What is claimed is:
1. A radiographic image conversion panel comprising:
a support including a heat resistant resin; and
a photostimulable phosphor layer provided on a face of the support by a vapor phase deposition method,
wherein an average surface roughness in the face of the support to be provided with the phosphor layer is above 0.05 m and not more than 0.5 m.
2. The panel of claim 1, wherein a maximum surface roughness in the face of the support to be provided with the phosphor layer is above 0.1 m and not more than 1.0 m.
3. The panel of claim 1, wherein the heat resistant resin is polyimide.
4. The panel of claim 1, wherein the phosphor layer comprises a photostimulable phosphor represented by the following Formula (1),
CsX:yA(1)
wherein the X represents Cl, Br or I, the A represents Eu, Sm, In, Tl, Ga or Ce, and a range of the y is not less than 1107 and not more than 1102.
5. A radiographic image conversion panel comprising:
a support having a substrate including a carbon fiber reinforced plastic and a heat resistant resin film including a heat resistant resin, the film being provided on a face of the substrate; and
a photostimulable phosphor layer provided by a vapor phase deposition method on a face of the film opposite side from the substrate,
wherein an average surface roughness in the face of the film to be provided with the phosphor layer is above 0.05 m and not more than 0.5 m.
6. The panel of claim 5, wherein a maximum surface roughness in a face of the support to be provided with the phosphor layer is above 0.1 m and not more than 1.0 m.
7. The panel of claim 5, wherein the heat resistant resin is polyimide.
8. The panel of claim 5, wherein in the substrate, two or more layers of unidirectional prepregs are laminated.
9. The panel of claim 5, wherein the phosphor layer comprises a photostimulable phosphor represented by the following Formula (1),
CsX:yA(1)
wherein the X represents Cl, Br or I, the A represents Eu, Sm, In, Tl, Ga or Ce, and a range of the y is not less than 1107 and not more than 1102.
10. A radiographic image conversion panel comprising:
a support having a substrate including a composite material of a matrix resin and a carbon fiber; and
a photostimulable phosphor layer provided on one face of the substrate by a vapor phase deposition method,
wherein a glass-transition temperature of the matrix resin is not less than 100 C. and not more than 300 C.
11. The panel of claim 10, wherein the matrix resin is one of an epoxy resin and a polyimide resin.
12. The panel of claim 11, wherein a weight ratio of the matrix resin against the substrate is not less than 15% and not more than 60%.
13. The panel of claim 10, wherein the phosphor layer comprises a photostimulable phosphor represented by the following Formula (1),
CsX:yA(1)
wherein the X represents Cl, Br or I, the A represents Eu, Sm, In, Tl, Ga or Ce, and a range of the y is not less than 1107 and not more than 1102.
14. A radiographic image conversion panel comprising:
a support having a substrate including a composite material of a matrix resin and a carbon fiber, and a heat resistant resin film provided on a face of the substrate; and
a photostimulable phosphor layer provided by a vapor phase deposition method on a face of the film opposite side from the substrate,
wherein a glass-transition temperature of the matrix resin is not less than 100 C. and not more than 300 C.
15. The panel of claim 14, wherein the matrix resin is one of an epoxy resin and a polyimide resin.
16. The panel of claim 15, wherein a weight ratio of the matrix resin against the substrate is not less than 15% and not more than 60%.
17. The panel of claim 14, wherein the heat resistant resin film comprises a polyimide resin.
18. The panel of claim 14, wherein the phosphor layer comprises a photostimulable phosphor represented by the following Formula (1),
CsX:yA(1)
wherein the X represents Cl, Br or I, the A represents Eu, Sm, In, Tl, Ga or Ce, and a range of the y is not less than 1107 and not more than 1102.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A method of generating at least a portion of a decision tree structure on a computer, the decision tree structure generation being performed by the computer to perform customer segmentation on a set of customers represented as one or more multidimensional data objects comprising data records of personal attributes stored on data storage device coupled to the computer, the method comprising the steps of:
representing data associated with the one or more multidimensional data objects as a node of the decision tree structure;
determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data; and
dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure,
wherein the discriminant measure is determined in accordance with a Fisher’s discriminant technique.
2. The method of claim 1, wherein the condition determining and data dividing steps are repeated at each subsequent node.
3. The method of claim 1, wherein the data is divided at a split plane determined to be perpendicular to a direction determined according to the Fisher’s discriminant technique.
4. The method of claim 3, wherein the split plane is determined to be a plane where an entropy measure is substantially optimized.
5. The method of claim 4, wherein the entropy measure is determined in accordance with a gini index.
6. A method of performing data classification on a computer, the data classification being performed by the computer to perform customer segmentation on a set of customers represented as multidimensional objects comprising data records of personal attributes stored on a data storage device coupled to the computer, the method comprising the steps of:
generating at least a portion of a decision tree structure from one or more multidimensional training data objects by: (i) representing data associated with the one or more multidimensional data objects as a node of the decision tree structure; (ii) determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data; and (iii) dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure; and
using the decision tree structure to predict one or more class variables associated with one or more multidimensional testing data objects in which only one or more feature variables are known,
wherein the class variable is found by using tree traversal on the decision tree structure, and
wherein a branch is selected to traverse based on an entropy measure.
7. The method of claim 6, wherein the branch yielding the lowest entropy measure is selected as the branch to traverse.
8. Apparatus for generating at least a portion of a decision tree structure perform customer segmentation on a set of customers represented as multidimensional data objects, the apparatus comprising:
a computer having a memory and a data storage device coupled thereto, wherein the data storage device stores a data store, the data store having one or more multidimensional data objects comprising data records of personal attributes; and
one or more computer programs, performed by the computer, for: (i) representing data associated with the one or more multidimensional data objects as a node of the decision tree structure; (ii) determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data; and (iii) dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure;
wherein the memory stores at least a portion of the decision tree structure, and
wherein the discriminant measure is determined in accordance with a Fisher’s discriminant technique.
9. The apparatus of claim 8, wherein the condition determining and data dividing operations are repeated at each subsequent node.
10. The apparatus of claim 8, wherein the data is divided at a split plane determined to be perpendicular to a direction determined according to the Fisher’s discriminant technique.
11. The apparatus of claim 10, wherein the split plane is determined to be a plane where an entropy measure is substantially optimized.
12. The apparatus of claim 11, wherein the entropy measure is determined in accordance with a gini index.
13. Apparatus for performing customer segmentation on a set of customers represented as multidimensional objects, the apparatus comprising:
a computer having a memory and a data storage device coupled thereto, wherein the data storage device stores a data store, the data store having the multidimensional objects comprising data records of personal attributes; and
one or more computer programs, performed by the computer for: (I) generating at least a portion of a decision tree structure from one or more multidimensional training data objects by representing data associated with the one or more multidimensional data objects as a node of the decision tree structure, determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data, and dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure; and (ii) using the decision tree structure to predict one or more class variables associated with one or more multidimensional testing data objects in which only one or more feature variables are known;
wherein the memory stores at least a portion of results associated with one or more of the data classification operations,
wherein the class variable is found by using tree traversal on the decision tree structure, and
wherein a branch is selected to traverse based on an entropy measure.
14. The apparatus of claim 13, wherein the branch yielding the lowest entropy measure is selected as the branch to traverse.
15. An article of manufacture comprising a program storage medium readable by a computer and embodying one or more instructions executable by the computer to perform method steps for generating at least a portion of a decision tree structure on the computer, the decision tree structure generation being performed by the computer to perform customer segmentation on a set of customers represented as one or more multidimensional data objects comprising data records of personal attributes stored on a data storage device coupled to the computer, the method comprising the steps of:
representing data associated with the one or more multidimensional data objects as a node of the decision tree structure;
determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data; and
dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure,
wherein the discriminant measure is determined in accordance with a Fisher’s discriminant technique.
16. The article of claim 15, wherein the condition determining and data dividing steps are repeated at each subsequent node.
17. The article of claim 15, wherein the data is divided at a split plane determined to be perpendicular to a direction determined according to the Fisher’s discriminant technique.
18. The article of claim 17, wherein the split plane is determined to be a plane where an entropy measure is substantially optimized.
19. The article of claim 18, wherein the entropy measure is determined in accordance with a gini index.
20. An article of manufacture comprising a program storage medium readable by a computer and embodying one or more instructions executable by the computer to perform methods steps for performing data classification on a computer, the data classification being performed by the computer to perform customer segmentation on a set of customers represented as multidimensional objects comprising data records of personal attributes stored on a data storage device coupled to the computer, the method comprising the steps of:
generating at least a portion of a decision tree structure from one or more multidimensional training data objects by: (I) representing data associated with the one or more multidimensional data objects as a node of the decision tree structure; (ii) determining a condition for dividing the data at the node into at least two subsequent nodes based on a discriminant measure which substantially maximizes separation between classes associated with the data; and (iii) dividing the data based on the condition, wherein the resulting nodes form at least a portion of the decision tree structure; and
using the decision tree structure to predict one or more class variables associated with one or more multidimensional testing data objects in which only one or more feature variables are known,
wherein the class variable is found by using tree traversal on the decision tree structure, and
wherein a branch is selected to traverse based on an entropy measure.
21. The article of claim 20, wherein the branch yielding the lowest entropy measure is selected as the branch to traverse.