Term Mining and
Terminology
Management
A Corporate Setting Perspective
Awareness
Globally active organizations whose core
business is not communications-related
(translation, localization, information
management, etc.) are generally unaware
of the benefits of performing terminology
management.
KaraWarburton, LISA, 2001
Term Mining andTerminology Management in a Corporate Setting Perspective
Translation-
oriented
terminology
Only 41% of localization-mature
organizations have some terminology
management policy in place, almost solely
translation-oriented
Term Mining andTerminology Management in a Corporate Setting Perspective
Scope
• Technical documentation
• Controlled languages
• Translation and localization
• Translation automation
• Content and Knowledge Management Systems
• Knowledge organization
• Taxonomies and ontologies
• Learning Management Systems
• Knowledge nugget (knowledge representation)
• Self-contained reusable educational entities (Learning Object Metadata, IEEE
1484.12.1)
• Marketing management
• Customer service
• SEM/SEO
• Sentiment analysis
Term Mining andTerminology Management in a Corporate Setting Perspective
Integrations
Documentation
CMS Website
Marketing
Service & Support
LMS
Term Mining andTerminology Management in a Corporate Setting Perspective
CVS
Costs
(IDC,2004)
• Productivity of knowledge workers
• 15% to 35% searching for information
• Successfully completed 50% of the time or less
• Only 21% found the information they needed 85% to 100% of the time
• $6 million a year looking for and not finding information
• 15% of time for duplicating existing information
• Opportunity costs
• Reworking existing information that could not be located
• $12 million a year
Term Mining andTerminology Management in a Corporate Setting Perspective
Terminology
cost multiplier
(JörgSchütz/RitaNübel)
Term Mining andTerminology Management in a Corporate Setting Perspective
Product data
Documentation development
Authoring
Editing
Approval
Localization
Maintenance
0.1 - 0.2
0.5
1.0
2.0
5.0
10.0
20.0
Indexing for
searching
• 15’ per query
• 3.75’ per day
• $500 per adult worker per year
• 160 hours of reading each and every
week
Term Mining andTerminology Management in a Corporate Setting Perspective
Costs/Benefits
• Huge costs in the short term
• $150 per terminological entry (J.D. Edwards, 2001)
• The practical value does not match the
technical value
Term Mining andTerminology Management in a Corporate Setting Perspective
Accuracy Fundamental accuracy of statement is
the one sole morality of writing.
Term Mining andTerminology Management in a Corporate Setting Perspective
Payback
• Cost reduction
• Authoring, localization, training, customer service
• Overhead
• Time reduction in the production cycle
• Immediate 1% payback for larger businesses
• Productivity increase
• Time-to-market
• Qualitative improvements
• Branding
• Safety
Term Mining andTerminology Management in a Corporate Setting Perspective
Controlled
languages
The most valuable of all talents is that of
never using two words when one will do.
Term Mining andTerminology Management in a Corporate Setting Perspective
Fatal errors
• The LinateAirport disaster (Oct 8, 2001)
• Deficiencies in the airport layout and procedures
• Violations of ICAO regulations
• Incorrect signs to runway
• Incorrect, uncorrected readback
• Non-standard phraseology
• Irrelevant term (extension) leading to fatal misunderstanding
Term Mining andTerminology Management in a Corporate Setting Perspective
Keywords
advertising
Rem tene, verba sequentur
(Keep to the subject, words will follow)
Marcus Porcius Cato (Cato the Censor)
Term Mining andTerminology Management in a Corporate Setting Perspective
The long tail
Rerum enim copia verborum copiam gignit
(All this gives rise to a plethora of words)
Cicero
Term Mining andTerminology Management in a Corporate Setting Perspective
Term mining
• Complex knowledge-intensive task
• Different approach for different scope
• Hard to grasp in a corporate setting perspective
• Business intelligence
Term Mining andTerminology Management in a Corporate Setting Perspective
Mining terms
• Linguistic approach
• Based on rules and dictionaries
• Collocations
• One language at a time
• Issues
• Loans
• Synonyms, variants,
abbreviations
• Ellipses
• Improper usage
• Bitext
• Knowledge bases
• Knowledge discovery
• Statistical approach
• Language independent
• Based on frequency
• Repeated sequences of
syntagmas
• The frequency threshold
must be specified
• Frequency does not necessarily
means importance
• Much “noise”
• Monolingual corpus
• Indices
• Controlled languages
• Keywords
• TQA
Term Mining andTerminology Management in a Corporate Setting Perspective
TaaS test drive
• Building a Localization Kit
• 13688 words, 142 repetitions
• memoQTerm Extraction
• Statistical analysis
• 815 term entries from the English document
• 647 term entries from translation memory
• Tilde Wrapper System for CollTerm (TWSC)
• Linguistic analysis enriched by statistical features
• 3046 term entries
• KilgrayTerminology extractor
• Statistical analysis
• 3218 term entries
Term Mining andTerminology Management in a Corporate Setting Perspective
Complexity
problems
If all you have is a hammer, everything
looks like a nail
Term Mining andTerminology Management in a Corporate Setting Perspective
Terminology
management
in the cloud
Pros
• ZeroTCO
• Availability and deployability
• Collaboration features
Cons
• Limited scalability
• Security issues
• Integration costs
Term Mining andTerminology Management in a Corporate Setting Perspective
ROI
The proof of performance, i.e. ROI
considerations, of terminology
management within the corporate setting
is a challenge for future projects.
Stefan Kremer, 2005
Term Mining andTerminology Management in a Corporate Setting Perspective
LWSpecial
Term Mining andTerminology Management in a Corporate Setting Perspective
Thank you
Term Mining andTerminology Management in a Corporate Setting Perspective
info@s-quid.it

Term Mining and Terminology Management in a Corporate Setting Perspective

  • 1.
    Term Mining and Terminology Management ACorporate Setting Perspective
  • 2.
    Awareness Globally active organizationswhose core business is not communications-related (translation, localization, information management, etc.) are generally unaware of the benefits of performing terminology management. KaraWarburton, LISA, 2001 Term Mining andTerminology Management in a Corporate Setting Perspective
  • 3.
    Translation- oriented terminology Only 41% oflocalization-mature organizations have some terminology management policy in place, almost solely translation-oriented Term Mining andTerminology Management in a Corporate Setting Perspective
  • 4.
    Scope • Technical documentation •Controlled languages • Translation and localization • Translation automation • Content and Knowledge Management Systems • Knowledge organization • Taxonomies and ontologies • Learning Management Systems • Knowledge nugget (knowledge representation) • Self-contained reusable educational entities (Learning Object Metadata, IEEE 1484.12.1) • Marketing management • Customer service • SEM/SEO • Sentiment analysis Term Mining andTerminology Management in a Corporate Setting Perspective
  • 5.
    Integrations Documentation CMS Website Marketing Service &Support LMS Term Mining andTerminology Management in a Corporate Setting Perspective CVS
  • 6.
    Costs (IDC,2004) • Productivity ofknowledge workers • 15% to 35% searching for information • Successfully completed 50% of the time or less • Only 21% found the information they needed 85% to 100% of the time • $6 million a year looking for and not finding information • 15% of time for duplicating existing information • Opportunity costs • Reworking existing information that could not be located • $12 million a year Term Mining andTerminology Management in a Corporate Setting Perspective
  • 7.
    Terminology cost multiplier (JörgSchütz/RitaNübel) Term MiningandTerminology Management in a Corporate Setting Perspective Product data Documentation development Authoring Editing Approval Localization Maintenance 0.1 - 0.2 0.5 1.0 2.0 5.0 10.0 20.0
  • 8.
    Indexing for searching • 15’per query • 3.75’ per day • $500 per adult worker per year • 160 hours of reading each and every week Term Mining andTerminology Management in a Corporate Setting Perspective
  • 9.
    Costs/Benefits • Huge costsin the short term • $150 per terminological entry (J.D. Edwards, 2001) • The practical value does not match the technical value Term Mining andTerminology Management in a Corporate Setting Perspective
  • 10.
    Accuracy Fundamental accuracyof statement is the one sole morality of writing. Term Mining andTerminology Management in a Corporate Setting Perspective
  • 11.
    Payback • Cost reduction •Authoring, localization, training, customer service • Overhead • Time reduction in the production cycle • Immediate 1% payback for larger businesses • Productivity increase • Time-to-market • Qualitative improvements • Branding • Safety Term Mining andTerminology Management in a Corporate Setting Perspective
  • 12.
    Controlled languages The most valuableof all talents is that of never using two words when one will do. Term Mining andTerminology Management in a Corporate Setting Perspective
  • 13.
    Fatal errors • TheLinateAirport disaster (Oct 8, 2001) • Deficiencies in the airport layout and procedures • Violations of ICAO regulations • Incorrect signs to runway • Incorrect, uncorrected readback • Non-standard phraseology • Irrelevant term (extension) leading to fatal misunderstanding Term Mining andTerminology Management in a Corporate Setting Perspective
  • 14.
    Keywords advertising Rem tene, verbasequentur (Keep to the subject, words will follow) Marcus Porcius Cato (Cato the Censor) Term Mining andTerminology Management in a Corporate Setting Perspective
  • 15.
    The long tail Rerumenim copia verborum copiam gignit (All this gives rise to a plethora of words) Cicero Term Mining andTerminology Management in a Corporate Setting Perspective
  • 16.
    Term mining • Complexknowledge-intensive task • Different approach for different scope • Hard to grasp in a corporate setting perspective • Business intelligence Term Mining andTerminology Management in a Corporate Setting Perspective
  • 17.
    Mining terms • Linguisticapproach • Based on rules and dictionaries • Collocations • One language at a time • Issues • Loans • Synonyms, variants, abbreviations • Ellipses • Improper usage • Bitext • Knowledge bases • Knowledge discovery • Statistical approach • Language independent • Based on frequency • Repeated sequences of syntagmas • The frequency threshold must be specified • Frequency does not necessarily means importance • Much “noise” • Monolingual corpus • Indices • Controlled languages • Keywords • TQA Term Mining andTerminology Management in a Corporate Setting Perspective
  • 18.
    TaaS test drive •Building a Localization Kit • 13688 words, 142 repetitions • memoQTerm Extraction • Statistical analysis • 815 term entries from the English document • 647 term entries from translation memory • Tilde Wrapper System for CollTerm (TWSC) • Linguistic analysis enriched by statistical features • 3046 term entries • KilgrayTerminology extractor • Statistical analysis • 3218 term entries Term Mining andTerminology Management in a Corporate Setting Perspective
  • 19.
    Complexity problems If all youhave is a hammer, everything looks like a nail Term Mining andTerminology Management in a Corporate Setting Perspective
  • 20.
    Terminology management in the cloud Pros •ZeroTCO • Availability and deployability • Collaboration features Cons • Limited scalability • Security issues • Integration costs Term Mining andTerminology Management in a Corporate Setting Perspective
  • 21.
    ROI The proof ofperformance, i.e. ROI considerations, of terminology management within the corporate setting is a challenge for future projects. Stefan Kremer, 2005 Term Mining andTerminology Management in a Corporate Setting Perspective
  • 22.
    LWSpecial Term Mining andTerminologyManagement in a Corporate Setting Perspective
  • 23.
    Thank you Term MiningandTerminology Management in a Corporate Setting Perspective info@s-quid.it