25.07.2013 Views

Download Complete Issue in PDF - Educational Technology & Society

Download Complete Issue in PDF - Educational Technology & Society

Download Complete Issue in PDF - Educational Technology & Society

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

the LOR, the number of selected learn<strong>in</strong>g objects <strong>in</strong> any particle can be bounded <strong>in</strong> an <strong>in</strong>tegrity rule,<br />

l max{e<br />

}, u m<strong>in</strong>{e<br />

}] , dur<strong>in</strong>g the generation of the <strong>in</strong>itial swarm. Moreover, to obta<strong>in</strong> a quality <strong>in</strong>itial swarm, a<br />

[ i<br />

i<br />

i = 1~<br />

n<br />

i=<br />

1~<br />

n<br />

selection rule is developed that can give higher selection probability to the learn<strong>in</strong>g objects that have difficulty levels<br />

closer to the target. Formally, the selection rule is def<strong>in</strong>ed as ( S di<br />

D ) S − − , where di is the degree of difficulty of<br />

learn<strong>in</strong>g object Oi and S is a constant.<br />

After <strong>in</strong>itiat<strong>in</strong>g the particle swarm, the module applies Equation 5 to measure the quality of each particle and then<br />

conducts particle iterations. In order to make sure that the best particle <strong>in</strong> each iteration survives, the elitist concept<br />

of the genetic algorithm (GA) has been <strong>in</strong>corporated <strong>in</strong>to EPSO (L<strong>in</strong>, Huang, & Cheng, 2010). If the best particle of<br />

the present iteration is worse than that of the previous iteration, the latter would replace the worst particle of the<br />

present iteration. By us<strong>in</strong>g Equation 10, each particle can update its velocity and position. Until the iteration<br />

term<strong>in</strong>ates, the draft is displayed <strong>in</strong> a web-based <strong>in</strong>terface and the <strong>in</strong>structor can check the results based on her or his<br />

own expertise, as shown <strong>in</strong> Figure 4. If the <strong>in</strong>structor is unsatisfied with the results, then she or he can require the<br />

module to produce another draft of the teach<strong>in</strong>g material or revise the draft <strong>in</strong> the wiki-based revision site, as shown<br />

<strong>in</strong> Figure 5.<br />

Figure 4. Screenshot of the draft generation <strong>in</strong>terface<br />

Figure 5. The draft of the teach<strong>in</strong>g material<br />

109

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!