designing a handwriting recognition based writing environment
Download
Skip this Video
Download Presentation
Designing a Handwriting Recognition Based Writing Environment

Loading in 2 Seconds...

play fullscreen
1 / 23

Designing a Handwriting Recognition Based Writing Environment - PowerPoint PPT Presentation


  • 525 Views
  • Uploaded on

Designing a Handwriting Recognition Based Writing Environment J C Read, S J MacFarlane, C Casey Department of Computing, University of Central Lancashire, Preston, UK Introduction Background Information Description of the Observational Study Findings from the study General Findings

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about 'Designing a Handwriting Recognition Based Writing Environment' - salena


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.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
designing a handwriting recognition based writing environment

Designing a Handwriting Recognition Based Writing Environment

J C Read, S J MacFarlane, C Casey

Department of Computing, University of Central Lancashire, Preston, UK

introduction
Introduction
  • Background Information
  • Description of the Observational Study
  • Findings from the study
    • General Findings
    • Findings relating to the Handwriting Recognition
    • Satisfaction measuring
  • Informing the design
  • Description of the Prototype
  • Conclusions

[email protected]

who why what where how
Who? Why? What? Where? How?
  • Janet C Read, lecturer and mother!
  • Elodie, PhD study
  • Speech, handwriting: - Human Language Technology, Free text not command.
  • Lancs..UK; white rural primary – age 7 – 9
  • Research, Observations, Usability studies

[email protected]

previous work
Previous Work
  • QWERTY keyboard difficult (WES2000)
  • HLT attractive to children, HR feasible (HCI2000, HCI2001)
  • Measuring Fun (CandF2000, CandF2001)
  • Participatory design (IDC2002)
  • Errors in HR interfaces (NordiCHI2002)

[email protected]

the observational study
The Observational Study
  • Children aged 7 and 8
  • Normal classroom activities
  • In twos
  • Laptop (HR), Desktop (QWERTY), Desk (Pencil)
  • Different writing tasks
  • Difficulties, Errors, Corrections

[email protected]

pen and paper errors
Pen and Paper Errors
  • Errors made – missing words, spelling, letters written backwards
  • Error prevention – asked, avoided, looked
  • Error discovery – reading back, self, teacher or another child
  • Error repair – rub out, scribble out, cross out, overwrite, re-write, squeeze in , change

[email protected]

qwerty errors
QWERTY - Errors
  • Errors made – missing words, spelling, hit wrong key
  • Error prevention – asked, avoided, looked
  • Error discovery – reading back, self, teacher or another child, wiggly lines!
  • Error repair – position and rub out, rub back to, rub all, retype, change

[email protected]

handwriting recognition overview
Handwriting Recognition - Overview
  • Hardware – Graphics tablet and pen
  • Software – Recognition software
    • Fuzzy computing
    • Disobedient – ambiguous
    • Character or word based
    • On line – ‘t’ stroke problems
  • Demonstration

[email protected]

handwriting recognition errors 1
Handwriting Recognition – Errors (1)
  • Child
    • Errors made – miss words, spellings, letters backwards, pen up
  • Computer
    • Errors made – Bad recognition, hardware
  • Child
    • Error prevention – ask, avoid, look
  • Computer
    • Error prevention – spell checker (not used)

[email protected]

handwriting recognition errors 2
Handwriting Recognition – Errors (2)
  • Child
    • Error discovery
      • Before Recognition – reading back, teacher, other child
      • After Recognition – as above + wiggly lines!
    • Error repair
      • Before recognition – scribble out, overwrite, insert letter
      • After Recognition – rub back to, rub all, rewrite all or some, use QWERTY

[email protected]

satisfaction measuring
Satisfaction Measuring
  • Errors do not imply dissatisfaction
  • WHY?
    • Sticky – addictive vs. nothing better
    • Funny – humour with recognition – easy to use

[email protected]

designing a prototype method
Designing a Prototype - method
  • Users
    • Children, environment, characteristics, mental models
  • Tasks
    • Goal oriented – hierarchy
  • System
    • States – dangerous states
  • Interface
    • UI design guidelines

[email protected]

child
Child
  • Classroom based – standard equipment, needs to be easy to use, robust, minimal help needed
  • Children – varied pen control, different levels of expertise with technology, different reading skills, poor or very good letter formation
  • Mental model – see tablet as paper – want to scribble out and insert missed words

[email protected]

user goal
User Goal
  • To produce good written work
    • Planning
    • Translation
    • Reviewing and Editing

(Hayes and Flower)

[email protected]

supporting the writer 1
Supporting the writer (1)
  • Ideas – pop up in clouds, can have many, child can re-order them and can put them away, use handwriting that is not recognised
  • Translation – training supported, lines can be drawn on screen or on the tablet (or both!); recognition can be immediate or delayed;

[email protected]

supporting the writer 2
Supporting the writer (2)
  • Reviewing – computer can read back recognised text, child can read recognised or script text; spellings may be highlighted in recognised text – teacher controls
  • Editing – child can edit with rubber and pen on script, or with keyboard on recognised text

[email protected]

system states
System States
  • Entry state
  • Recognition state
  • Edit state
    • DANGEROUS STATES
      • Pens that point
      • Cursors that confound
      • Spaces that stop

[email protected]

interface design 1
Interface Design (1)
  • Full writing screen
  • Ability to place new pages
  • Menus at the bottom
  • Haptic boundary preferred
  • Tablet matched to screen
  • Pen can be turned on and off

[email protected]

interface design 2
Interface design (2)
  • Video clip facility
  • Teachers screen
  • Assistant
  • Customisable
  • Training activities

[email protected]

and so
And so……………
  • The designs for a product for a small group of users, for a narrow application
  • Keyboard interface
  • Error repair
  • Speech recognition

[email protected]

thank you
Thank you

Janet C Read

University of Central Lancashire

Preston

Up North!

England

[email protected]

[email protected]

ad