Targeted variation of machine learning input data

1. A machine learning system, comprising:

a computing device including an artificial neural network executing a machine learning algorithm;

wherein the computing device includes at least one processor and memory storing computer-executable instructions that, when executed by the at least one processor cause the computing device to:

determine input data for the artificial neural network;

identify one or more biases in the input data, the one or more biases including prioritizing quantitative values in the input data over Boolean values and strings in the input data;

group, based on the one or more biases, the input data into one or more input data groups, wherein the one or more input data groups are groups of quantitative values, Boolean values and strings;

apply, based on the one or more biases, weights to the one or more input data groups to generate weighted one or more input data groups by associating each of the one or more input data groups with a level of importance;

transmit, to the artificial neural network, the weighted one or more input data groups;

train the artificial neural network to generate output data based on the weighted one or more input data groups received via one or more input nodes;

determine, based on the output data, one or more output biases in the output data; and

apply, based on the one or more output biases in the output data, second weights to the one or more input data groups.

2. The machine learning system of claim 1 , wherein the instructions further cause the computing device to:

group, based on the one or more output biases in the output data, the input data into one or more second input data groups.

3. The machine learning system of claim 1 , wherein grouping the input data includes group the input data by discarding one or more portions of the input data.

4. The machine learning system of claim 1 , wherein applying weights to the one or more input data groups includes converting a first type of data into a second type of data.

5. A method comprising:

receiving, by a computing device having at least one processor and memory storing computer-executable instructions, input data for an artificial neural network executing a machine learning algorithm;

identifying, by the at least one processor, one or more biases in the input data, the one or more biases including prioritizing quantitative values in the input data over Boolean values and strings in the input data;

grouping, by the at least one processor and based on the one or more biases, the input data into one or more input data groups, wherein the one or more input data groups are groups of quantitative values, Boolean values and strings;

applying, by the at least one processor and based on the one or more biases, weights to the one or more input data groups to generate weighted one or more input data groups by associating each of the one or more input data groups with a level of importance;

transmitting, by the at least one processor and to the artificial neural network, the weighted one or more input data groups;

training, by the at least one processor, the artificial neural network to generate output data based on the weighted one or more input data groups received via one or more input nodes;

determining, by the at least one processor and based on the output data, one or more output biases in the output data; and

applying, by the at least one processor and based on the one or more output biases in the output data, second weights to the one or more input data groups.

6. The method of claim 5 , further comprising:

grouping, by the at least one processor and based on the one or more output biases in the output data, the input data into one or more second input data groups.

7. The method of claim 5 , wherein grouping the input data includes discarding one or more portions of the input data.

8. The method of claim 5 , wherein applying weights to the one or more input data groups includes converting a first type of data into a second type of data.

9. An apparatus comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the apparatus to:

receive input data for an artificial neural network executing a machine learning algorithm;

identify one or more biases in the input data, the one or more biases including prioritizing quantitative values in the input data over Boolean values and strings in the input data;

group, based on the one or more biases, the input data into one or more input data groups, wherein the one or more input data groups are groups of quantitative values, Boolean values and strings ;

apply, based on the one or more biases, weights to the one or more input data groups to generate weighted one or more input data groups by associating each of the one or more input data groups with a level of importance;

transmit, to the artificial neural network, the weighted one or more input data groups;

train the artificial neural network to generate output data based on the weighted one or more input data groups received via one or more input nodes;

determine, based on the output data, one or more output biases in the output data; and

apply, based on the one or more output biases in the output data, second weights to the one or more input data groups.

10. The apparatus of claim 9 , wherein the memory stores instructions that, when executed by the one or more processors, further cause the apparatus to:

group, based on the one or more output biases in the output data, the input data into one or more second input data groups.

11. The apparatus of claim 9 , wherein grouping the input data comprises discarding one or more portions of the input data.

12. The apparatus of claim 9 , wherein applying weights to the one or more input data groups comprises converting a first type of data into a second type of data.