevant media products for use in outreach activities promoting American culture and highlighting America’s role in international affairs.

Manual & Automatic Subject Tagging in PLOS
Helen Atkins, Director, Publishing Services, Public Library of Science (PLOS)
The PLOS thesaurus provides the taxonomy, a hierarchy, and a wealth of synonyms to use as subject matter tags for PLOS articles. It is used in paper submission, the production process, finding peer reviewers, navigation and search on the website, setting up RSS feeds so readers can create a custom scientific journal flow of their own, and in an ever-increasing number of other places as new options and functionality occur to the developers. Continuous auto-tagging of articles and taxonomist curation of the taxonomy ensures researchers can find their way around the scientific research held within PLOS.

Implementing a Taxonomy for the Common Core
Raj Cary, VP of Technology/Architecture, Triumph Learning
Much has been written about the Common Core Standards and their implementation across the entire United States educational curriculum for more uniform expected outcomes and ways to measure educational attainment. How can you be sure you are meeting all of the requirements? Is there a way to ensure that what is being taught will indeed allow students to meet the requirements of the standards? What about remedial areas? How do you know when they are needed? This unique application of a taxonomy for education exceeds the expectations of the Common Core and makes learning fun!

Integrate: Leverage Both Your Taxonomy & Peer Review System
Anna Jester, Director of Sales & Marketing, eJournalPress
Jester discusses the integration between Data Harmony and eJournalPress, allowing terms to be associated with a manuscript at submission and also once a manuscript has been accepted. One of the many benefits publications reap when using terms to correctly identify manuscripts is accurate assignment of potential reviewers during the peer-review process. This not only ensures the appropriate reviewers are asked to participate but can help widen the reviewer pool, preventing and alleviating reviewer fatigue.

Navigation vs. Backend Taxonomies: Case Study
Ari Kramer, Knowledge Management Officer, Robert Wood Johnson Foundation
Joseph Busch, Principal, Taxonomy Strategies
As part of an effort to expand their reach, increase engagement in their work among target audiences, and raise the visibility of their brand, the Foundation has been working on improving how “Topics” are framed, organized, and used in the RWJF website. Using a combination of evidence-based methodologies such as topic usage and search log analysis, and more qualitative methodologies such as interviews, competitive analysis, and textual analysis, the Foundation has been refining a set of broad web topics that distill the detailed and large set of backend metadata topics and keywords that originate in the RWJF’s project information management system. This talk discusses the process and considerations associated with developing a navigation taxonomy that can be provisioned with content that is organized and categorized using a backend taxonomy based on the RWJF web taxonomy and other related navigation and backend taxonomy experience.

Categories that Convert: 5 Keys to a Profitable Taxonomy
11:45 a.m. - 12:00 p.m.
Stewart Coerver, Principal Consultant, Codifyd
Constructing a taxonomy that drives search engine traffic and increases conversions involves more than just implementing a logical structure. Join Codifyd's presentation and learn how to create a taxonomy that spurs cross-selling opportunities and creates seamless browsing for more web traffic, better conversions, and higher online revenue.

12:00 p.m. - 1:00 p.m.
1:00 p.m. - 2:00 p.m.
From a Hierarchical Classification to a Faceted Taxonomy: Case Study
Arthur Smith, Lead Data Analyst for APS Journal Information Systems, American Physical Society
Joseph Busch, Principal, Taxonomy Strategies
Since 1975, APS Physical Review article submission, editorial assignments, and journal tables of contents have been based on the Physics and Astronomy Classification Scheme (PACS) developed by the American Institute of Physics (AIP), a five- to six-level hierarchical classification system. In 2013, APS decided to replace PACS with a new faceted scheme because AIP is no longer maintaining it, APS desires a single categorization scheme for all types of content (journal articles, conference papers, and website content), and to develop new capabilities for topic-based online services such as faceted navigation, targeted alerts, personalized subscriptions, etc. The proposed scheme includes several discrete sets of categories (facets) whose values can be combined to express concepts such as existing PACS codes, as well as new concepts such as those that have already been added by APS along with new categories that have not yet emerged or have been difficult to express with the existing PACS. APS is using the PoolParty thesaurus management tool to build, maintain, and publish the new faceted taxonomy using the SKOS standard. This presentation discusses the process that has been developed for the APS Taxonomy so that it is, as far as possible, backward-compatible with the legacy content categorized using PACS and is extensible and scalable to support new information services.

Elastic XML: XML Taxonomy Transformation using XSLTs
Jim Sweeney, Senior Product Manager, Taxonomy & Ontology Solutions, Synaptica LLC, USA

The Canadian Library of Parliament uses an Integrated Library System (ILS) and an associated OPAC (Online Public Access Catalogue) to manage and make accessible the Library’s collection. The ILS system requires a specific version of XML schema called MARC21 to operate. Seeing a need for a more streamlined solution than running a manual conversion each time the taxonomy is updated, they asked the Synaptica team to develop a turnkey method to transform outputs from their taxonomy management tool into MARC XML records, which they in turn would convert to MARC21 using the widely available MARCEdit toolset. Over the long term, MARC21 may be phased out in favor of an alternate XML schema. However, the provided solution will allow for the insertion of any XSLT (eXtensible Stylesheet Language Transformation) as part of the transformation process to provide for the transformation to any alternate XML output that may be used for any taxonomy consuming application.

Foundation Center Case Study: Adventures in Taxo Modernization & Negotiation
Stephanie Lemieux, President & Principal Consultant, Dovecot Studio
The Foundation Center is a nonprofit that maintains the most comprehensive data on U.S., and increasingly global, philanthropy to produce knowledge tools and research for the philanthropic sector. The Center has recently undertaken a large-scale revamp of its grant classification taxonomy—a 30-year-old nationally adopted classification structure. The aim was to update the classification scheme to reflect modern fields of interest and also to simplify how the Center manages its data. The Foundation Center also needed to get buy-in on a new, faceted model for the taxonomy that would streamline data capture but which represented a fairly drastic change in how grants were being coded. As hundreds of grant-making organiza- tions use the taxonomy to classify their activities, collecting and integrating comments from such a large pool of stakeholders across different sectors created a big challenge. Changes to categories also have far-reaching impact on what conclusions may be drawn from the data, since the taxonomy both reflects and shapes how the organizations think about the subject matter. For example, changing the list of population groups that will be coded for forces one to question social constructs around gender, sexual identity, and ethnicity. This case study illustrates how the Foundation Center approached the task of updating its classification scheme, including creating a new, modern data model; understanding how taxonomy shapes the data and reflects sociological views; getting buy-in from internal stakeholders; collecting and reacting to feedback from third- party organizations; and making the taxonomy publicly viewable and user-friendly.

2:00 p.m. - 2:45 p.m.
Keep the Complexity. Simplify With SKOS
James Morris, Solution Architect, Semaphore by MarkLogic MarkLogic Corporation
The "S" in SKOS stands for "simple".  As a W3C standard, SKOS is a taxonomist's common denominator, enabling vocabulary interoperability and interchangeability in a networked world. However, because of its simplicity, SKOS is sometimes criticized for not being able to support the necessary sophistication of some of our ontologies. At the same time, some valuable ontologies modeled in languages like OWL,  can seem impractical without an extensive conversion effort. By applying Linked Data techniques, Morris demonstrates that Semantic Web principles allow us to have it both ways. Adopting SKOS to make vocabularies more widely usable does not require sacrificing the complexity of an ontology.  

Retrieving & Using SKOS Taxonomy Data
Bob DuCharme, Technical Writer, Commonwealth Computer Research, Inc.
In addition to lots of data from Wikipedia infoboxes, the public web service DBpedia includes data about the categories into which concepts described by Wikipedia pages have been sorted. Much of this data conforms to the W3C SKOS standard for vocabulary management, a standard used by the Library of Congress, the New York Times, and many other organizations. This DBpedia data includes information about preferred labels, broader terms, and related terms, and being in SKOS means that modern taxonomy management tools can often take advantage of this data directly. Is all of this data clean and orderly? Not necessarily; it’s based on Wikipedia. However, other W3C standards besides SKOS can help you to extract, sort, and manage this data so that you and your enterprise can get the best value out of it using both open source and commercial tools. DuCharme looks at how to retrieve taxonomies ranging from horror movies to product categories and how to connect them to people, places, and things.

Semantic Indexing of Unstructured Documents Using Taxonomies & Ontologies
Jans Aasman, CEO, Franz Inc.
Life science companies and healthcare organizations use RDF/SKOS/OWL-based vocabularies, thesauri, taxonomies, and ontologies to organize enterprise knowledge. There are many ways to use these technologies but one that is gaining momentum is to semantically index unstructured documents through ontologies and taxonomies. This talk discusses two use cases and demonstrates two projects using a combination of SKOS/OWL- based taxonomies and ontologies, entity extraction, fast text search, and Graph Search to create a semantic retrieval engine for unstructured documents. The first project organized all science-related artifacts in Malaysia through a taxonomy of scientific con- cepts. It indexed all papers, people, pat- ents, organizations, research grants, etc., and created a user-friendly taxonomy brows- er to quickly find relevant information, such as, “ How much research funding has been spent on a certain subject during the last 3 years and how many patents resulted from this research?” The second project involved a large socioeconomic content publisher that has millions of documents in at least eight different languages. Reusing documents for new publications was a painful process given that keyword search and LSI techniques were mostly inadequate to find the document fragments that were needed. Fortunately, the organization had begun developing a large, SKOS-based taxonomy that linked common concepts to various preferential and alternative labels in many languages. This taxonomy indexed millions of document fragments, and the speaker shows how to perform relevancy search and retrieval based on taxonomic concepts.

Using Agile to Build a Taxonomy/Ontology
2:45 p.m. - 3:00 p.m.
Evelyn L. Kent, Principal, Bacon Tree Consulting
Learn how using an Agile approach to building a taxonomy or ontology creates a sound basis for classifying content and gets you started quickly. Kent talks about how ontologies can grow organically from your organization's needs, making them agile, functional and practical. She illustrates with lessons learned from building a mulit-faceted, general news ontology that needed to classify often ambiguous content.

Coffee Break
3:00 p.m. - 3:15 p.m.
3:15 p.m. - 4:15 p.m.
What’s Cooking? How Taxonomy & Metadata Improve the Search Experience
Ben Licciardi, Manager, PwC
Using cooking and recipe sites as examples, Licciardi explores how taxonomy and metadata help drive faceted search experiences, facilitate personalization, and optimize Google search results. He explores how backend metadata informs frontend search, user-interface design, and content display and distribution. He takes a look at how recipe sites are leveraging a particular type of metadata— microdata—to enable faceted search experiences in Google. Although this talk focuses on the tastier side of web content, the principles outlined are applicable to a broad range of industries and websites.

Taxonomy in Enterprise Search: Making the Business Case
Mindy Carner, Senior Manager, Optimity Advisors
Many companies turn to enterprise search to help them sift through terabytes of content. But many decision makers do not see the value in including taxonomy in a search system that is advertised to work “out of the box.” The real problem is that many decision makers believe that when they purchase an out-of-the-box enterprise search system, they are going to get something that works for their content the way that online search engines work for the web. This just isn’t correct. Carner walks you through the differences between enterprise search appliances and online search, explores how taxonomy fills the gaps presented by an out-of-the-box search appliance, and raises the ROI on this project from “Well, it’s better than what we had before,” to “a vastly improved search system that consistently returns high value results.” She shares the tools you need to return to your stakeholders and present a strong business case for including taxonomy in an enterprise search development project from the very beginning.

SEO Strategy: It Begins With a Robust Digital Taxonomy
Tina Johnson Marcel, Senior Content Lead, Siteworx, LLC
When developing a digital taxonomy strategy with clients, a very consistent and recurring theme is SEO: “We want to improve search.” Educating stakeholders on the difference between taxonomy and search and how they work together can prove challenging. Taxonomy is not search; however, taxonomy can enhance search. The key is finding a way to illustrate how the two can deliver an intuitive and flexible digital experience. In short, it’s all about semantics. Marcel provides language and tools you can use to explain to clients how an effective SEO strategy begins with a robust digital taxonomy. Learn how to help clients visualize the co-dependent relationship between the two specialties and how those connections—when implemented correctly—can drive the ability to refine search and browser results, creating an optimal and contextual experience to help users find the content they need.

Using a Predictive Model in Search
Beth Maser, Director of Records and Information Systems, PPC
Giancarlo Crocetti, Associate Director, Information Access, Boehringer-Ingelheim; Professor, St. Johns University
Andrew Fast, Chief Scientist, Elder Research, Inc.
This case study illustrates how an organization’s data and applied predictive analytics can predict outcomes. With a layer of taxonomy over the predictive analytics, the predictions improved by 10%. Hear how the two tools can be combined to be extremely powerful.

4:15 p.m. - 5:00 p.m.
Three Steps Toward Connected Taxonomy Management
Daniel Mayer, CMO, Expert System Enterprise
Taxonomy offers the promise of improved content organization, search, and navigation. But with enterprise content growing by 50% a year, automation is a key to scalable metadata enrichment and to making taxonomy’s promise come true. The good news is that integrating taxonomy work, semantic enrichment, and domain experts’ feedback is possible and bridges the gaps that are present in these work-flows when they are disconnected. Mayer showcases how these workflows can be connected in practice and provides concrete examples of the resulting productivity and quality benefits.

Auto-Cat: Now That I Have It, How Do I Manage It?
Paula R McCoy, Managing Editor, Science & Taxonomy, ProQuest
Once you have gone through the process of defining what you want to do with auto-categorization, evaluating and selecting vendors, and then installing the software, is your work done? By no means! Actually implementing auto-cat within your operations means you have to manage it. And managing it requires decisions about how you will maintain it from Day 1 onward. This presentation provides guidance on how to manage the early days of a new auto-cat process; how to determine who and how many people you need to handle it; how to assess the results of auto-cat and then ensure that you have continuous improvement; and how to justify that your auto-cat has been a success.

Text Analytics: A Tool for Taxonomy Development
Tom Reamy, Chief Knowledge Architect & Founder, KAPS Group Author, Deep Text
Text analytics and taxonomy have been partners for a long time, as text analytics auto-categorization typically requires a taxonomy. However, as new capabilities are added to text analytics software, including advanced text mining capabilities, text analytics can now be used to greatly enhance the entire process of taxonomy development in a number of ways. This talk explores how these new capabilities were used in two projects to develop taxonomies. Text mining was used to uncover not just the most frequent topics in a text corpus, but by combining that with other analytics (categorization, entity extraction, and summarization) KAPS was able to characterize these topics according to such dimensions as basic generality, their centrality to the corpus using advanced relevance and clustering techniques, and the activities associated with various topics (using verbs as much as nouns). An additional success was to use text analytics to first identify dynamic communities within the organization based on their communication and then develop audience-specific taxonomies based on an analysis of the writings of those communities.

6:00 p.m. - 7:00 p.m.