Secret decision tree test apparatus, secret decision tree test system, secret decision tree test method, and program
A secret decision tree test device configured to evaluate a division condition at each of a plurality of nodes of a decision tree when learning of the decision tree is performed by secret calculation, the secret decision tree test device includes a memory; and a processor configured to execute inputting a numerical attribute value vector composed of specific numerical attribute values of items of data included in a data set for learning of the decision tree, a label value vector composed of label values of the items of the data, and a group information vector indicating grouping of the items of the data into the nodes; and calculating, using the numerical attribute value vector, the label value vector, and the group information vector, first to fourth frequencies, to evaluate the division condition using the first to fourth frequencies.
1 . A secret decision tree test device configured to evaluate a division condition at each of a plurality of node of a decision tree when learning of the decision tree is performed by secret calculation, the secret decision tree test device comprising:
a memory; and
a processor configured to execute:
inputting a numerical attribute value vector composed of specific numerical attribute values of items of data included in a data set for learning of the decision tree, the specific numerical attribute values including three or more mutually different numerical attribute values that provide a plurality of threshold candidates, a label value vector composed of label values of the items of data, and a group information vector indicating grouping of the items of the data into the nodes;
calculating, using the numerical attribute value vector, the label value vector, and the group information vector, a first frequency of data belonging to each group, a second frequency of data for each of the label values in said each group, a third frequency of data belonging to a division group obtained by dividing said each group by a division condition indicating comparison between the numerical attribute value and a threshold value selected from the plurality of threshold candidates, and a fourth frequency of data for each of the label values in the division group; and
calculating an evaluation value for evaluating the division conditions using the first frequency, the second frequency, the third frequency, and the fourth frequency.
2 . The secret decision tree test device according to claim 1 , wherein the processor calculates the third frequency and the fourth frequency in each of a plurality of the division conditions for said each group.
3 . The secret decision tree test device according to claim 1 , wherein the processor further executes
rearranging numerical attribute values included in the numerical attribute value vector and label values included in the label value vector in ascending order of the numerical attribute values for said each group indicated in the group information vector;
creating, for each value that can be taken by the label value, a bit vector indicating a position of a label value that matches the value that can be taken by the label value, among the label values included in the label value vector;
calculating a first determination vector for determining a number of data items of numerical attribute values being less than or equal to the threshold value by performing an aggregation function cumulative sum operation of elements included in the bit vector for said each group indicated in the group information vector; and
calculating a second determination vector for determining a number of data items of numerical attribute values being greater than the threshold value by using the first determination vector and the bit vector,
wherein the processor calculates the fourth frequency using the first determination vector and the second determination vector.
4 . The secret decision tree test device according to claim 3 , wherein the processor calculates a total vector which is an element of a total for each group and calculates the second determination vector by subtracting the first determination vector from the total vector by performing an aggregate function summation operation of each element included in the bit vector in accordance with a group represented by the group information vector.
5 . A secret decision tree test system configured to evaluate a division condition at each of a plurality of nodes of a decision tree when learning of the decision tree is performed by secret calculation, the secret decision tree test system comprising:
a computer including a memory and a processor configured to execute:
inputting a numerical attribute value vector composed of specific numerical attribute values of each items of data included in a data set for learning of the decision tree, the specific numerical attribute values including three or more mutually different numerical attribute values that provide a plurality of threshold candidates, a label value vector composed of label values of the items of data, and a group information vector indicating grouping of the items of data into the nodes;
calculating, using the numerical attribute value vector, the label value vector, and the group information vector, a first frequency of data belonging to each group, a second frequency of data for each of the label values in said each group, a third frequency of data belonging to a division group obtained by dividing said each group by a division condition indicating comparison between the numerical attribute value and a threshold value selected from the plurality of threshold candidates, and a fourth frequency of data for each of the label values in the division group; and
calculating an evaluation value for evaluating the division condition using the first frequency, the second frequency, the third frequency, and the fourth frequency.
6 . A non-transitory computer-readable recording medium having computer-readable instructions stored thereon, which when executed, cause a computer to function as the secret decision tree test device according to claim 1 .
7 . A secret decision tree test method of evaluating a division condition at each of a plurality of nodes of a decision tree when learning of the decision tree is performed by secret calculation, executed by a computer including a memory and a processor, the secret decision tree test method comprising:
inputting a numerical attribute value vector composed of specific numerical attribute values of items of data included in a data set for learning of the decision tree, the specific numerical attribute values including three or more mutually different numerical attribute values that provide a plurality of threshold candidates, a label value vector composed of label values of the items of data, and a group information vector indicating grouping of the items of the data into the nodes;
calculating, using the numerical attribute value vector, the label value vector, and the group information vector, a first frequency of data belonging to each group, a second frequency of data for each of the label values in said each group, a third frequency of data belonging to a division group obtained by dividing said each group by a division condition indicating comparison between the numerical attribute value and a threshold value selected from the plurality of threshold candidates, and a fourth frequency of data for each of the label values in the division group; and
calculating an evaluation value for evaluating the division condition using the first frequency, the second frequency, the third frequency, and the fourth frequency.