Device, method and application for establishing a current load level

1. A device for calculating a current load level of a user, comprising:

a mobile end unit having:

at least one signal data generating sensor integrated into the mobile end unit,

a plurality of available applications for use by the user, and

an evaluation unit provided in the mobile end unit or in a central server,

wherein:

the mobile end unit has a further application designed for calculating biometric data about the user, at least from the at least one signal data produced by said at least one sensor, and from user data of said plurality of available applications used by the user, and making the biometric data available to the evaluation unit,

wherein the biometric data from the at least one signal data produced by said at least one sensor, and user data of said plurality of available applications, is divided into a plurality of categories,

wherein category-specific load levels are ascertained by the evaluation unit by means of an arithmetic mean or a weighted mean of features relating to the biometric data pertaining to each category in the plurality of categories,

the evaluation unit is designed for determining the current load level of the user from the biometric data by applying a method carried out in a network of artificial neural networks that includes a plurality of artificial neural networks that interact with each other,

a plurality of processors are arranged on the mobile end unit or on the central server, which are designed for calculating the plurality of artificial neural networks in parallel, and

at least one graphics card with at least one graphics card processor is arranged on the mobile end unit or on the central server and the at least one graphics card processor supports the calculation of the artificial neural networks, and

wherein the determined current load level of the user is displayed to the user via the mobile end unit in the form of a consolidated load level obtained from a combination of category-specific load levels by the evaluation unit forming the arithmetic mean or weighted mean of the category-specific load levels.

2. The device according to claim 1 , wherein the evaluation unit is designed for determining the current load level of the user from the biometric data by using a method that is carried out on the plurality of artificial neural networks, which are designed as a Convolutional Deep Belief Network.

3. The device according to claim 1 , wherein the further application can be stored as a Subscriber Identification Module (SIM) application in the memory area on a SIM card that can be operated in the mobile end unit and can be executed by a separate execution unit integrated on the SIM card.

4. The device according to claim 3 , wherein the evaluation unit is designed for determining the current load level of the user from the biometric data by using a method that is carried out on the plurality of artificial neural networks, which are designed as a Convolutional Deep Belief Network.

5. The device according to claim 1 , wherein the biometric data can be divided into a plurality of categories, wherein the categories relate to one or more of:

sleep,

speech,

motor skills,

social interaction,

economic data,

personal information, and

questionnaire data.

6. The device according to claim 1 , wherein:

each artificial neural network includes two neuron layers, one input layer and one hidden layer,

the input layer includes a plurality of input neurons

the hidden layer includes a plurality of hidden neurons, and

wherein the input neurons can be determined by the biometric data.

7. The device according to claim 6 , wherein the evaluation unit is designed for determining the current load level from at least one output neuron, which is identifiable with at least one hidden neuron of at least one artificial neural network.

8. The device according to claim 6 , wherein:

the evaluation unit is designed for determining the input layer of at least one of the artificial neural networks through the hidden layers of a plurality of other artificial neural networks.

9. The device according to claim 8 , wherein the evaluation unit is designed for determining the current load level from at least one output neuron, which is identifiable with at least one hidden neuron of at least one artificial neural network of the plurality of artificial neural networks.

10. The device according to claim 6 , wherein the evaluation unit cooperates with the plurality of processors and the respective processor is designed for calculating neurons for at least one of the plurality of artificial neural networks.

11. The device according to claim 1 , wherein the at least one sensor integrated into the mobile end unit comprises at least one of:

a gyroscope,

an acceleration sensor, and

a light sensor.

12. A method for calculating a current load level of a user of a mobile end unit, comprising:

starting a further application installed on the mobile end unit so that this is carried out on the mobile end unit,

calculating biometric data of the user by means of the further application, wherein the biometric data is recorded at least from user data that is recorded from using a plurality of applications present and available on the mobile end unit by the user, and calculated from at least one signal data produced by at least one sensor integrated into the mobile end unit,

wherein the biometric data from the at least one signal data is produced by said at least one sensor, and user data of said plurality of available applications, is divided into a plurality of categories,

evaluating the biometric data with an evaluation unit provided in the mobile end unit or in a central server for determining the current load level,

wherein:

category-specific load levels are ascertained by means of an arithmetic mean or a weighted mean of features relating to the biometric data pertaining to each category in the plurality of categories,

the evaluation unit determines the current load level with the aid of a network of artificial neural networks,

the network of artificial neural networks comprises a plurality of artificial neural networks that interact with each other,

the plurality of artificial neural networks calculates in parallel with a plurality of processors,

the plurality of processors is arranged on the mobile end unit or on the central server,

at least one graphics card processor of at least one graphics card supports the calculation of the artificial neural networks,

wherein the at least one graphics card with the at least one graphics card processor is arranged on the mobile end unit or on the central server, and

wherein the determined current load level of the user is displayed to the user via the mobile end unit in the form of a consolidated load level obtained from a combination of category-specific load levels by forming the arithmetic mean or weighted mean of the category-specific load levels.

13. The method according to claim 12 , wherein the evaluation unit trains each artificial neural network on the basis of biometric data and from load levels evaluated for the biometric data determined, which improves the quality of the artificial neural network.

14. The method according to claim 12 , wherein:

the further application, as long as it is carried out on the mobile end unit, verifies that signal data is provided at least by the at least one sensor integrated into the mobile end unit, or that usage data is provided by the at least one application available on the mobile end unit for calculating the biometric data,

the further application calculates the biometric data at least from the provided signal data or the provided usage data during a successful verification and equips this with a time stamp, and

the evaluation unit determines the current load level from the biometric data, the time stamp of which is currently valid.

15. The method according to claim 14 , wherein the evaluation unit trains each artificial neural network on the basis of biometric data and from load levels evaluated for the biometric data determined, which improves the quality of the artificial neural network.

16. A non-transitory computer-operable medium that stores computer-executable code associated with an application for a mobile end unit, wherein the computer-executable code, when executed, causes at least one processor of the end unit to carry out the method according to claim 12 .

17. The device according to claim 1 , wherein for each category, a category-specific ascertainment time interval is defined, and the signal data and use data that are relevant to a category are processed using category-specific time intervals to produce conditioned signal data and conditioned use data.

18. The method according to claim 12 , wherein for each category, a category-specific ascertainment time interval is defined, and the signal data and use data that are relevant to a category are processed using category-specific time intervals to produce conditioned signal data and conditioned use data.