Data processing system and data processing method
A data processing system that calculates a correlation coefficient between multiple variables includes a storage unit in which a table in which multiple correlation coefficients are stored is stored, a first selection unit that makes a selection to determine whether or not calculation of a first correlation coefficient that is a correlation coefficient between a first variable and a second variable is indispensable, a second selection unit that selects a third variable which is a variable that results from storing in the table a second correlation coefficient that is a correlation coefficient representing a relationship with the first variable, and a third correlation coefficient that is a correlation coefficient representing a relationship with the second variable, and a first determination unit that determines whether or not the calculation of the first correlation coefficient is indispensable, based on the second correlation coefficient and the third correlation coefficient.
1. A data processing system that calculates a correlation variable between multiple variables, the data processing system comprising:
a storage unit in which a table in which multiple correlation coefficients are stored is stored;
a first selection unit that makes a selection to determine whether or not calculation of a first correlation coefficient that is a correlation coefficient between a first variable and a second variable is indispensable;
a second selection unit that selects a third variable which is a variable that results from storing in the table a second correlation coefficient that is a correlation coefficient representing a relationship with the first variable, and a third correlation coefficient that is a correlation coefficient representing a relationship with the second variable;
a first determination unit that determines whether or not the calculation of the first correlation coefficient is indispensable, based on the second correlation coefficient and the third correlation coefficient; and
an input unit into which a threshold that is at a lower limit of the first correlation coefficient is input,
wherein, based on the second correlation coefficient and the third correlation coefficient, the first determination unit determines whether or not there is a likelihood that the first correlation coefficient will exceed the threshold.
2. The data processing system according to claim 1 , further comprising:
a calculation unit that calculates the first correlation coefficient in a case where the first determination unit cannot determine that the calculation is unnecessary,
wherein the second selection unit further selects a different variable that results from storing in the table a correlation coefficient which is a correlation coefficient representing a relationship with the first variable and a correlation coefficient which is a correlation coefficient representing a relationship with the second variable, and
wherein, based on a correlation coefficient of the different variable, the first determination unit further determines whether or not the calculation of the first correlation coefficient is indispensable.
3. The data processing system according to claim 1 ,
wherein, in a case where multiple variables each of which satisfies a condition under which the variable is the third variable are present, the second selection unit selects the multiple variables sequentially, starting from the variable of which the second correlation coefficient is greater or of which the third correlation coefficient is greater.
4. The data processing system according to claim 1 ,
wherein each of the multiple variables is normalized.
5. The data processing system according to claim 1 , further comprising:
a second determination unit that causes the first selection unit to make a selection in a case where when the threshold is changed, the post-change threshold is smaller than a minimum value of the correlation coefficient that is stored in the table,
wherein the input unit is constituted in such a manner that the threshold is changeable.
6. The data processing system according to claim 1 , further comprising:
a display unit on which the multiple variables, the correlation coefficient that exceeds the threshold, among the multiple correlation coefficients, and a link that connects between the multiple variables each of which has the correlation coefficient that exceeds the threshold, are displayed.
7. A data processing method of calculating a correlation coefficient between multiple variables, the data processing method comprising:
determining whether or not calculation of a first correlation coefficient that is a correlation coefficient between a first variable and a second variable is indispensable;
selecting a third variable that is a variable which results from storing a second correlation coefficient that is a correlation coefficient representing a relationship with the first variable and a third correlation coefficient that is a correlation coefficient representing a relationship with the second variable, in a table in which multiple correlation coefficients are stored;
determining whether or not the calculation of the first correlation coefficient is indispensable, based on the second correlation coefficient and the third correlation coefficient; and
determining, based on the second correlation coefficient and the third correlation coefficient, whether or not there is a likelihood that the first correlation coefficient will exceed a threshold that is at a lower limit of the first correlation coefficient.