Neural Networks - Revision history

Thwien: /* Questions and Answers */

2008-12-18T12:29:39Z

Questions and Answers

← Older revision		Revision as of 12:29, 18 December 2008
Line 25:		Line 25:
	** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.		** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.
	** '''''Network traffic estimation:''''' Prediction how much traffic is to be expected for a sever cluster to a specific time of a day.		** '''''Network traffic estimation:''''' Prediction how much traffic is to be expected for a sever cluster to a specific time of a day.
			** '''''CRAP (Change Risk Analysis and Prediction:''''' Analysing of error patterns in programming structures to predict possible errors in selected modules or methods could happen while refactoring software based on experiences the neural network did while testing the software.
	** '''''Cancer detection:''''' Cancer detection on several medical test results.		** '''''Cancer detection:''''' Cancer detection on several medical test results.
	** '''''Air traffic control'''''		** '''''Air traffic control'''''

Thwien: /* Questions and Answers */

2008-07-03T13:47:15Z

Questions and Answers

← Older revision		Revision as of 13:47, 3 July 2008
Line 34:		Line 34:
	** '''''Weather forecast'''''		** '''''Weather forecast'''''
	** '''''Music composition:''''' A neural network can learn rules of composition and create own music based on them. (But until today I have never heard neural composed music.)		** '''''Music composition:''''' A neural network can learn rules of composition and create own music based on them. (But until today I have never heard neural composed music.)
			** '''''Compiler optimization:''''' It's planned to extend the compiler ''gcc'' by neural networks to learn how to compile code with higher performance without manually optimizing the source code.

	* '''How to train an artificial neural network?'''		* '''How to train an artificial neural network?'''

Thwien: /* Questions and Answers */

2008-01-22T21:06:53Z

Questions and Answers

← Older revision		Revision as of 21:06, 22 January 2008
Line 25:		Line 25:
	** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.		** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.
	** '''''Network traffic estimation:''''' Prediction how much traffic is to be expected for a sever cluster to a specific time of a day.		** '''''Network traffic estimation:''''' Prediction how much traffic is to be expected for a sever cluster to a specific time of a day.
			** '''''Cancer detection:''''' Cancer detection on several medical test results.
	** '''''Air traffic control'''''		** '''''Air traffic control'''''
	** '''''Robot control'''''		** '''''Robot control'''''

Thwien: /* Questions and Answers */

2008-01-18T21:27:41Z

Questions and Answers

← Older revision		Revision as of 21:27, 18 January 2008
Line 24:		Line 24:
	** '''''Stock market:''''' Prediction of stock values by market rules and psychological reactions.		** '''''Stock market:''''' Prediction of stock values by market rules and psychological reactions.
	** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.		** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.
			** '''''Network traffic estimation:''''' Prediction how much traffic is to be expected for a sever cluster to a specific time of a day.
	** '''''Air traffic control'''''		** '''''Air traffic control'''''
	** '''''Robot control'''''		** '''''Robot control'''''

Thwien: /* Questions and Answers */

2008-01-13T15:03:04Z

Questions and Answers

← Older revision		Revision as of 15:03, 13 January 2008
Line 14:		Line 14:

	* '''What are examples for technical use of artificial neural networks?'''		* '''What are examples for technical use of artificial neural networks?'''
	** '''''Prediction of power consumption:''''' The local electricity factory in Duesseldorf, Germany is using a multilayer perceptron for daily prediction of power consumption in the city referring to temperature, humidity, weekday, and others.		** '''''Prediction of power consumption:''''' The local electricity factory in Duesseldorf, Germany is using a [[multilayer perceptron]] for daily prediction of power consumption in the city referring to temperature, humidity, weekday, and others.
	** '''''Prediction of selling products:''''' Prediction on how many articles of a product in a supermarket will be sold in one week. With this information it is possible to optimize ordering and storage.		** '''''Prediction of selling products:''''' Prediction on how many articles of a product in a supermarket will be sold in one week. With this information it is possible to optimize ordering and storage.
	** '''''Prediction of incoming calls to a call centre:''''' Prediction on the among of daily calls in a call centre to plan how many co-workers have to work that day.		** '''''Prediction of incoming calls to a call centre:''''' Prediction on the among of daily calls in a call centre to plan how many co-workers have to work that day.

Thwien: /* Questions and Answers */

2008-01-13T13:57:48Z

Questions and Answers

← Older revision		Revision as of 13:57, 13 January 2008
Line 7:		Line 7:
	* '''What are the differences between a biological and an artificial neural network?'''		* '''What are the differences between a biological and an artificial neural network?'''

	Between both networks are huge differences. The neurons of a biological neural networks works parallel massively and such a network has billions or trillions of neurons connected massively. In comparison the artificial neural networks has just a few count of neurons depending on the among of data to associate by the network, e.g. 20 neurons. Such an artificial network is not connected by accident, but as a controlled and planned layer structure, where is an input layer, one or a few hidden layers and an output layer (if a multilayer perceptron topology is used). As an artificial neural network is a computer software it works sequentially and therefore there is a big difference to a biological neural network, of course.		Between both networks are huge differences. The neurons of a biological neural networks works parallel massively and such a network has billions or trillions of neurons connected massively. In comparison the artificial neural networks has just a few count of neurons depending on the among of data to associate by the network, e.g. 20 neurons. Such an artificial network is not connected by accident, but as a controlled and planned layer structure, where is an input layer, one or a few hidden layers and an output layer (if a [[multilayer perceptron]] topology is used). As an artificial neural network is a computer software it works sequentially and therefore there is a big difference to a biological neural network, of course.

	* '''What is the benefit of an artificial neural network?'''		* '''What is the benefit of an artificial neural network?'''
Line 51:		Line 51:
	* '''Which kinds of artificial neural networks are existing?'''		* '''Which kinds of artificial neural networks are existing?'''

	There are several. For example: multilayer perceptron, self-organizing maps (Kohonen net), Hoppfield net, generic maps, and further more.		There are several. For example: [[multilayer perceptron]], self-organizing maps (Kohonen net), Hoppfield net, generic maps, and further more.

	== Information about Neural Networks ==		== Information about Neural Networks ==

Thwien: /* Questions and Answers */

2008-01-10T10:24:12Z

Questions and Answers

← Older revision		Revision as of 10:24, 10 January 2008
Line 14:		Line 14:

	* '''What are examples for technical use of artificial neural networks?'''		* '''What are examples for technical use of artificial neural networks?'''
			** '''''Prediction of power consumption:''''' The local electricity factory in Duesseldorf, Germany is using a multilayer perceptron for daily prediction of power consumption in the city referring to temperature, humidity, weekday, and others.

			** '''''Prediction of selling products:''''' Prediction on how many articles of a product in a supermarket will be sold in one week. With this information it is possible to optimize ordering and storage.
	There are a few very interesting uses of artificial neural networks. For example, the '''local electricity factory in Duesseldorf, Germany''' is using a multilayer perceptron for daily prediction of power use in the city referring to temperature, humidity, weekday, etc. Another example is the prediction '''how many articles of a product in a supermarket will be sold''' in one week. With this information it is possible to optimize ordering and storage. Or such a network is used to '''predict the among of daily calls in a call centre''' to plan how many co-workers have to work that day. Or the German Post is using neural networks in '''recognition of post codes (PLZ)''' written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again. Artificial neural networks are also used to '''find a numeric solution of difficult mathematical functions'''. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule. A '''general classification''' can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on. Another example is a '''network router which learns the fastest routing''' of internet packets. Therefore this router can optimize routing decisions. Also neural networks are used to '''detect spam mails''' in mail clients or on mail servers. Also in the this areas neural networks are used: '''Stock market, credit assignments, air traffic control, robot control, game strategy control, noise tolerance of analogue modems, scheduling buses and trams, air planes and elevators, optimization of traffic flows, weather forecast, music composition.'''
			** '''''Prediction of incoming calls to a call centre:''''' Prediction on the among of daily calls in a call centre to plan how many co-workers have to work that day.
			** '''''Post code recognition:''''' The German Post is using neural networks in recognition of post codes (PLZ) written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again.
			** '''''Solving mathematical functions:''''' Artificial neural networks are also used to find a numeric solution of difficult mathematical functions. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule.
			** '''''Classification:''''' A general classification can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on.
			** '''''Intelligent network routing:''''' A network router which learns the fastest routing of internet packets could be solved by using neural networks. Therefore this router can optimize routing decisions.
			** '''''Detecting spam mails:''''' Neural networks are also used to detect spam mails in mail clients or on mail servers.
			** '''''Stock market:''''' Prediction of stock values by market rules and psychological reactions.
			** '''''Credit assignments:''''' Determine bad credit risk of bank customers by including different aspects.
			** '''''Air traffic control'''''
			** '''''Robot control'''''
			** '''''Game strategy control'''''
			** '''''Noise tolerance of analogue modems'''''
			** '''''Scheduling buses, trams, air planes and elevators'''''
			** '''''Optimization of traffic flows'''''
			** '''''Weather forecast'''''
			** '''''Music composition:''''' A neural network can learn rules of composition and create own music based on them. (But until today I have never heard neural composed music.)

	* '''How to train an artificial neural network?'''		* '''How to train an artificial neural network?'''

Thwien at 19:08, 9 January 2008

2008-01-09T19:08:40Z

← Older revision		Revision as of 19:08, 9 January 2008
Line 15:		Line 15:
	* '''What are examples for technical use of artificial neural networks?'''		* '''What are examples for technical use of artificial neural networks?'''

	There are a few very interesting uses of artificial neural networks. For example, the '''local electricity factory in Duesseldorf, Germany''' is using a multilayer perceptron for daily prediction of power use in the city referring to temperature, humidity, weekday, etc. Another example is the prediction '''how many articles of a product in a supermarket will be sold''' in one week. With this information it is possible to optimize ordering and storage. Or such a network is used to '''predict the among of daily calls in a call centre''' to plan how many co-workers have to work that day. Or the German Post is using neural networks in '''recognition of post codes (PLZ)''' written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again. Artificial neural networks are also used to '''find a numeric solution of difficult mathematical functions'''. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule. A '''general classification''' can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on. Another example is a '''network router which learns the fastest routing''' of internet packets. Therefore this router can optimize routing decisions. Also neural networks are used to '''detect spam mails''' in mail clients or on mail servers. Also in the this areas neural networks are used: '''Stock market, credit assignments, air traffic control, robot control, game strategy control, noise tolerance of analogue modems, scheduling buses and trams, air planes and elevators, optimization of traffic flows, weather forecast.'''		There are a few very interesting uses of artificial neural networks. For example, the '''local electricity factory in Duesseldorf, Germany''' is using a multilayer perceptron for daily prediction of power use in the city referring to temperature, humidity, weekday, etc. Another example is the prediction '''how many articles of a product in a supermarket will be sold''' in one week. With this information it is possible to optimize ordering and storage. Or such a network is used to '''predict the among of daily calls in a call centre''' to plan how many co-workers have to work that day. Or the German Post is using neural networks in '''recognition of post codes (PLZ)''' written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again. Artificial neural networks are also used to '''find a numeric solution of difficult mathematical functions'''. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule. A '''general classification''' can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on. Another example is a '''network router which learns the fastest routing''' of internet packets. Therefore this router can optimize routing decisions. Also neural networks are used to '''detect spam mails''' in mail clients or on mail servers. Also in the this areas neural networks are used: '''Stock market, credit assignments, air traffic control, robot control, game strategy control, noise tolerance of analogue modems, scheduling buses and trams, air planes and elevators, optimization of traffic flows, weather forecast, music composition.'''

	* '''How to train an artificial neural network?'''		* '''How to train an artificial neural network?'''

Thwien at 19:07, 9 January 2008

2008-01-09T19:07:37Z

← Older revision		Revision as of 19:07, 9 January 2008
Line 31:		Line 31:
	* '''Does have artificial neural networks has disadvantages?'''		* '''Does have artificial neural networks has disadvantages?'''

	Yes, there are a few problems. Today are existing a number of neural network topologies. Each topology is perfect to solve a special kind of problems. Neural networks are slow in training, because training can be several ten thousands of loops. And a neural network can be overfit if it is overtrained.		Yes, there are a few problems. Today are existing a number of neural network topologies. Each topology is perfect to solve a special kind of problems. Neural networks are slow in training, because training can be several ten thousands of loops. And a neural network can be overfit if it is overtrained. A problem is also the extrapolation of data if data is not presented properly to the neural network.

	* '''Which kinds of artificial neural networks are existing?'''		* '''Which kinds of artificial neural networks are existing?'''

Thwien at 19:05, 9 January 2008

2008-01-09T19:05:26Z

← Older revision		Revision as of 19:05, 9 January 2008
Line 15:		Line 15:
	* '''What are examples for technical use of artificial neural networks?'''		* '''What are examples for technical use of artificial neural networks?'''

	There are a few very interesting uses of artificial neural networks. For example, the '''local electricity factory in Duesseldorf, Germany''' is using a multilayer perceptron for daily prediction of power use in the city referring to temperature, humidity, weekday, etc. Another example is the prediction '''how many articles of a product in a supermarket will be sold''' in one week. With this information it is possible to optimize ordering and storage. Or such a network is used to '''predict the among of daily calls in a call centre''' to plan how many co-workers have to work that day. Or the German Post is using neural networks in '''recognition of post codes (PLZ)''' written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again. Artificial neural networks are also used to '''find a numeric solution of difficult mathematical functions'''. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule. A '''general classification''' can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on. Another example is a '''network router which learns the fastest routing''' of internet packets. Therefore this router can optimize routing decisions. Also neural networks are used to '''detect spam mails''' in mail clients or on mail servers.		There are a few very interesting uses of artificial neural networks. For example, the '''local electricity factory in Duesseldorf, Germany''' is using a multilayer perceptron for daily prediction of power use in the city referring to temperature, humidity, weekday, etc. Another example is the prediction '''how many articles of a product in a supermarket will be sold''' in one week. With this information it is possible to optimize ordering and storage. Or such a network is used to '''predict the among of daily calls in a call centre''' to plan how many co-workers have to work that day. Or the German Post is using neural networks in '''recognition of post codes (PLZ)''' written on letters by computer or manually. After recognition a computer readable code is printed on the letter so further recognition is not necessary again. Artificial neural networks are also used to '''find a numeric solution of difficult mathematical functions'''. The network can associate the output of a function getting one or more unknown inputs. This is just possible by learning known inputs and outputs. The artificial network has an "idea" of the mathematical rule. A '''general classification''' can be done also with a neural network. For example such a network can detect by an image if the person displayed is looking left or right, is laughing or not, male or female, and so on. Another example is a '''network router which learns the fastest routing''' of internet packets. Therefore this router can optimize routing decisions. Also neural networks are used to '''detect spam mails''' in mail clients or on mail servers. Also in the this areas neural networks are used: '''Stock market, credit assignments, air traffic control, robot control, game strategy control, noise tolerance of analogue modems, scheduling buses and trams, air planes and elevators, optimization of traffic flows, weather forecast.'''

	* '''How to train an artificial neural network?'''		* '''How to train an artificial neural network?'''