JP2000194710A - Device and method for synthesizing concept data having tree structure in multidimensional database - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device for synthesizing concept data with which categories can be easily synthesized without fail with respect to the tree structure of a multidimensional database. SOLUTION: A category synthesizing means 3 reads categories A and B out of a storage means 4, reads category synthesis rules out of a storage means 5 and automatically synthesizes a new category C while using these categories A and B and category synthesis rules. The category C synthesized by this category synthesizing means 3 is preserved in a storage means 6. It is described in the category synthesis rules to 'delete a node (b) of the second category and to make a child (c) of the node (b) into child of the first category (a) in the case of the same node name at the node (a) and the node (b)'. Besides, the algorithm of this category synthesizing means can be updated by a user input means 2 and can be confirmed by an information display means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an apparatus and method for synthesizing concept data having a tree structure in a multidimensional database having hierarchical structure data for analysis.

[0002]

2. Description of the Related Art Before describing the related art, definitions of terms used in this specification will be described. 1 and 2 are explanatory diagrams related to terms in a tree structure. FIG.
, Those indicated by a circle are referred to as nodes. A straight line between nodes is called a branch. As shown in FIG. 1, a node corresponding to the root of the tree is called a root, and a node located at the forefront in the direction opposite to the root with respect to the branch is called a leaf. FIG.
As shown in (2), of the two nodes sandwiching the branch, the one closer to the root is called a parent and the one farther away is called a child. The number of branches between the root and the leaf is called the length from the root to the leaf. Further, among the concept data, particularly, the concept data in which the length from the root to the leaf is constant in the form of a tree structure is called a category. In a tree structure having categories, each node between the root and the leaf (including the root and the leaf) is assigned a level, and the root level is set to the highest value (level 2 in the example of FIG. 1). , The leaf level is assigned to the lowest level (normally, the level value is “0” as shown in FIG. 1).

Next, a conventional technique will be described. In general, constructing a new multidimensional database from a multidimensional database having multiple hierarchically structured data is based on OLAP (O
n Line Analytical Processing) is a common task. For example, there is a case where data for May and June are added to data for April.

When constructing a new multidimensional database, it is often necessary to combine categories. Examples of the category include year and month (year-half-year-quarter-month) and region (Japan-prefecture-municipalities).

[0005] However, there is a restriction that categories must not have duplicate node names at the same level, must not form a loop, and must have a constant length from the root to the leaves. Can not be easily.

Heretofore, the category synthesizing operation has been performed manually while considering such category constraints.

[0007]

However, the task of manually synthesizing a category is not only troublesome, but also can result in the creation of a category that does not meet the category constraints.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problem, and has as its object to provide a conceptual data synthesizing apparatus and method for synthesizing categories easily and without errors.

[0009]

In order to achieve the above objects, a multi-dimensional database conceptual data synthesizing apparatus according to a first invention comprises a first category and a second category. Storage means for storing a first category, a second category, and a category combination rule, and a category combination means for automatically combining a third category using the first category, the second category, and the category combination rule. Storage means for storing the synthesized third category, wherein the category synthesis rules include “first
If the node name of the category a is the same as that of the node b of the second category, the node b is deleted and the child c of the node b is replaced with the node a.
To the child of ".

[0010] In the multi-dimensional database according to the second aspect of the present invention, the conceptual data synthesizing apparatus has a tree structure.
Storage means for storing the category, the second category, and the category synthesis rule, and category synthesis means for automatically newly synthesizing the third category using the first category, the second category, and the category synthesis rule. And storage means for storing the third category synthesized by the category synthesis means.
It can be described as "change the node name of the node b when the node name of the node a of the first category and the node b of the second category are the same."

The concept data synthesizing device having a tree structure in the multidimensional database according to the third aspect of the present invention comprises:
Storage means for storing the category, the second category, and the category synthesis rule, and category synthesis means for automatically newly synthesizing the third category using the first category, the second category, and the category synthesis rule. And storage means for storing the third category synthesized by the category synthesis means.
It can be described as "If the node name of the node a of the first category is the same as the node b of the second category, the node a is deleted and the child c of the node a is made a child of the node b."

[0012] Also, a concept data synthesizing apparatus having a tree structure in a multidimensional database according to a fourth aspect of the present invention comprises:
Storage means for storing the category, the second category, and the category synthesis rule, and category synthesis means for automatically newly synthesizing the third category using the first category, the second category, and the category synthesis rule. And storage means for storing the third category synthesized by the category synthesis means.
It can be described as "change the node name of the node a when the node name of the node a of the first category and the node b of the second category are the same."

A method for synthesizing conceptual data having a tree structure in a multidimensional database according to a fifth aspect of the present invention is described in the first aspect.
Storing the category, the second category, and the category combining rule in a file; and automatically combining a new third category using the first category, the second category, and the category combining rule. And storing the third category synthesized in the category synthesis step in a file. The category synthesis rules include “a node name of the first category and a node b of the second category. Are the same, node b is deleted and child c of node b is made child of node a. "

Further, a method for synthesizing conceptual data having a tree structure in a multidimensional database according to a sixth aspect of the present invention, comprises:
Storing the category, the second category, and the category combining rule in a file; and automatically combining a new third category using the first category, the second category, and the category combining rule. And storing the third category synthesized in the category synthesis step in a file. The category synthesis rules include “a node name of the first category and a node b of the second category. Are the same, the node name of the node b is changed. "

A method for synthesizing conceptual data having a tree structure in a multidimensional database according to a seventh aspect of the present invention is described in the first aspect.
Storing the category, the second category, and the category combining rule in a file; and automatically combining a new third category using the first category, the second category, and the category combining rule. And storing the third category synthesized in the category synthesis step in a file. The category synthesis rules include “a node name of the first category and a node b of the second category. Are the same, the node a is deleted and the child c of the node a is made a child of the node b. "

Further, a method for synthesizing conceptual data having a tree structure in a multidimensional database according to an eighth aspect of the present invention is described in the first aspect.
Storing the category, the second category, and the category combining rule in a file; and automatically combining a new third category using the first category, the second category, and the category combining rule. And storing the third category synthesized in the category synthesis step in a file. The category synthesis rules include “a node name of the first category and a node b of the second category. Are the same, the node name of the node a is changed. "

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 Before describing the present invention, definitions of terms used in this specification will be described. 1 and 2 are explanatory diagrams related to terms in a tree structure. In FIG. 2, the ones indicated by the circles are called nodes. A straight line between nodes is called a branch. Also, as shown in FIG. 1, a node corresponding to the root of the tree is called a root,
The node located at the forefront in the direction opposite to the root with respect to the branch is called a leaf. Further, as shown in FIG. 2, of the two nodes sandwiching the branch, the one closer to the root is called a parent, and the one farther away is called a child. The number of branches between the root and the leaf is called the length from the root to the leaf. Further, among the concept data, particularly, the concept data in which the length from the root to the leaf is constant in the form of a tree structure is called a category.

The term "level" is defined as the length from the leaf to the node (the number of levels) (FIG. 2). That is, for each node between the leaf and the root (including the root and the leaf), assignment is performed so that the level value increases in the direction from the leaf to the root, and the leaf level is set to the lowest (as shown in FIG. 1). And the root level is assigned to the highest value (level 2 in the example of FIG. 1).

FIG. 3 is a schematic block diagram of a conceptual data synthesizing apparatus according to the present invention. The present apparatus comprises: an information display unit 1 for displaying information to a user; a user input unit 2 for receiving an input from the user; a storage unit 5 for storing a category synthesis rule; Storage means 4 for storing category B as a second category, category combining means 3 for realizing a category combining method for combining category C as a third category from category A and category B, and category combining means And storage means 6 for storing the category C synthesized by step 3. The category combining method of the category combining means 3 can be realized by a program.

The information display means 1 is generally a display. The user input means 2 is generally a keyboard and a mouse, a trackball, and a pen. In addition, storage means 4 for category A and category B, storage means 6 for category C,
The storage means 5 for the category combination rules is, for example, a disk. In FIG. 3, these are described as being stored on separate disks, but all of them may be stored on one disk. Also, it may be stored on a disk connected to another computer on the network.

A category is represented by a set of branches constituting the category. Branches are represented by (start point, end point). According to this method, the categories in FIG. 4 are ｛(a, b), (a,
c), (b, d), (b, e), (c, f), (c,
g) It can be represented by｝. Since the category has a constant distance from the root to the leaf, once the node is determined, the length (level) to the leaf is uniquely determined.

The category synthesizing means 3 stores the categories A and B from the storing means 4 for the categories A and B,
The category combining rule is read from the category combining rule storage means 5. Then, category C is generated based on the category synthesis algorithm. This category C is stored in the storage means 6.

FIG. 5 is a flowchart showing the operation of the category synthesizing means 3. 9 to 17 show that the level is 2
FIG. 8 is an explanatory diagram showing a category synthesis process in the tree structure of FIG. Next, the operation of the category synthesizing means 3 will be described with reference to FIGS.

First, the category synthesizing means 3 reads the category A, the category B, and the category synthesizing rule (step S101).

Next, the union of category A and category B is taken as category C (step S102). At this time, all the category A node names are added to the original data as “: A”.
Change to the one with. Similarly, all categories B
Is changed to the original data with “: B” added to the original data. At this time, a new branch is added so that level 0 data of category A becomes level 0 data of category C, and level 0 data of category B becomes level 0 data of category C. So that the category C has a tree shape. That is, when the maximum level of category A is LA, the root of category A is ra, the maximum level of category B is LB, and the root of category B is rb, (r, ra_1), (ra
_1, ra_2), ..., (ra_m-1, ra_m), (r, rb_1), (rb_1, rb_
2), ..., (rb_n-1, rb_n) are added to category C. Where ra_m = ra, rb_n = rb, m = LB-min (LA, LB) + 1, n = LA
-min (LA, LB) +1. Category A and Category B in FIG.
When combining, LA = 2, LB = 2, (r, a1: A), (r, b
1: B) is added.

Since the category C created at this time retains {0: A} and {: B} at the node name of level 0, it cannot be used as it is as conceptual data of the multidimensional database. Then, the category C is converted into a form that can be used as conceptual data of the multidimensional database.

Steps S104 to S110 are repeated in the range of L from the maximum level of category B to 0 (step S103). Let b be a node at level L of category B (step S104). Let a be a node of category A (step S105). a and b are compared (step S106), and if a and b are different names, step S109
Go to If the names are the same, the procedure goes to step S107, where the level in category A and the level in category B are compared. If the level in the category A is the same as the level in the category B, the process goes to S109 and performs processing according to the category combination rule (S108).

If the level in category A and the level in category B are not the same, the name of b in category C is changed.

The processing from step S105 to step S109 is performed for all elements a of category A.
The processing from step S104 to step S110 is performed on all the nodes at level L of category B. Further, the processing from step S103 to step S111 is performed for all levels.

Next, unnecessary branches of category C will be removed.

From level 1 to the highest level of C: For L, "If the child of e2 is a leaf at the branch (e1, e2) of level L of C, the branch (e1, e2) is removed from C". The operation is repeated for all the branches (steps S112 to S11)
4).

Finally, {: A} and {: B} are removed from all the node names of category C (step S115). This completes the synthesis of category C.

A specific example will be described. Category A in FIG.
= ((A1, a2), (a1, a3), (a2, a4), (a2, a5), (a3, a6), (a3, a
7)} and category B = {(b1, b2), (b1, a3), (b2, a4), (b2, a5),
Consider the synthesis of (a3, b6), (a3, b7)}. First, a category C is created as a union of categories A and B (step S101). At this time,: A is added after the category A element, and: B is added after the category B element. Further, necessary branches are added so that the category C has a tree shape. The category C thus created is {(r, a1: A), (r, b1: B),
(a1: A, a2: A), (a1: A, a3: A), (b1: B.b2: B), (b1: B, a3: B), (a
2: a, a4: A), (a2: A, a5: A), (a3: A, a6: A), (a3: A, a7: A), (b2:
B, a4: B), (b2: B, a5: B), (a3: B, b6: B), (a3: B, b7: B)}. (FIG. 10)

Next, all categories A and B
Check name and level for elements of. a3: A and a
3: B is the same name and at the same level, so category B
Make the child of a3: B on the side a child of a3: A. And a3: B
Cut off a branch with a parent or child. Then category C is
((r, a1: A), (r, b1: B), (a1: A, a2: A), (a1: A, a3: A), (b1: B, b
2: B), (a2: A, a4: A), (a2: A, a5: A), (a3: A, a6: A), (a3: A, a7:
A), (a3: A, b6: B), (a3: A, b7: B), (b2: B, a4: B), (b2: B, a5:
B)} (FIG. 11).

Similarly, since a4: A and a4: B are at the same level with the same name, a process of moving the children of a4: B to a4: A is performed. However, since there is no child in a4: B, it is only necessary to remove the branch having a4: B as a child. As a result, the category C becomes {(r, a1: A), (r, b1:
B), (a1: A, a2: A), (a1: A, a3: A), (b1: B, b2: B), (a2: A, a4:
A), (a2: A, a5: A), (a3: A, a6: A), (a3: A, a7: A), (a3: A, b6:
B), (a3: A, b7: B), (b2: B, a5: B)} (FIG. 12).

The same applies to a5: A and a5: B, and the branches having a5: B as a child are removed. The category C thus created is as follows. C = {(r, a1: A), (r, b1: B), (a1: A, a2:
A), (a1: A, a3: A), (b1: B, b2: B), (a2: A, a4: A), (a2: A, a5:
A), (a3: A, a6: A), (a3: A, a7: A), (a3: A, b6: B), (a3: A, b7:
B)} (FIG. 13).

Next, unnecessary branches are removed. The category C thus created is as follows. C = {(a1: A, a2:
A), (a1: A, a3: A), (a2: A, a4: A), (a2: A, a5: A), (a3: A, a6:
A), (a3: A, a7: A), (a3: A, b6: A), (a3: A, b7: A)} (FIG. 14).

Finally,: A,: B is removed from the node name, and the process ends. The category C thus created is as follows. C = {(a1, a2), (a1, a3), (a2, a4), (a2, a5), (a3, a6),
(a3, a7), (a3, b6), (a3, b7)} (FIG. 15). This is the combined category.

Conventionally, synthesis of conceptual data having a tree structure in a multidimensional database has been performed manually, which may be troublesome and may cause an error. It can be done easily and without errors.

Embodiment 2 FIG. 3 is a schematic configuration diagram of the present embodiment. The description of FIG. 3 has been made in the first embodiment, and will not be repeated. FIG. 6 is a flowchart of the category combining program according to the second embodiment. The only difference between FIG. 5 and FIG. 6 is the difference between step S108 and step S208. In the second embodiment, if the level in category A and the level in category B are the same, the node name of b is changed. The user inputs a new node name (step S208).

Conventionally, synthesis of conceptual data having a tree structure in a multidimensional database has been performed manually, which may be cumbersome and may cause an error. It can be done easily and without errors.

Embodiment 3 FIG. 3 is a schematic configuration diagram of the present embodiment. The description of FIG. 3 has been made in the first embodiment, and will not be repeated. FIG. 7 is a flowchart of the category combining program according to the third embodiment. The difference between FIG. 5 and FIG. 7 is only the difference between S108 and S308. In the third embodiment, if the level in category A is the same as the level in category B, a is removed and the child of a becomes the child of b (step S308).

Conventionally, synthesis of conceptual data having a tree structure in a multidimensional database has been manually performed, which may be troublesome and may cause an error. It can be done easily and without errors.

Embodiment 4 FIG. FIG. 3 is a schematic configuration diagram of the present embodiment. The description of FIG. 3 has been made in the first embodiment, and will not be repeated. FIG. 8 is a flowchart of a category combining program according to the fourth embodiment. The only difference between FIG. 5 and FIG. 8 is the difference between S108 and S408. In the fourth embodiment, if the level in category A and the level in category B are the same, the node name of a is changed. The user inputs the new node name from the user input unit (step S408).

Conventionally, synthesis of conceptual data having a tree structure in a multidimensional database has been performed manually, which may be cumbersome and may cause an error. It can be done easily and without errors.

[0046]

According to the first and fifth aspects of the present invention, the first category, the second category, and the category combining rule are stored in a file, and the first category, the second category, and the category combining rule are used. Automatically combine the third category and save it to a file.
If the node name of the category a is the same as that of the node b of the second category, the node b is deleted and the child c of the node b is replaced with the node a.
To be a child of "
Since synthesis of concept data having a tree structure in a multidimensional database is automatically performed, it can be performed easily and without errors.

According to the second and sixth aspects of the present invention, the category combination rule includes “the node a of the first category and the second
When the node name is the same in the node b of the category, the node name of the node b is changed. ", And automatically synthesizes conceptual data having a tree structure in a multidimensional database. Therefore, it can be performed easily and without error.

According to the third and seventh aspects of the invention, the category combination rule includes “the node a of the first category and the second
When the node name is the same in the node b of the category, the node a is deleted and the child c of the node a is made a child of the node b. " Is automatically performed, so that it can be performed easily and without error.

According to the fourth and eighth aspects of the present invention, the category combination rule includes “a node a of the first category and a second
If the node name is the same at the node b of the category, the node name of the node a is changed. "
Since synthesis of concept data having a tree structure in a multidimensional database is automatically performed, it can be performed easily and without errors.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating terms of a tree structure.

FIG. 2 is a diagram illustrating terms of a tree structure.

FIG. 3 is a schematic configuration diagram of an embodiment of an information processing system having a category combining device according to the present invention.

FIG. 4 is an example of a category.

FIG. 5 is a flowchart of a process according to the first embodiment of the present invention.

FIG. 6 is a flowchart of a process according to the second embodiment of the present invention.

FIG. 7 is a flowchart of a process according to a third embodiment of the present invention.

FIG. 8 is a flowchart of a process according to a fourth embodiment of the present invention.

FIG. 9 shows a category synthesis process (1).

FIG. 10 is a category synthesis process (2).

FIG. 11 shows a category synthesis process (3).

FIG. 12 shows a category synthesis process (4).

FIG. 13 shows a category synthesis process (5).

FIG. 14 shows a category synthesis process (6).

FIG. 15 shows a category synthesis process (7).

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 information display means 2 user input means 3 category synthesis means 4 storage means (category A, category B) 5 storage means (category synthesis rule) 6 storage means (category C)

Claims (8)

[Claims] 1. Hierarchical data used for analyzing data in a multidimensional database is called conceptual data, and conceptual data in which the length of a leaf from a root is constant in a tree structure is called a category. A synthesizing device for conceptual data having a tree structure in a multidimensional database in which a part to be called is referred to as a node; storage means for storing a first category, a second category, and a category synthesis rule; A category synthesizing unit for automatically synthesizing a third category automatically using a category and a category synthesizing rule; and a storage unit for storing the third category synthesized by the category synthesizing unit, wherein the category synthesizing rule is In the case where the node name is the same between the node a of the first category and the node b of the second category, the node b is deleted and the node b is deleted. Synthesizer concept data having a tree structure in a multidimensional database, characterized in that the child c can be described to be a child of the node a "on. 2. Hierarchical structure data used for analyzing data in a multidimensional database is referred to as concept data. In particular, concept data having a constant length from a root to a leaf in a tree structure is referred to as a category. A synthesizing device for conceptual data having a tree structure in a multidimensional database in which a part to be called is referred to as a node; storage means for storing a first category, a second category, and a category synthesis rule; A category synthesizing unit for automatically synthesizing a third category automatically using a category and a category synthesizing rule; and a storage unit for storing the third category synthesized by the category synthesizing unit, wherein the category synthesizing rule is "Change the node name of node b when the node name of node a of the first category and the node b of the second category are the same. That "
An apparatus for synthesizing conceptual data having a tree structure in a multidimensional database, characterized in that it can be described as: 3. Hierarchical structure data used for analyzing data in a multidimensional database is referred to as conceptual data. In particular, conceptual data having a constant length from the root to the leaf in a tree structure is referred to as a category. A synthesizing device for conceptual data having a tree structure in a multidimensional database in which a part to be called is referred to as a node; storage means for storing a first category, a second category, and a category synthesis rule; A category synthesizing unit for automatically synthesizing a third category automatically using a category and a category synthesizing rule; and a storage unit for storing the third category synthesized by the category synthesizing unit, wherein the category synthesizing rule is "If the node a of the first category and the node b of the second category have the same node name, the node a is deleted and the node a is deleted. Synthesizer concept data having a tree structure in a multidimensional database, characterized in that the child c can be described to be a child of the node b "on. 4. Hierarchical structure data used for analyzing data in a multidimensional database is referred to as conceptual data. In particular, conceptual data having a constant length from a root to a leaf in a tree structure is referred to as a category. A synthesizing device for conceptual data having a tree structure in a multidimensional database in which a part to be called is referred to as a node; storage means for storing a first category, a second category, and a category synthesis rule; A category synthesizing unit for automatically synthesizing a third category automatically using a category and a category synthesizing rule; and a storage unit for storing the third category synthesized by the category synthesizing unit, wherein the category synthesizing rule is "Change the node name of node a if the node name is the same for node a of the first category and node b of the second category. That "
An apparatus for synthesizing conceptual data having a tree structure in a multidimensional database, characterized in that it can be described as: 5. Hierarchical data used for analyzing data in a multidimensional database is referred to as conceptual data. In particular, conceptual data in which the length from the root to the leaf is constant in a tree structure is referred to as a category. Storing a first category, a second category, and a category synthesis rule in a file, wherein the first category, the second category, and the category synthesis rule are stored in a file. Automatically synthesizing a new third category by using the category and the category synthesizing rule, and storing the third category synthesized by the category synthesizing step in a file. The rule states that “when the node names of the first category node a and the second category node b are the same, Remove the b, a method of synthesizing concept data having a tree structure in a multidimensional database, characterized in that the child c of the node b can be described to be a child of the node a ". 6. Hierarchical data used for analyzing data in a multidimensional database is referred to as conceptual data, and in particular, conceptual data having a fixed length from the root to the leaf in a tree structure is referred to as a category. Storing a first category, a second category, and a category synthesis rule in a file, wherein the first category, the second category, and the category synthesis rule are stored in a file. Automatically synthesizing a new third category by using the category and the category synthesizing rule, and storing the third category synthesized by the category synthesizing step in a file. The rule states that “when the node names of the first category node a and the second category node b are the same, The method of synthesis concept data having a tree structure in a multidimensional database, characterized in that can be described as changing the node name of the b ". 7. Hierarchical data used for analyzing data in a multidimensional database is referred to as conceptual data, and in particular, conceptual data in which the length of a leaf from a root is constant in a tree structure is referred to as a category. Storing a first category, a second category, and a category synthesis rule in a file, wherein the first category, the second category, and the category synthesis rule are stored in a file. Automatically synthesizing a new third category by using the category and the category synthesizing rule, and storing the third category synthesized by the category synthesizing step in a file. The rule states that “when the node names of the first category node a and the second category node b are the same, Remove the a, method of synthesis concept data having a tree structure in a multidimensional database, characterized in that the child c of the node a can be described to be a child of the node b. " 8. Hierarchical structure data used for analyzing data in a multidimensional database is referred to as concept data. In particular, concept data having a constant length from a root to a leaf in a tree structure is referred to as a category. Storing a first category, a second category, and a category synthesis rule in a file, wherein the first category, the second category, and the category synthesis rule are stored in a file. Automatically synthesizing a new third category by using the category and the category synthesizing rule, and storing the third category synthesized by the category synthesizing step in a file. The rule states that “when the node names of the first category node a and the second category node b are the same, The method of synthesis concept data having a tree structure in a multidimensional database, characterized in that can be described as changing the node name of a ".