Top contributor recommendation for cloud analytics

1. A computer-implemented method, the method comprising:

receiving all numeric values of a specified target measure column of a first dataset including a plurality of records, the first dataset having a plurality of columns including the specified target measure column and a plurality of non-numeric, dimension columns for the records of the first dataset;

discretizing each of the received numeric values of the target specified measure column into a plurality of bins, the plurality of bins being a pre-defined value, each of the bins having an equal interval width, and each of the bins having an index number;

generating a bin index column that contains a determined bin index number for each numeric value of the specified target measure column of each record in the first dataset;

determining a bin probability that represents a probability of each of the numeric values of the specified target measure column of the first dataset being in each of the bins based on the generated bin index column;

determining, based on the generated bin index column, a dimension score for each dimension column of the first dataset in each bin;

forming, based on the determined dimension score, a dimension score matrix for the first dataset; and

saving the determined bin probability and the dimension score matrix as metadata for the first dataset.

2. The method of claim 1 , further comprising:

receiving an indication of one or more specified variable values, the specified variable values each being a value selected from one or more of the plurality of non-numeric, dimension columns of the first dataset;

retrieving values in the bin index column of the records related to the specified one or more variable values;

determining a second bin probability, based on the retrieved bin index column values of the records related to the specified one or more variable values, that represents a probability of a value in the retrieved bin index column of the records related to the specified one or more variable values being in each of the bins;

deriving, by a first calculation, an approximated dimension score vector of a subset of the first dataset related to the specified one or more variable values based on the determined bin probability of a value being in each of the bins for the first dataset, the determined dimension score matrix for the first dataset, and the determined second bin probability; and

saving an output of the approximated dimension score vector.

3. The method of claim 2 , wherein a weighted dimension score vector is used in deriving the approximated dimension score vector.

4. The method of claim 2 , further comprising:

determining a first set of dimensions in the approximated dimension score vector having a highest value relative to each other, the number of dimension in the set being predefined; and

presenting the determined set of dimensions to a user.

5. The method of claim 4 , further comprising:

determining a second set of dimensions in the approximated dimension score vector having a highest value, the number of dimension in the second set being predefined and fewer than the number of dimensions in the first set; and

presenting the determined second set of dimensions to a user.

6. The method of claim 2 , wherein the first calculation to derive the approximated dimension score vector is substituted with a second calculation based on at least the determined bin probability for the first dataset and the determined dimension score vector for the first dataset.

7. A non-transitory, computer readable medium having executable instructions stored therein that, when executed by a computer processor cause the processor to perform a method, the method comprising:

receiving all numeric values of a specified target measure column of a first dataset including a plurality of records, the first dataset having a plurality of columns including the target measure column and a plurality of non-numeric, dimension columns for the records of the first dataset;

discretizing each of the received numeric values of the target specified measure column into a plurality of bins, the plurality of bins being a pre-defined value, each of the bins having an equal interval width, and each of the bins having an index number;

generating a bin index column that contains a determined bin index number for each numeric value of the specified target measure column of each record in the first dataset;

determining a bin probability that represents a probability of each of the numeric values of the specified target measure column of the first dataset in each of the bins based on the generated bin index column;

determining, based on the generated bin index column, a dimension score for each dimension column of the first dataset in each bin;

forming, based on the determined dimension score, a dimension score matrix for the first dataset; and

saving the determined bin probability and the dimension score matrix as metadata for the first dataset.

8. The medium of claim 7 , further comprising:

receiving an indication of one or more specified variable values, the specified variable values each being a value selected from one or more of the plurality of non-numeric, dimension columns of the first dataset;

retrieving values in the bin index column of the records related to the specified one or more variable values;

determining a second bin probability, based on the retrieved bin index column values of the records related to the specified one or more variable values, that represents a probability of a value in the retrieved bin index column of the records related to the specified one or more variable values being in each of the bins;

deriving, by a first calculation, an approximated dimension score vector of the first dataset related to the specified one or more variable values based on the determined bin probability of a value being in each of the bins for the first dataset, the determined dimension score matrix for the first dataset, and the determined second bin probability; and

saving an output of the approximated dimension score vector.

9. The medium of claim 8 , wherein a weighted dimension score vector is used in deriving the approximated dimension score vector.

10. The medium of claim 8 , further comprising:

determining a first set of dimensions in the approximated dimension score vector having a highest relative value, the number of dimension in the first set being predefined; and

presenting the determined first set of dimensions to a user.

11. The medium of claim 10 , further comprising:

determining a second set of dimensions in the approximated dimension score vector having a highest value, the number of dimension in the second set being predefined and fewer than the number of dimensions in the first set; and

presenting the determined second set of dimensions to a user.

12. The medium of claim 8 , wherein the first calculation to derive the approximated dimension score vector is substituted with a second calculation based on at least the determined bin probability for the first dataset and the determined dimension score vector for the first dataset.

13. A system, the system comprising:

a computer processor, and

computer memory, coupled to the computer processor, storing instructions that, when executed by the computer processor cause the computer processor to:

receive all numeric values of a specified target measure column of a first dataset including a plurality of records, the first dataset having a plurality of columns including the target measure column and a plurality of non-numeric, dimension columns for the records of the first dataset;

discretize each of the received numeric values of the specified target measure column into a plurality of bins, the plurality of bins being a pre-defined value, each of the bins having an equal interval width, and each of the bins having an index number;

generate a bin index column that contains a determined bin index number for each numeric value of the specified target measure column of each record in the first dataset;

determine a bin probability that represents a probability of each of the numeric values of the specified target measure column of the first dataset being in each of the bins based on the generated bin index column;

determine, based on the generated bin index column, a dimension score for each dimension column of the first dataset in each bin;

form, based on the determined dimension score, a dimension score matrix for the first dataset; and

save the determined bin probability and the dimension score matrix as metadata for the first dataset.

14. The system of claim 13 , further comprising:

receiving an indication of one or more specified variable values, the specified variable values each being a value selected from one or more of the plurality of non-numeric, dimension columns of the first dataset;

retrieving values in the bin index column of the records related to the specified one or more variable values;

determining a second bin probability, based on the retrieved bin index column values of the records related to the specified one or more variable values, that represents a probability of a value in the retrieved bin index column of the records related to the specified one or more variable values being in each of the bins;

deriving, by a first calculation, an approximated dimension score vector of the first dataset related to the specified one or more variable values based on the determined bin probability of a value being in each of the bins for the first dataset, the determined dimension score matrix for the first dataset, and the determined second bin probability; and

saving an output of the approximated dimension score vector.

15. The system of claim 14 , wherein a weighted dimension score vector is used in deriving the approximated dimension score vector.

16. The system of claim 14 , further comprising:

determining a first set of dimensions in the approximated dimension score vector having a highest value relative to each other, the number of dimension in the first set being predefined; and

presenting the determined first set of dimensions to a user.

17. The system of claim 16 , further comprising:

determining a second set of dimensions in the approximated dimension score vector having a highest value, the number of dimension in the second set being predefined and fewer than the number of dimensions in the first set; and

presenting the determined second set of dimensions to a user.

18. The system of claim 14 , wherein the first calculation to derive the approximated dimension score vector is substituted with a second calculation based on at least the determined bin probability for the first dataset and the determined dimension score vector for the first dataset.