<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/">
	<channel>
		<title><![CDATA[Forums - IntelligenceLab]]></title>
		<link>http://mitov.com/forum/</link>
		<description><![CDATA[Forums - http://mitov.com/forum]]></description>
		<pubDate>Mon, 11 May 2026 03:39:49 +0000</pubDate>
		<generator>MyBB</generator>
		<item>
			<title><![CDATA[Radial Basis Function]]></title>
			<link>http://mitov.com/forum/thread-310.html</link>
			<pubDate>Wed, 25 Apr 2012 04:11:39 -0400</pubDate>
			<guid isPermaLink="false">http://mitov.com/forum/thread-310.html</guid>
			<description><![CDATA[Hello,<br />
<br />
can anybody explain me how the radial basis function works?<br />
<br />
I can configure the parameters MinInfluenceField, MaxInfluenceField and Background, but how do they exaclty influence the radial basis function?<br />
<br />
When I look at the theory of the radial basis function there are many different functions that can be used, but I can't find anything about something like a MinInfluenceField, MaxInfluenceField or Background.<br />
<br />
And what is the actual basis function used here?]]></description>
			<content:encoded><![CDATA[Hello,<br />
<br />
can anybody explain me how the radial basis function works?<br />
<br />
I can configure the parameters MinInfluenceField, MaxInfluenceField and Background, but how do they exaclty influence the radial basis function?<br />
<br />
When I look at the theory of the radial basis function there are many different functions that can be used, but I can't find anything about something like a MinInfluenceField, MaxInfluenceField or Background.<br />
<br />
And what is the actual basis function used here?]]></content:encoded>
		</item>
		<item>
			<title><![CDATA[Wrong classification with RProp in neural network demo]]></title>
			<link>http://mitov.com/forum/thread-309.html</link>
			<pubDate>Tue, 24 Apr 2012 11:37:24 -0400</pubDate>
			<guid isPermaLink="false">http://mitov.com/forum/thread-309.html</guid>
			<description><![CDATA[Hi all,<br />
<br />
am I the only one who gets wrong classifications with the RProp training method in the neural network demo.<br />
<br />
In 28% of the cases the classification ist wrong. Backpropagation works fine for me. I attached some screenshots with the wrong classification.<br />
<br />
Is this an error in the component or is this normal for RProp?<br />
<br />
greetings<br />
Benjamin<br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=2" target="_blank">RProp_Error.PNG</a> (Size: 40.89 KB / Downloads: 1)
<!-- end: postbit_attachments_attachment --><br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=3" target="_blank">RProp_Error2.PNG</a> (Size: 39.28 KB / Downloads: 0)
<!-- end: postbit_attachments_attachment --><br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=4" target="_blank">RProp_Error3.PNG</a> (Size: 40.19 KB / Downloads: 0)
<!-- end: postbit_attachments_attachment -->]]></description>
			<content:encoded><![CDATA[Hi all,<br />
<br />
am I the only one who gets wrong classifications with the RProp training method in the neural network demo.<br />
<br />
In 28% of the cases the classification ist wrong. Backpropagation works fine for me. I attached some screenshots with the wrong classification.<br />
<br />
Is this an error in the component or is this normal for RProp?<br />
<br />
greetings<br />
Benjamin<br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=2" target="_blank">RProp_Error.PNG</a> (Size: 40.89 KB / Downloads: 1)
<!-- end: postbit_attachments_attachment --><br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=3" target="_blank">RProp_Error2.PNG</a> (Size: 39.28 KB / Downloads: 0)
<!-- end: postbit_attachments_attachment --><br /><!-- start: postbit_attachments_attachment -->
<br /><img src="images/attachtypes/image.gif" border="0" alt=".png" />&nbsp;&nbsp;<a href="attachment.php?aid=4" target="_blank">RProp_Error3.PNG</a> (Size: 40.19 KB / Downloads: 0)
<!-- end: postbit_attachments_attachment -->]]></content:encoded>
		</item>
		<item>
			<title><![CDATA[General questions about ILKNearestNeighbor]]></title>
			<link>http://mitov.com/forum/thread-287.html</link>
			<pubDate>Tue, 10 Apr 2012 08:44:39 -0400</pubDate>
			<guid isPermaLink="false">http://mitov.com/forum/thread-287.html</guid>
			<description><![CDATA[Hi,<br />
<br />
I started working with the ILKNearestNeighbor unit. I looked at the wiki and I modified the example to learn a little about its functionality.<br />
<br />
I have two questions about ILKNearestNeighbor:<br />
Why is the example not working with k &lt; 6?<br />
Is it possible to change the distance function? Found nothing about that in the wiki...]]></description>
			<content:encoded><![CDATA[Hi,<br />
<br />
I started working with the ILKNearestNeighbor unit. I looked at the wiki and I modified the example to learn a little about its functionality.<br />
<br />
I have two questions about ILKNearestNeighbor:<br />
Why is the example not working with k &lt; 6?<br />
Is it possible to change the distance function? Found nothing about that in the wiki...]]></content:encoded>
		</item>
		<item>
			<title><![CDATA[General Info]]></title>
			<link>http://mitov.com/forum/thread-91.html</link>
			<pubDate>Wed, 04 Jan 2012 08:36:57 -0500</pubDate>
			<guid isPermaLink="false">http://mitov.com/forum/thread-91.html</guid>
			<description><![CDATA[Good Day <br />
<br />
I know there is not much info on the Intelligence Lab at present but i believe this will be updated soon.<br />
<br />
Until then, can anyone recommend some general information on the Neural Network Concept so we have a better understanding when it comes time to do major development.<br />
<br />
Any info would be greatly appreciated!<br />
<br />
Kind Regards<br />
<br />
Donovan Boddy]]></description>
			<content:encoded><![CDATA[Good Day <br />
<br />
I know there is not much info on the Intelligence Lab at present but i believe this will be updated soon.<br />
<br />
Until then, can anyone recommend some general information on the Neural Network Concept so we have a better understanding when it comes time to do major development.<br />
<br />
Any info would be greatly appreciated!<br />
<br />
Kind Regards<br />
<br />
Donovan Boddy]]></content:encoded>
		</item>
		<item>
			<title><![CDATA[how to make speech recognition]]></title>
			<link>http://mitov.com/forum/thread-75.html</link>
			<pubDate>Sat, 17 Dec 2011 23:56:02 -0500</pubDate>
			<guid isPermaLink="false">http://mitov.com/forum/thread-75.html</guid>
			<description><![CDATA[I will make a speech recognition system the which can<br />
Recognize vowels (A, I, U, E, O), using a backpropagation<br />
neural network system. <br />
I use the components IntelligenceLab 5.0.1 for Delphi, AudioLab 5.0.1 for Delphi and I am using Delphi XE2 program.<br />
<br />
I encounter problems when applying it, I use Openwire Editor and<br />
components that I use:<br />
1. ALAudio1  (to take sound from a microphone)<br />
2. ALSpectrum1  (spectrum components)<br />
3. ILNeuralNetworkBackpropTrain1  (backprop training algorithm components)<br />
4. ILNeuralNetwork1  (neural network clasifier)<br />
<br />
to the output pins of the component spectrum can I connect with the<br />
input of clasifier neuralnetwork1, but I can not connect the output<br />
spectrum with component input ILNeuralNetworkBackpropTrain1.<br />
My question whether the steps that I am doing is correct or whether<br />
there is still a mistake? or need additional program listings?<br />
<br />
greetings Hymae]]></description>
			<content:encoded><![CDATA[I will make a speech recognition system the which can<br />
Recognize vowels (A, I, U, E, O), using a backpropagation<br />
neural network system. <br />
I use the components IntelligenceLab 5.0.1 for Delphi, AudioLab 5.0.1 for Delphi and I am using Delphi XE2 program.<br />
<br />
I encounter problems when applying it, I use Openwire Editor and<br />
components that I use:<br />
1. ALAudio1  (to take sound from a microphone)<br />
2. ALSpectrum1  (spectrum components)<br />
3. ILNeuralNetworkBackpropTrain1  (backprop training algorithm components)<br />
4. ILNeuralNetwork1  (neural network clasifier)<br />
<br />
to the output pins of the component spectrum can I connect with the<br />
input of clasifier neuralnetwork1, but I can not connect the output<br />
spectrum with component input ILNeuralNetworkBackpropTrain1.<br />
My question whether the steps that I am doing is correct or whether<br />
there is still a mistake? or need additional program listings?<br />
<br />
greetings Hymae]]></content:encoded>
		</item>
	</channel>
</rss>