Re: Tagging Assets Covering the Subject Universe
- Robin, I feel your pain.
I'm in the same boat, looking to support a folksonomy for a single
facet(subject terms/keywords) amongst a lot of other structured data.
I'm working on a DAM deployment that focuses on visual resources and
I'm relying on the user base to do much of the cataloging work. Most
of our facets are very structured, and capture valuable information
that is useful once you've found the object of your desire, but few of
them are descriptive enough to get you there in the first place.
Because these are images auto-content extraction tools are not much help.
Because my user base is composed of artists, designers and
photographers who lament typing - I've been told to keep the
requirements low - while keeping findability and functionality as high
as say - Getty Images. I've also been told not to train them using
confusing words like "Disambiguation", "Term Lemmitization" and "Noun".
We are already taking "free-form" tags and mapping them to approved
terms to reconcile language ambiguities in search and indexing - that
is less of an issue. When a unique tag is added to the system the
taxonomist is notified and takes care of any needed relationships or
establishes a new term based on the user "suggestion". The
"suggestion" remains in the database while the index expands to
include the relationships - the only problem is that the feedback loop
to the users (visibility of the metadata) does not include the
expanded relationships and may be training them to use the
"suggestion" rather than the approved term.
What we are struggling to develop, is an indexing aid that takes their
term "suggestions" and pushes them towards approved vocabulary and
minimizes the amount of work that the taxononmist (me) has to do in
building new relationships into the thesaurus.
I've proven to users that clicking on approved terms is faster than
having them write unnapproved ones. I've also used static and paper
models to show then how they can use their 'suggestions' to navigate
to approved terms, I just can't find a tool to support the suggestion
process that is sexy enough for my users (again all very visual
My questions are:
A) What are the indexing aid programs out there that can be easily
integrated (anybody with Documentum experience?)and do they support a
user feedback function that drives them towards approved terms based
on their "suggestions".
B) How many other's are considering this hybrid approach with a single
facet when the rest of their data is very structured? I'm starting to
hear more noise about it recently and looking for people to share my
pain and push for tools to be developed.
Marketing Content Manager
--- In TaxoCoP@yahoogroups.com, "dalyrl" <rdaly@...> wrote:
> Hi all,
> I am supporting the development of the taxonomy/metadata model for a
> children's publisher's digital asset management system. FYI, we have
> not yet selected a DAM, nor any taxonomy software. The model will
> cover facets that will be controlled by the taxonomy, e.g. title,
> publication date, product type, genre, audience, reading level. In
> addition, very limited subject categorization will be controlled with
> the taxonomy. The dilemma that I am facing is how to deal with
> subject descriptors (keywords).
> Since we are not limited to a single subject-domain such as art,
> medicine, etc., but are limited rather by audience (children), the
> subject matter covered is "encyclopedic." The tagging will most
> likely be done by editors, not trained indexers. Other than LC
> Subject Headings, does anyone have a suggestion on a thesaurus or
> tool that could support the subject descriptor tagging? LCSH would
> of course cover all the subjects, but people cringe at the thought as
> it tends not to be non-librarian friendly.
> We are considering taking a hybrid approach - controlled metatags for
> everything except the subject descriptors, allowing "free form"
> tagging. The problem is how to deal with the search at the other
> end. Without a thesaurus to interface with, and no controls placed
> on the subject descriptors, how best do we handle synonyms, variant
> spellings, etc.?
> I have just learned of WordNet (Princeton's lexicon), and am looking
> at Siderean as a search possibility. My knowledge of Siderean is
> limited at this point. Would it use some sort of built-in lexicon
> like WordNet to automatically match synonyms to compensate for
> the "free form" tagging? We are also considering allowing social
> tagging of assets by users (company employees only) when they access
> an asset from the DAM.
> Thanks in advance for any insight!
> Robin Daly
- Dearest Taxonomy FriendsAt last I have something to share with you:)
Please see below a job that was sent to me - alas not my skill set therefore sending it out to the Taxo universe as there maybe someone out there who is interested. All yours and good luck.
_________________________________________________________________DETAILS BELOW:Recruiter:Andrew Young
Pivotal Solutions, Inc.
Director of Technical Recruitment
9 East 37th Street , 2nd Floor
New York , NY 10016
http://www.linkedin.com/in/YoungAndrewMy name is Andrew Young with Pivotal Solutions, Incorporated (PSI). PSI is an industry leader in IT Resourcing, we provide IT consulting & permanent staffing solutions to Fortune 500s, major banks, brokerage, insurance & accounting firms, manufacturers, retailers and other successful public and private enterprises.I have a Enterprise Content Management Taxonomy Analyst position available with our client listed below. If you are qualified and interested in this opportunity or know someone who may be interested please give me a call, answer the questions below, and send an updated resume.Title: Enterprise Content Management Taxonomy Analyst
City: Houston, TX
Position type: Permanent OpportunityDescription
This position is a member of a small team that is responsible for the systems and processes that enable the businesses to improve their ability to find up-to-date, relevant and trusted information, eliminate out-of-date and duplicate information, and preserve information needed for legal and regulatory compliance. The Taxonomy Analyst is responsible for working with Marathon stakeholders to maintain, extend and implement the enterprise-wide classification structure and associated metadata. This person also works with the ECM records consultant, records analysts and the Law Organization to determine the length of time that various types of content should be kept by Marathon to meet legal, regulatory and business requirements.Responsibilities:
1. Work with Marathon stakeholders to develop and maintain a comprehensive enterprise-wide classification structure (taxonomy) and metadata schemas that interface with the corporation's ECM solutions..2. Define and maintain the processes used to update Marathon's enterprise-wide taxonomy and retention schedule.
3. Collaborate with the IT organization to ensure that Marathon's enterprise-wide taxonomy is easy to use, is intuitive and provides tangible benefit via search and desktop user interfaces.4. Assist organizations in classifying their content so that it can be easily located and managed.
5. Track and coordinate the work of content catalogers throughout the business segments of Marathon with the goal of synchronizing metadata standards for consistency and technical interoperability.6. Champion and influence the implementation of enterprise-wide standards in content classification.
7. Develop and provide input to ECM training to help business segments learn how to use ECM classification tools to store and locate their content.8. Keep abreast of relevant concepts, practices, and trends in classification, content management, search engines/portals, database and data modeling structures, Semantic Web and Internet standards.Requirements
1. Degree: Bachelors Required
2. Major: Information Science, Computer Science, Engineering or Library Science preferred.
3. Computer Proficiencies: Solid understanding of content management, content classification, taxonomies, thesauri, metadata, XML, RDF, Web, search, portal and collaboration technologies.4. Certifications: Project management and Microsoft certifications are a plus.
5. Work Experience: 3-5 years of experience in: information architecture, library, document, data, and/or content management, taxonomy, thesaurus, metadata modeling, faceted classification construction and maintenanceIf you have the required experience and interest, Please answer the questions below, along with an updated resume:*** What is your visa status (US Citizen, Greencard, H1-b)???*** What is your current salary / bonus level???*** Why are you looking for a new position???*** How many years of IT experience do you have???*** How much experience in years do you have as a Enterprise Content Management Taxonomy Analyst and when was the last year you have used it???*** How much experience in years do you have with information architecture and when was the last year you have used it???*** How much experience in years do you have working with metadata and when was the last year you have used it???*** How much experience in years do you have with taxonomies and when was the last year you have used it???If you are qualified and interested in this opportunity or know someone who may be interested please give me a call and send an updated resume to Andrew Young at psi@....
Luggage? GPS? Comic books?
Check out fitting gifts for grads at Yahoo! Search.