assessing the usability of machine translated content: a user-centred study using eye tracking

14
Assessing the Usability of Machine Translated Content: A User-Centred Study using Eye Tracking Dr. Stephen Doherty & Dr. Sharon O’Brien Centre for Next Generation Localisation School of Applied Language & Intercultural Studies Dublin City University

Upload: lester-george

Post on 01-Jan-2016

22 views

Category:

Documents


0 download

DESCRIPTION

Assessing the Usability of Machine Translated Content: A User-Centred Study using Eye Tracking. Dr. Stephen Doherty & Dr. Sharon O ’ Brien Centre for Next Generation Localisation School of Applied Language & Intercultural Studies Dublin City University. Outline. Introduction Research Aims - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Assessing the Usability of Machine Translated Content: A User-Centred Study using Eye Tracking

Dr. Stephen Doherty & Dr. Sharon O’BrienCentre for Next Generation Localisation

School of Applied Language & Intercultural Studies

Dublin City University

Page 2: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Outline

Introduction

Research Aims

Methods

Results

Conclusions

Page 3: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Introduction

Increased need for translation

Diversity of content and users

Rise in prevalence of machine translation [MT] both off- and online

Mixed reports of quality – attitudes and expectations

Divergence in R&D – translation studies/computer science

Evaluation metrics – human and automatic

Our focus here is on usability

Page 4: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Research Aims

To investigate if there are differences in usability between the English [source language] and the unedited machine translated target languages [FR, DE, SP, JP].

Or in other words: how usable is machine translated content?

Adoption of the ISO/TR 16982 definition of usability

Importance of ecological validity: real materials and users

Page 5: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Methods

User-centred approach [n = 30]; task driven – ‘new user’ scenario

Eye tracking [tobii 1750]:Fixation count and average duration

Attentional shifts; percentage time in each window

Textual regressions

Page 6: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking
Page 7: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Methods

Post-task questionnaire; five-point LikertComprehension

Task completion

Potential improvement

Future reuse

Recommendation

Recall

Page 8: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Methods

UsabilitySatisfaction

Efficiency [task success/task time]

Page 9: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Eye Tracking

Task timeLowest for EN [sig. JP]

Fixation count and average durationLowest for EN [sig. JP] for both

Attentional shifts; percentage time in each windowEN and FR spent most time in task window

EN fewest shifts of attention [sig. JP]

Textual regressionsRaw number and distance: EN and SP [sig. JP]

‘Long’ regressions: JP [sig. all others]

Page 10: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Questionnaire Results

ComprehensionEN rated highest [sig. for FR and JP]

Task completionEN rated highest [sig. for JP]

Potential improvementSP & EN rated as needing least improvement, but could still be improved upon

Future reuseFR & EN rated highest

RecommendationEN rated highest [sig. for JP and DE]

RecallEN scored highest [sig. for JP and DE]

Page 11: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Usability Results

SatisfactionEN rated highest [sig. for FR, DE, and JP]

Task completionEN and SP more successful [sig. JP]

EfficiencyEN most efficient [sig. JP and DE]

Page 12: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Conclusions

So, just how usable is raw MT?Similar results for EN, SP, and FR

DE and JP more problematic [MT system]

Functionally usable [more than just ‘gisting’]

UX best for EN users

MT viable for certain pairs

Human intervention necessary to ensure best UX

Page 13: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Questions?

[email protected] [email protected]

This research is supported by the Science Foundation Ireland (Grant 07/CE/I1142) as part of the Centre for Next Generation Localisation (www.cngl.ie) at Dublin City University.

Page 14: Assessing the Usability of Machine Translated Content:  A User-Centred Study using Eye Tracking

Predictors of Positive UX

Satisfied users: comprehension & task time

Satisfied users: recommend to others

Task completion: textual regressions

Cognitive effort: instructions aiding task completion