Facilitating selection of categorical features in machine learning

1. A computer-executable method, comprising:

receiving, by a system comprising a processor, input data including a plurality of categorical variables and an output data including a single target numerical variable, wherein the plurality of categorical variables comprises a first categorical variable and a second categorical variable;

building, by the system, a predictive model configured to:

estimate significance of each categorical variable to the single target numerical variable by converting each categorical variable to each numerical variable, wherein the converting comprises:

determining, by the system, a list of unique values of each categorical variable;

calculating, by the system, respective mean values of the single target numerical variable for unique values in the list of unique values of each categorical variable; and

sorting, by the system, the list of unique values by the respective mean values, resulting in each sorted list, wherein each sorted list corresponds to each numerical variable;

calculate, by the system, respective derivatives of the respective mean values in each sorted list considering the respective mean values as a function and a number of the respective mean values in each sorted list as an independent variable;

determine, by the system, a minimum derivative value over a sorted list of the first categorical variable, wherein the minimum derivative value indicates a most significant categorical feature;

determine, by the system, a maximum derivative value over a sorted list of the second categorical variable, wherein the maximum derivative value indicates a least significant categorical feature;

output, by the system, an output data including the first categorical variable and excluding the second categorical variable; and

generating, by the system, one or more predictive models using the output data to automatically implement one or more actions in a network process operated by network equipment with respect to the single target numerical variable.

2. The method of claim 1 , wherein the determining the minimum derivative value further comprises:

determining a respective slope of a respective mean value; and

determining a minimum value of the respective slope of the respective mean value.

3. The method of claim 1 , wherein the calculating of the respective derivatives comprises:

calculating the respective derivatives as a slope based on determining a ratio of a mean values change to an item number change between any two items on the sorted list.

4. The method of claim 1 , further comprising:

calculating, by the system, a quadratic mean value of the single target numerical variable for each unique value of each categorical variable.

5. The method of claim 1 , further comprising:

calculating, by the system, an arithmetic mean value of the single target numerical variable for each unique value of each categorical variable.

6. The method of claim 1 , further comprising:

calculating, by the system, a geometric mean value of the single target numerical variable for each unique value of each categorical variable.

7. The method of claim 1 , further comprising:

calculating, by the system, a weighted arithmetic mean value of the single target numerical variable for each unique value of each categorical variable.

8. The method of claim 1 , further comprising:

calculating, by the system, a weighted arithmetic mean value of the single target numerical variable normalized by a normalization process.

9. A system, comprising:

a processor; and

a memory that stores executable instructions that when executed by the processor, facilitate performance of operations, comprising:

receiving input data including a plurality of categorical variables and a single target numerical variable, wherein the plurality of categorical variables comprises a first categorical variable and a second categorical variable;

building a predictive model configured to:

estimate a respective significance of the plurality of categorical variables by converting each categorical variable to each numerical variable, wherein the converting comprises:

determining, by the system, a list of unique values of each of the plurality of categorical variables;

calculating, by the system, respective mean values of the single target numerical variable for unique values in the list of unique values of each categorical variable; and

sorting, by the system, the list of unique values by the respective mean values, resulting in each sorted list, wherein each sorted list corresponds to each numeral variable;

calculate, by the system, respective derivatives of the respective mean values in each sorted list considering the respective mean values as a function and a number of the respective mean values in each sorted list as an independent variable;

determine, by the system, a minimum derivative value over a sorted list of the first categorical variable, wherein the minimum derivative value indicates a most significant categorical feature;

determine, by the system, a maximum derivative value over a sorted list of the second categorical variable, wherein the maximum derivative value indicates a least significant categorical feature; and

output an output data including the first categorical variable and excluding the second categorical variable; and

generating one or more predictive models using the output data to automatically implement one or more actions in a network process operated by network equipment with respect to the single target numerical variable.

10. The system of claim 9 , wherein the determining the minimum derivative value further comprises:

determining a respective slope of a respective mean value; and

determining a minimum value of the respective slope of the respective mean value.

11. The system of claim 9 , wherein the operations further comprises determining a variable significance value of the first categorical variable.

12. The system of claim 9 , wherein the operations further comprises:

calculating a quadratic mean value of the single target numerical variable for each unique value of each categorical variable.

13. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:

receiving input data including a plurality of categorical variables and an output data including a single target numerical variable, wherein the plurality of categorical variables comprises a first categorical variable and a second categorical variable;

building a predictive model configured to:

estimate significance of each categorical variable to the single target numeral variable by converting each categorical variable to each numeral variable, wherein the converting further comprises;

determining a list of unique values of each of the plurality of categorical variables;

calculating respective mean values of the single target numerical variable for unique values in the list of unique values of each categorical variable; and

sorting the list of unique values by the respective mean values, resulting in each sorted list, wherein each sorted list corresponds to each numeral variable;

calculate respective derivatives of the respective mean values in each sorted list considering the respective mean values as a function and a number of the respective mean values in each sorted list as an independent variable;

determine a minimum derivative value over a sorted list of the first categorical variable, wherein the minimum derivative value indicates a most significant categorical feature;

determine a maximum derivative value over a sorted list of the second categorical variable, wherein the maximum derivative value indicates a least significant categorical feature; and

output an output data including the first categorical variable and excluding the second categorical variable; and

generating one or more predictive models using the output data to automatically implement one or more actions in a network process operated by network equipment with respect to the single target numerical variable.

14. The non-transitory machine-readable medium of claim 13 , wherein the determining the minimum derivative value further comprises:

determining a respective slope of a respective mean value; and

determining a minimum value of the respective slope of the respective mean value.

15. The non-transitory machine-readable medium of claim 13 , wherein the calculating of the respective derivatives comprises:

calculating the respective derivatives as a slope based on determining a ratio of a mean values change to an item number change between any two items on the sorted list.

16. The non-transitory machine-readable medium of claim 13 , further comprising:

calculating a quadratic mean value of the single target numerical variable for each unique value of each categorical variable.

17. The non-transitory machine-readable medium of claim 13 , further comprising:

calculating an arithmetic mean value of the single target numerical variable for each unique value of each categorical variable.

18. The non-transitory machine-readable medium of claim 13 , further comprising:

calculating a geometric mean value of the single target numerical variable for each unique value of each categorical variable.

19. The non-transitory machine-readable medium of claim 13 , further comprising:

calculating a weighted arithmetic mean value of the single target numerical variable for each unique value of each categorical variable.

20. The non-transitory machine-readable medium of claim 13 , further comprising:

calculating a weighted arithmetic mean value of the single target numerical variable normalized by a normalization process.