1 / 17

(FP7-PEOPLE-2010-ITN-263954)

OPTIMALE Symposium Rennes, 6 June 2013. The handling of translation metadata in TM/MT environments: Insights from process research. (FP7-PEOPLE-2010-ITN-263954). Carlos da Silva Cardoso Teixeira Universitat Rovira i Virgili (Tarragona, Spain). Defining metadata.

chase-ochoa
Download Presentation

(FP7-PEOPLE-2010-ITN-263954)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. OPTIMALE Symposium Rennes, 6 June 2013 The handling of translation metadata in TM/MT environments: Insights from process research (FP7-PEOPLE-2010-ITN-263954) Carlos da Silva Cardoso TeixeiraUniversitat Rovira i Virgili (Tarragona, Spain)

  2. Defining metadata

  3. Defining metadata (cont’d)

  4. Motivation • Integration ofTM and MT  different workflows • Translation vs. Post-editing vs. Mixed (interactive) (pretranslated) (?) Metadata  Actual impact ? .

  5. Research questions Speed: Will you translate faster? Quality: Will you translate better? Effort: Will you feel more tired? • Explanation:Do metadataplay a role? How do the differences between the environments affect translators? Satisfaction ?

  6. How to test 10 expert translators(English > Spanish) 2 comparable source texts (~520 words, 28 segments) 4 types of translation proposals 7 x Exact matches 7 x High Fuzzy (85-99%) 7 x Low Fuzzy (70-84%) 7 x Machine translation 2 environments • a visual or interactive environment • a blind or pre-translation environment

  7. Visual vs. Blind in TM/2

  8. Data collection methods Keystroke logging Screen recording Eye tracking Retrospective interviews Quality review

  9. Results: Speed • Types of environment: V: Visual environment (metadata available) B: Blind environment (metadata not available) • Types of proposals: • E: Exact matches • H: High-range fuzzy matches (85%-99%) • L: Low-range fuzzy matches (70%-84%) • M: Machine Translation

  10. Results: Effort (typing) • Types of environment: V: Visual environment (metadata available) B: Blind environment (metadata not available) • Types of proposals: • E: Exact matches • H: High-range fuzzy matches (85%-99%) • L: Low-range fuzzy matches (70%-84%) • M: Machine Translation

  11. Results: Quality • Types of environment: V: Visual environment (metadata available) B: Blind environment (metadata not available) • Types of proposals: • E: Exact matches • H: High-range fuzzy matches (85%-99%) • L: Low-range fuzzy matches (70%-84%) • M: Machine Translation

  12. Results: Comparison SPEED-1 EFFORT (TYPING) QUALITY-1 • Types of environment: V: Visual environment (metadata available) B: Blind environment (metadata not available) • Types of proposals: • E: Exact matches in TM/2 • H: High-range fuzzy matches (85%-99%) in TM/2 • L: Low-range fuzzy matches (70%-84%) in TM/2 • M: Machine-Translation proposals in TM/2

  13. Interviews (1)

  14. Interviews (2)

  15. Reality vs. Perception Conjectures Higher cognitive effort when no metadata (≠ typing effort) Different cognitive strategies for processing MT and TM proposals (E, H, L) Satisfaction ~ familiarity, access to information, possibility to choose

  16. Implications/applications • Translators: job satisfaction, or at least task satisfaction • Vendorsand clients: productivity, payment schemes • Tool manufacturers: increase usability of tools  sales • MT researchers: quality estimation for MT proposals • Translator training: teach new required skills • Intellectual significance: how we process information?

  17. Thank you! Merci ! Obrigado! Carlos S. C. Teixeira Intercultural Studies Group Universitat Rovira i Virgili (Tarragona, Spain)

More Related