Interactive Design of Web Sites with a Genetic Algorithm - PowerPoint PPT Presentation

Loading...

PPT – Interactive Design of Web Sites with a Genetic Algorithm PowerPoint presentation | free to download - id: 7315ea-MzRjY



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Interactive Design of Web Sites with a Genetic Algorithm

Description:

Interactive Design of Web Sites with a Genetic Algorithm A. Oliver, N. Monmarch , G. Venturini Laboratoire d Informatique, Universit de Tours – PowerPoint PPT presentation

Number of Views:18
Avg rating:3.0/5.0
Slides: 27
Provided by: univ190
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Interactive Design of Web Sites with a Genetic Algorithm


1
Interactive Design of Web Sites with a Genetic
Algorithm
  • A. Oliver, N. Monmarché, G. Venturini
  • Laboratoire dInformatique, Université de Tours
  • http//www.antsearch.univ-tours.fr/webrtic

2
Talk overview
  • The problem
  • The Interactive Genetic Algorithm
  • Style optimization
  • Layout optimization
  • Results

3
The problem context
  • Dynamic Site Publisher
  • On line web site editor
  • For non HTML users

4
The problem
  • First solution predefined templates

5
The problem
  • But
  • How many templates ? Which ones ?
  • Many sites to few differences between sites

6
What can we do ?
  • A simple and intuitive method
  • Take into account aesthetic preferences
  • Subjective notion
  • Can not be easily expressed
  • Interactive method
  • Generate solutions
  • Let the user choose the nicest ones
  • Generate new solutions according to the choices

7
Interactive Genetic Algorithm
8
Interactive Genetic Algorithm
  • Adaptation to variations in user choices
  • Intuitive use (select prefered individuals)
  • Several works in this field

9
Proposed IGA
  • Different coding for style (colors) and layout
    (position of pictures)
  • Comon algorithm
  • Generate the initial population of N individuals
  • Display the N individuals
  • Stop and apply style/layout on site or continue
  • Select Nselect individuals (by the user)
  • Generate next population
  • Goto step 2

10
Proposed IGA
  • Step 2 Display individuals

11
Proposed IGA
  • Step 5
  • Case Nselect
  • 0 random generation
  • 1 generate N-1 individuals by mutation (Pmut)
  • N do nothing
  • 2N-1 crossover(Pcross) and mutation

12
Style optimization
13
Style optimization
  • Genetic operators
  • Random generation
  • Mutation
  • Crossover
  • Luminance test

14
Layout optimization
  • N objects (text, pictures)
  • Bin-packing like problem
  • Constraints
  • Correct html tables
  • Affect objects to cells

15
Layout optimization
  • Variable length representation

16
Layout optimization
  • Genetic operator random generation

17
Layout optimization
  • Genetic operator mutation

18
Layout optimization
  • Genetic operator crossover

19
Results
20
Results
21
Results
22
Results
23
Results
24
Results
25
Results
26
Conclusion and future work
  • Interactive Genetic Algorithm for layout and
    style optimization of a web site/page
  • Dependencies between objects in page
  • Powerpoint presentations
About PowerShow.com