JP2003216619A - Computer processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a very convenient information control technique. <P>SOLUTION: A computer processing apparatus includes a decision part for deciding situations prior to retrieval; a specifying part for specifying according to the situations the attributes of a cell to be retrieved; a retrieval part for retrieving cells with the attributes from within data space; and a processing part for constructing, based on the retrieval result, the cell space involving the desired attributes. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention [0001] TECHNICAL FIELD The present invention relates to information using cell information theory.
Technology for managing. [0002] 2. Description of the Related Art Conventionally, database models have been
Rational databases are common. Relay show
Null database recognizes data dependencies
Relevance of data is premised on the existence of an information manager
Is defined and fixed in advance by the information manager. Individual
When managing data within a closed group such as people or companies
In that case, even if the data dependency is statically fixed,
No problems, but on the web, which has grown rapidly in recent years
To manage the information of the built cyber world, relay
The null database is no longer available. Cyber world
Now, on the web that continues to work independently and at a dizzying pace
The site is organically linked through the WWW,
These rapid and complex changes in information are related to relational data.
Database management limits are exceeded. [0003] DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
In light of the current situation,
We have developed a new model that overturns the book. Its purpose is more
It is to provide highly convenient information management technology. [0004] Certain aspects of the invention include:
The present invention relates to a computer processing device. This device is prior to search
Therefore, the judgment unit for judging the situation and the inspection according to the situation.
A specification section that specifies the attribute of the cell to be searched, and includes the above attribute.
A search unit that searches for cells in the data space, and
Build a cell space containing desired attributes based on the search results
And a processing unit. The search unit determines whether the data space has the attribute.
The cells may be disassembled from the viewpoint of nothing. This device is
Holds the cell operation performed in the search section or the processing section
It may further include a holding unit. Another aspect of the invention is also a computer processor.
Regarding the table This device is a
Database for storing information in a form, and the database
And a management unit that manages the zero cell
The cell ID is stored in the
In order to specify, the effective range is added to the cell and managed.
It Yet another aspect of the present invention is a computer.
Regarding a processing device. This device is a device that stores educational elements.
The database and the educational elements are
Management of the curriculum through cell calculation
And a science department. Any combination of the above components
The method of expressing the present invention, an apparatus, a system, a recording medium,
Something converted between computer programs, etc.
It is also effective as an aspect of the present invention. [0009] BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of the present invention will be described below.
I will explain based on. First of all, as the basis of the embodiment
Cell Information Model (CIM: Cellu)
The basics of lar Information Model)
State the theory as a prerequisite technology, and then implement concrete implementation.
The form will be described. [First Prerequisite Technology] [1-0] Summary Cyber world is an information world formed on the Web
And it is constantly changing over time. Wide range of low
Kal's activity is to create a cyber world.
Fuse each other on the global Web. Cell in this paper
Present a theoretical situation model. "cell
The information model named “Ra model” is a local model
Integrate globally. Cellular model in cell space
Applied to database design
Various electronic commerce in space to glue cell space
Can be realized as a situation model according to
It [1-1] Foreword Cyberworld is a website for various local activities.
The result is composed of global fusion. locker
Cyber site by changing daily site
Do is a vast world where information constantly changes over time.
It And in a new world like this,
A rationale for delegation is needed. Currently used as a general-purpose database
It is a relational database. Only
The relational data of a relational database
Data dependency in situations that need to change without changing
Situation where an information manager who is aware of
Is assumed. This data model is for example
In a closed space, grasp the data dependency,
Can be modeled, but like the cyber world
A vast world in which information changes over time
Seems not very suitable for. In the database
Use the huge and irregular data on the Web to change
Correspondence is required. There is a huge and irregular data
Available to respond to complex changes in the cyber world
Enable modeling that accurately characterizes common parts
Apply the cellular model. In this paper, cell theory is applied.
The situation model of electronic commerce using
It [1-2] Cellular model Cellular model is for modeling information on the Web
To minimize the modeling size
It uses the concept. Define invariants in a hierarchy
Can model various situations. one
On the other hand, the relational data model is based on set theory.
However, because it is abstract, it cannot project reality into personality. When a database of information on the Web is created
In this case, a huge amount of information is constantly changing over time, so it is essential.
If you can't define all the attributes in advance,
At the same time, it is not possible to determine the dependency relationship between tables
Yes. Therefore, an outline called equivalence of cell theory
Applying a reminder, the equivalency data
It becomes possible to determine the dependency relationship of data. The structure of this design
The cell spatial structure, work the cell joining (cell comp
position), cell decomposition (celldecompo)
The position). Keep homotopy again
By doing so, you can restore the deformed cell to its original state.
It There are boundaries in the cellular data model.
Closed cells, open cells without borders
Closed in each n dimension Is BⁿIs the boundary of. Open cells are closed cells
One attribute included, as well as the open cell itself
Is treated as a cell having one attribute. Cell model of the cell
Cellde, if bounds and join invariants are preserved
composition is possible. Object perception
Another is identification mapping
(Quantity mapping)
Be done. The binding is called an attaching map c
ongoing mapping, ject
It is defined by ive mapping. [1-3] Web Information by Cellular Model
Modeling When modeling, first the nature of the composition of the cyber world
It is necessary to characterize Mostly many W
Cyber as a result of local activity on the eb site
-World X is created. Information, unlike information in a company
It is unlikely that an administrator will give you an initial schema. Cyber
Local Web site to clarify World X
To search for specific information (information
on mining). Of course information
There is no management of innings. And on the web
Integrate what you find from information on distributed sites
To do. This is generally "information by design.
It is called "mining". Information mining on the Web
Is Whitehead inductive sche
Based on me, the local Web world is
Work to be fully integrated into Iverworld. n-dimensional service
Everworld XⁿIs And can be configured. Whitehead inductive
  Cyber world built on the scheme
Guarantees the following relationships. X⁰⊆ X¹⊆ X^Two⊆. ． ． ⊆ X^n-1⊆ Xⁿ⊆. ． ． ⊆
X How is it viewed from the perspective of cyberworld verification?
Cyberworlds of even dimensions are cybers of lower dimensions
This is valid because it contains the world. In the above, the identification is based on the equivalence relation.
Identified by equivalence classes
To be done. In other words, "equalence class
Generally speaking, "identification by es" is a relational model.
Corresponds to Dell's JOIN operation. Hence the cell
One of the usefulness of the Lah model is seen in this aspect. We
Cyber world on b is very complex and changes significantly
However, the integration power of the cellular model is true for the Web information model.
Supply the theoretical basis. [1-4] Situation model Is it a given cyber world on the Web?
Encounter a situation where it is necessary to create a new cyber world from
I have something to do. To model various Web situations
See the status of e-commerce. Online shop here
A typical situation will be described with reference to the example below. Online here
Take 3 possible situations in Yop
increase. ・ Consumers look for products. ・ Stores will acquire the needs of consumers as marketing
It ・ The store aims to expand sales and uses the sales information to identify the customer's purchasing tendency.
Find out. These situations are set and shown below. (Situation 1) I bought a product
Online products where customers who want
Browse the internet to find a shop. (Situation 2) Online dealing with products
Shop browses customer list for marketing
To do. (Situation 3) Online dealing with products
Shops perform data mining to expand sales. Here, p, q, and
It is defined as an r-dimensional cyber world as follows.
It B_s ^p  : shop B_c ^q  : client B_m ^r  : Product (Situation 1): Hope the store
When there is a product that meets the conditions (usually the lowest price)
Identifies the name of the product you are interested in with the product in the store. this
In the situation the information schema is a cell due to identification work
by decomposition operation
Is formed. cell decompositio
n operation is a mapping f. Given n-dimensional cell BⁿAttaching ma
pg_iTo hold Repeated times. Attaching in each cell disassembly
Keeping the mapping makes cell decomposition homotopic
can do. 1. cell decompositio
n Store B_s ^p, Customer B_c ^q, Product B_m ^rMinutes each
To identify the e-commerce associated with the attribute by
Equivalent cells B
_a ^t(A = s, c, m) and the rest. cel
l decomposition operation
Is And g, h, and k are the following
ation function. The homotopy is retained by these. For example, the attribute is quotient
B if it is the product name, product price, or product ID_a ^t= B_a ^Threeso
is there. 2. Cell by cell attachment
composition B of the product by the attaching map_m ^rAnd store
B_s ^pThe customer's B_c ^qAttach to. At here
Each taching map is Equivalent cells B_a ^tTo
Identify. Here, a = s, c, m. Now information
Keema is configured as a structure of electronic commerce on the Web
Since it has been done, some e-shop B_s ^pAttribute B_s
^tThe lowest value (price) in electronic commerce related to
Search for the attribute "commodity price". Site 2: First, We
Purchase candidate list B on b_mi ^riSent e-custom
omer B_c ^qSearch the web. This purchase candidate
Strikes usually have only partial information about the product. 1. cell decompositio
n E-customers B searched_ci ^qi(I =
1. ． ． n) Purchase candidate list B_mi ^riAnd the rest
It is divided by the cell disassembly operation. Cell decomposition
Peration is And here Is. Also, g is And this is cell decomposition
  homotopy against cell deformation due to operation
Identification f for holding
It is an unction. Apparently customer purchase candidate list B
_mi ^riProducts on the web are products B_m ^rThe service
Busset B_mi ^ri⊆ B_m ^rAnd r_iConsidered ≦ r
Be done. 2. Cell by cell attachment
composition identification function f
_mPurchase candidate product B guided by_mi ^riIs equivalent to
E-merchandisc B_m ^rIdentify attributes of
And found customer B_ci ^qiThe product B_m ^rAttack on
I will. At that time, identification ma
p is (∂ ... ∂B_c ^qWhere ∂ is q_i-R_iRepeated times
And ∂. ． ． ∂B_m ^r∂ is r-r_iRepeated times
You Also ∂. ． ． ∂B_m ^rIs B_mi ^riIs equivalent to
It ) With this, in order to expand sales, store B_s ^pPurchase candidate product information for
Can get all the information of the customer including
It Position 3: e-shop
A condition to extract only the desired information from a huge amount of data
Encounter a situation. For example, the attributes A, B, C, D, E
The situation is that they want the information they have. First, the table
The cell with the attribute number 0 is searched for in the data space.
A zero-dimensional cyber world X made up of zero-dimensional cells
⁰Is Becomes [0030] Find from between. [0031] Search from space. By repeating such work,
N-dimensional cyber world XⁿCan be configured
It I want to add more to the information I have added so far By these work, XⁿIs open cell e^j
_i(J = 1 to n, element of entity),
It is possible to grasp the whole picture of data related to desired information.
Wear. [1-5] Concrete example So far, we have three
I have presented these as examples of online bookstores below
Show. In addition, B in each case_s ^p, B_c ^q, B
_m ^rHave different meanings. Position 1: "Customer is on the Web
Search the selling sites to buy books on. Customers sell
Find the site you are doing and get more information about the book
I want to put it in. " At this time, the sales site cell is an individual online bookstore.
A cell that collects information about Customer cell: B_c ⁷ [Name] [Age] [Gender] [Address] [Request title]
[Author] [Publisher] Selling site cell: B_s ⁷ [Handling site] [Title] [Author] [Publisher] [Sales
Selling price] [Number of sales] [Stock] Product information cell: B_m ⁸ [ISBN Code] [Title] [Author] [Publisher]
[Genre] [List price] [Issue date] [Pre-published] Customer cell, sales site cell, product information cell
Are disassembled and joined by the equivalence relation. First,
Equivalent cells and other parts by
Divide into cell decomposition m
ap f_c: identification function
g: cell decomposition map
f_s: identification function
h: cell decomposition map
f_m: identification function
k: Then, according to the attaching map
Join sales site cell and product information cell to customer cell
It attaching map p_s: attaching map p_m: The integrated cell by these can be expressed by the following formula. As a result, customer cells become sales site cells and product information cells.
Joined to, customers get product sales site and product detail information
You will be able to. Site 2: "The online shop wins the customer's candidate list
It At this time, I also want to get information about the customer himself
It " Here, it is assumed that the list has three customers. Web first
Find the customer who sent you the candidate list above. Customer 1 cell B_c1 ⁶: [Name] [Age] [Gender] [Address] [Title] [Author
Person] Purchase candidate cell B_m1 ^Two: [Title] [Author] Customer 2 cell B_c2 ^Three: [Name] [Address] [Title] Purchase candidate cell B_m2 ¹: [title] Customer 3 cell B_c3 ^Four: [Age] [Gender] [Genre] [Publisher] Purchase candidate cell B_m3 ^Two: [Genre] [Publisher] Product information cell: B_m ⁸ [ISBN Code] [Title] [Author] [Publisher]
[Genre] [List price] [Issue date] [Pre-published] Selling site cell: B_s ⁶ [Title] [Author] [Publisher] [Sales Price] [Sales
[Number] [Stock] A customer cell is set as a purchase candidate cell and other parts
Divide. cell decomposition ma
pf_c1: identification function g
₁: cell decomposition map f
_c2: identification function g
_Two: cell decomposition map f
_c3: identification function g
_Three: In the product information cell having the same value as the purchase candidate cell
Identify the attributes and attach the three customer cells to the product information cell.
I will. attaching map f_m1: attaching map f_m2: attaching map f_m3: The integrated cell by these can be expressed by the following formula. By obtaining these from the sales site, online
The shop is the product that the customer wants to purchase and customer information
To get more information about the product
Can be earned. Site 3.1: "When online shops do data mining,
Find the data you need from a large amount of data. By this
Select specific customers who will be effective for specific products
It " Here, consider three products. Items to look for: [Name] of the customer who purchased the specific product
[Age] [Gender] [Address] First, as a table of products A, B and C,
A cell of sex 0 is searched from the data space. At this time B⁰ ₁, B⁰ _Two, B⁰ _ThreeAre each B⁰ ₁: Product A table B⁰ _Two: Product B table B⁰ _Three: Product C table And Also used later in this situation
Is Bⁿ ₁, Bⁿ _Two, Bⁿ _ThreeAre products A and
The information is about B and C. Next, the customer who purchased these products
The cell with the attribute number 1 that contains the [name] of
Search from the data space. With the work up to this point, the one-dimensional section with the customer's [name] as an attribute
X consisting of le¹Is built. Next, information "age" is added. Buy
The customer's [age] is included as an attribute, and
A cell with 2 attributes including the possible attributes from the data space
Search for. It With the work up to this point, use the customer's [name] and [age] as attributes
X consisting of two-dimensional cells^TwoIs built. Next, the information "sex" is added. Buy
Includes the [gender] of the customer as an attribute, and
From the data space
Search for. Take a response. Assign the customer's [Name], [Age], and [Gender] with the work up to this point.
X consisting of three-dimensional cells^ThreeIs built. Finally, the information [address] is added. Purchase
It includes the [address] of the customer who made the
Is the cell with the number of attributes 4 including the attributes that can become a boundary a data space?
Search from Take action. With the work up to this point, the customer's [name], [age], [gender], and [resident]
X] consisting of a four-dimensional cell whose attribute is^FourIs built,
It is possible to understand the data of the necessary items related to these.
Wear. For example, by checking the correlation from here
For example, “A person who purchases product A is a man in his 40s living in Tokyo.
There are many
Can derive a customer model. Site 3.2: "When online shops do data mining,
Find the data you need from a large amount of data. By this
Promote additional purchase for each customer by knowing the purchasing tendency of products
And decide which products are effective. " Consider three customers here. Items to look for: [Title] of books purchased by a specific customer
[Author] [Publisher] [Genre] [Price] First, as a table of customers A, B, and C,
A cell of sex 0 is searched from the data space. X⁰= {B⁰ ₁, B⁰ _Two, B⁰ _Three} At this time B⁰ ₁, B⁰ _Two, B⁰ _ThreeAre each B⁰ ₁: Customer A's table B⁰ _Two: Customer B's table B⁰ _Three: Customer C table And Also used later in this situation
Is Bⁿ ₁, Bⁿ _Two, Bⁿ _ThreeIs customer A,
The information is about B and C. Next, regarding these customers,
A cell with an attribute number of 1 containing [Title] as an attribute
Search from the data space. Assign the [Title] of the book purchased by the customer in the work so far
X consisting of one-dimensional cells¹Is built. Next, the information of [author] is added. The customer
Includes the [author] of the purchased book as an attribute, and
The data space is a cell with 2 attributes including the attributes that can be the boundary
Search from. Take With the work up to this point, the [Title] and [Author] of the customer are set as attributes.
X consisting of 2D cells^TwoIs built. Next, the information "Publisher" is added. client
Contains the [Publisher] of the book purchased by
And the data of the cell with the attribute number 3 including the attribute that can be the boundary
Search from space. Take action. With the work up to this point, the [Title], [Author], and [Publish] of the customer
X consisting of three-dimensional cells with the attribute "company"^ThreeIs built
It Next, information of "genre" is added. Customer
Include the [genre] of the book purchased by the customer as an attribute.
In addition, a cell with 4 attributes including the attributes that can be the boundary
Search from the data space. And X^ThreeTake action. With the work up to this point, the [Title], [Author], and [Publish] of the customer
Company] [Genre] as an attribute X consisting of four-dimensional cells^Four
Is built. Finally, the information [price] is added. Purchase
It includes the [price] of the customer who made the purchase as an attribute, and
Whether the cell with the attribute number 5 including the attribute that can become a boundary is a data space
Search from ∂B⁵ _iAnd X^FourTake action. With the work up to this point, the [title]
5th with attributes of [person] [publisher] [genre] [price]
Original cell X⁵Is built and required regarding these
You can grasp the data of the item. Correlation from here
By checking the person in charge, for example, "Customer B says XX
I often buy the author's books. "" Non-fiction works
"I often buy" "I buy only books up to about 1000 yen
To know the purchasing trends of a particular customer
It is possible to encourage customers to purchase additional products that meet their conditions.
You can get a means to aim for sales promotion with. [1-6] Conclusion In this paper, we use a cell theory for electronic commerce.
Presented the simulation model. As presented here
If you apply cellular modeling to
You can model any situation
It We would like to study how to implement this model in the future.
Get [Second Prerequisite Technology] [2-0] Summary The cyber world has a large amount of data and is extremely complex.
And changes repeatedly with speed. Introduce cell model
This allows us to properly characterize equivalence relations,
Furthermore, by introducing the homotopy theory,
Save the correspondence information of the elements of complex data operation words, and
Elements that are otopic invariant can also be incorporated. So
This makes it possible to handle complex schema changes, and
Large amount of data is effective even under the conditions often seen on
A model that can be operated efficiently can be provided. [2-1] Purpose The cyber world is on various local websites
Composed as a result of the activity. The data on the Web is
Because it is updated and changed by people all over the world,
The data changes with extreme complexity and speed. Present
Currently, the database contains huge and irregular data on the Web.
It is required to utilize the data and respond to the changes.
Introduction of Cellular Model
Defined at the level of theory and defined only at that level
Relationships between tables that could not be done, relationships within tables
homotopy, quatient mapping
(Identification function)
By dynamically configuring, to the complicated changes in the cyber world
Appropriate modeling to support and accurately characterize common parts
Of a data model that enables
For the purpose of provision. [2-2] Requirements for the cyber world The cyber world is intentional or natural on the Web
To be designed. Also, sometimes the world is generated without design.
It is a world group, a world consisting of only information. many
If relevant to activities on various local websites
Then, the cyber world is created. Cyber as information
The world is a real world or a virtual world.
And sometimes both. Cyber world is extreme
Because it repeats changes at various speeds and complexity,
Without proper modeling, the cyber world would be chaotic
It will be beyond the control and understanding of people.
I am thought. Effectiveness of using large amounts of data in the cyber world
Efficient data manipulation and flexibility to handle complex changes
And need modeling that accurately characterizes common parts
To do. [2-2-1] Cyber world and relay show
Null model Here, the flexibility is mainly mentioned above,
The relational model used by
Think about the problems of managing the world. Relay
The information model is that the information manager knows the data dependency.
It is a data model that is based on this assumption. That is,
A world where information is constantly changing, like the Iber world
Well, not all attributes have to be pre-defined
It is not possible to determine the dependencies between the tables first.
It is a suitable data model for modeling such a world
Not really. Also, to handle complicated targets, relay
The use of regional models is weak in terms of descriptive power and is particularly relevant.
Difficulties in managing multiple actual conditions with a database
There is. This is because the relational model is based on set theory.
Because of its abstractness, it is possible to accurately project reality.
Because there is no. Also, when managing information on the Web
In addition, each local Web Site has its own pipe.
The database managed by the administrator can be integrated and managed.
I can't. This is because the data management method
Need to be redesigned to be different in different databases
It is. In other words, the relational model is
It can be said that it is not possible to cope with the complicated changes in. Also
From the viewpoint of data manipulation, the input rule specified by the administrator
You must enter according to the rules. In other words, in flexibility
It can be said that it is lacking. [2-3] Cellular Model
To the relational model Here, in terms of flexibility and efficiency of data manipulation,
Cellular Model and relational model
Let's compare. Cellular model smell
For example, by using the homotopy theory,
Cell space can be configured. This is a relational model
Is very effective in terms of flexibility compared to
Can be said. Relationship of data in relational model
Connectivity is represented as a table. Relationship between tables
Based on data dependencies by database administrator
Is normalized. In other words, the relevance of data is
The data administrator pre-defines it as a data dependency.
Fixed and fixed. Also, the relational model
In this case, since homotopy is not saved, the table
By projecting (projection)
After taking out, it is possible to keep the relation with the original table.
Can not. In the following, a relational model will be described through a concrete example.
Data manipulation by Dell and cellular mode
Let's look at the difference in data manipulation depending on l. Data manipulation is
Each model is used to
Student name and phone number corresponding to raw number 00r3242
Ask for. At this time, the hands in the relational model
The order is 1. Select 2. Projection (see Figure 3) In the case of cell model
Is 1. Projection (projection, cell decomposition) 2. Select (see Figure 4) The required attributes can be obtained by the procedure. this
The relational model is, as mentioned above, the table
As shown in this example
To extract necessary information from a table
If you think about the data by keeping tuples in the table,
Keep the relationship. Therefore, when selecting, all
All attributes are referenced to the student number
By selecting the tuple with the number 00r3242,
Whereas attribute (column) name and phone number are projected,
In the case of the cell model, first, the attributes required by projection, students
Project the number (cell decomposition) and extract only the extracted attributes
, Select the tuple with student number 00r3242
I know what to do. And decompositio
By saving n map f,
py theory can be used, and the name and phone number corresponding to the student number
You can get the story number. This is a projection
action), only the cell space containing the required attributes
Even if divided, the relationship with the original cell space is homotopic.
Possible by being able to relate by y theory
Operation. In other words, use the homotopy theory
By the relational model, a fixed table
In the cellular model, the relationship between
Equivalent cells are empty according to
You can disassemble the cell from between, and respond to the user's request.
Thus, the cell space can be dynamically constructed. Also, in this example,
Consider the efficiency in terms of conservation of the mapping by theory
View. Preserving the mapping by homotopy theory is
In addition to guaranteeing spatial relationships, reuse of cell space
It also helps. For example, suppose in this cell space
In contrast, the name of the student corresponding to student number 00r3232.
When you ask for the phone number and
The student number 00r3232 is projected as the name by projecting the raw number.
Check all attributes by selecting the phone number
Therefore, it is not necessary to project the attribute of the student number,
By going back, the relationship between cell spaces can be derived. this
For the relational model,
Even if the operation requires a shape, the data operation is
I have to repeat the work, but the cellular
  In the case of model, follow the saved map
Efficient because the cell space used once can be reused
It is possible to process various requests and effectively use information. here
So, an example of operation for one table (cell space)
As mentioned above, a large amount of data is
Since there is, the amount of operation will naturally increase. Therefore, the preservation of the map
Reuse of cell space by existence is common in the cyber world.
It is thought that the effectiveness will increase. [2-4] Cell model in electronic commerce
Typical example of application On the Web, new services are available from the already existing cyber world.
Encounter a situation where you need to make up the Iber world
It Such a situation is caused by electronic commerce on the Web.
Can be seen. As a specific example, typical e-busines
The customer in the electronic commerce seen in s etc. is on the Web
So, let's consider an example of searching for products based on conditions and purchasing from the store.
I will try. In this example, the business execution entity is the business
To identify common areas with customers, products, and stores
The customer information, product information, store information
Consolidate information. In this example, customer C is on the Web
So, I think I buy a TV from a store. On the web
Under these circumstances, customers and product TV,
You don't have to get the exact information of the stores you sell. This
Of the customer's cyber world in the q-dimensional cell space B_c
^qAnd B for product cell space_t ^rAs the r-dimensional rhino
The bar world, the cyber world of stores, the p-dimensional cell space B_a
^PIt is defined as Here, in order to materialize each cell,
The space is defined as follows. Store cell space: B_a ^P(P = 9) = B_a ⁹= {Store name, EM
AIL, bank account, URL, product name, product category, price,
Size, manufacturer} Product cell space: B_t ^r(R = 5) = B_t ⁵= {Product name,
Product classification, price, size, manufacturer} Customer cell space: B_c ^q(Q = 5) = B_c ⁵= {Name, residence
Office, MAIL, desired product category, desired price} Here, the store cell space B_a ⁹And product cell space
B_t ⁵The relationship between is described. Various local webs
A collection of information on the cell space of the store released above,
Store cell space B_a ⁹And In other words, the cell space of the store is
Information disclosure to buy and sell products on the local Web
Cell space B that collects the data of the cell space of the store_a ⁹
Shall be managed in the database. Also
All local stores have the same number of attributes in the web space
I can't say that, so here is the 9th dimension of the cell space of the store.
Is the store cell space on the local Web with the most attributes.
Defined to be the same as the number of attributes,
If it does not exist, the attribute is undefined and the logical symbol ⊥
It shall be represented by ("undefined"). Since
By the above definition, the store cell space B_a ⁹Is a local W
A cell that contains the product information on eb as a partial cell space
It becomes a space. That is, the product cell space B_t ⁵Is the store
Le space B_a ⁹It can be seen that there is the following relationship with. B_t ⁵⊂
B_a ⁹ From this relationship, the cell space of the store can be set as the product information.
Cell decomposition into a cell space and a cell space containing store information
It FIG. 5 shows how the cell is disassembled. As described above, the product cell space B_t ⁵Is the store
Cell space B_a ⁹Established as a cell space that collects product information of
To mean. Fm used here is decomp
positionmap, where m is an identifier
is an ion function. As mentioned above, this strange
By storing the mapping in the form, this cell
It is possible to keep the deformation of space to be homotopic.
In other words, the product partial cell space obtained by cell decomposition
The cell space can be obtained as follows. This cell is empty
Mapping f used for transformation between_mAnd m are surjecti
Since it is a ve and continuous transformation, it is a reverse copy.
There is a statue. From the above relation, the deformation of this cell space is h
It is an omotopy equivalent. Therefore
The integration means separates the product cell space and the customer cell space from each other.
By decomposing, the common part is extracted and the information integration is cel
An integrated cell is constructed by lattachment.
And the integrated cell and the cell space of the store are homotop
It can be related by using y theory. Since
The specific procedure is shown below. Used to transform the cell space below
All mappings that are used shall be preserved. The customer said,
Product is offering the best price
Identify yourself as a product you are interested in. " this
Information schema in the situation, identification operation, cell decomposition
Is formed by. 1. Cell decompo
position) Cell space for each customer and product, B_c ⁵, B_a ⁵, E
quivalent cells B_e ^t(E = a, c)
And cell decomposition into a cell space containing the rest of the attributes
It At this time, the customer C purchases the product at the price
Products under two conditions: = 30,000 yen or less, product classification = TV
Is searched on the Web, equivalent
cells B_e ^tIs B_e ^TwoBecomes Fc and ft used here are decompositio
n map. Furthermore, g, k, are identi
fiction function,
Is defined as FIG. 6 shows how the cell is disassembled. 2. Next, cell attachment
A cell is constructed by Customer cell space B_c ⁶And goods
Cell space B_t ⁵The equivalent cell
s B_e ^TwoAs a qurtient space, a
As shown below by tachingmap p
Get a response. Then, the Integrate cell was constructed. FIG. 7 shows how cells are constructed. Furthermore, this Integrate cell
And the store information cell space B of the store_a ⁵Relationship with
The information function m In addition, the cell space of the store should be the same as the product information cell space.
Decompo used for cell decomposition into cell space including information
position map f_mUsing the inverse mapping of If you use the relationship Therefore, the final integrated cell space B is FIG. 8 shows how cells are constructed. In other words, in this situation, hometop
Relationship between store cell space based on y theory and product information cell space
Dating between tables in the relational model
Plays the same role as the association. However, Lille
The relation between the tables of the
Must be fixed when designing
Iga, relationship between cell spaces by homotopy theory
The attachment can be related dynamically. In this case
Depending on the conditions under which the customer searches for products, the product cell space is also reviewed.
It can be configured dynamically according to customer's request.
The obtained cell space, product cell space, and
The customer's cell space is also related by homotopy theory.
You can see that As a result, the structure of electronic commerce is
Can be configured above. And the customer is the one who likes the most price
Information on the stores that sell these TVs
You can get and sell the account number of the store and the product name of the target TV.
In addition, the store acquires the information necessary for buying.
By obtaining information such as name and email address
You can move to trading. In this case, the customer
It was explained on the premise that the search is performed according to the condition for
The customer specifies a store, browses the products handled by a store,
You may want to do this, but in this case, the store information
Information separation of the common part of the customer space and the customer space
Information is integrated by disassembly and cell attachment
Done by. Furthermore, the relationship between the integrated cell and the product information is
Integration by relating by homotopy theory
The cell space can be configured and customers can
You can browse existing products. In this e-commerce
Used decomposition maps and a
touching maps, identificat
By holding the ion function,
The cell space used in this electronic commerce is hot
Can be kept opic. To save the map
Therefore, it is possible to save the relationship of these cell spaces.
If there is a similar request,
Relationships can be easily derived by tracing the numbers
It Also, save all mappings in this variant
As a result, each operation is of course homotopic.
It can be treated as an equivalent in y. Protect this map
The cell space, that is, the information is transformed into various forms depending on the existence.
Even after setting, the relationship of the cell space is guaranteed and easily
It is possible to derive the relationship of the cell space. Tsuma
By storing the mapping in data manipulation,
And the transformation of the cell space is equivalent to homography
It is possible to determine that
It is possible to maintain relevance. Here, the integrated cell space B and each cell
From the relationship of the maps used to transform the space, homotopy
Prove the equivalence. Mapping p, fc, f
t, is continues and is subjective
And the existence of the inverse map is clear. Also as mentioned above
, M, f_mThe inverse map of m^-1, F_m ^-1Also exists. identification function m
From the reverse mapping of Then, Fc^-1, Ft-¹, Fa-¹Is obviously
H exists because it exists_c= Fc-¹  H_t= Ft-¹H_a=
Fa-¹Then, these transformations are FIG. 9 is a diagram showing homotopy equivalence. Further, this store information cell space B_a ^FourAnd the product
Information cell space B_t ⁵From the cell space B of the store_a ⁹Transformation to
Is as described above. From the above relationships, this e-commerce
The transformation of the cell space in the transaction is
It can be proved that it is quivalence. Also, all
The existence of the inverse map of all maps is guaranteed from the continuity of the map
Therefore, the intermediate stage of deformation is similarly guaranteed. Here, based on this case, cellu
Data manipulation by relational DBMS and relational model
Describe the differences in data manipulation at Dell. c
Illustration of cell space operation by ellular model
11 shows. For relational models, a collection of mathematics
Based on the consensus, all the links between the data in the table are
Based on set theory
Therefore, do not operate while keeping the tapple.
If it is not possible to maintain the relevance of the data, for data manipulation
When projecting the attribute of interest, the original table
The relationship with can not be maintained. A large amount on the Web
Cellula because it is thought that the data of
r Only necessary attribute data after projecting DBMS
The referencing procedure can be said to be efficient. Also, in the RDBMS, the database administrator
However, when designing the database, we intended the relationship between the tables.
Dynamically associated tables
I can't configure it. In other words, even inside the table
The relationship between the two is also fixed and depends on flexibility.
It The data on the Web is updated by people all over the world.
Therefore, the relations associated with the database design are also
It can change, but consider how to respond to it.
Then, it is impossible to deal with this huge and irregular data.
Can be considered. But the cellular DBMS is
As shown in this case, data manipulation begins with cell decomposition.
Extract the common part in each cell space by (projection)
put out. Then, the extracted partial cell space and the original cell space
Is guaranteed by homotopy theory.
(FIG. 11). That is, the relationships between data are dynamically configured
You can do so, such irregular changes on the Web
It will be possible to deal with. And this case
Each cell is decomposed into a cell space that contains the attribute of interest, such as
The common part of the space is called a qoutient space
By joining, the necessary information that meets the conditions can be obtained.
it can. Furthermore, the mapping preservation by Homotopy theory
Existence is the reuse of information in addition to the relationship between cell spaces
It can also be used for that. Here, in the cellular DBMS,
Let's look at the table operation. Each copy
The relationship between the image tables is similar to the mapping described above. [2-5] As a data model of the cell model
Effectiveness As an example, consider the effective use of the cell space just constructed.
It 1) The cell space can be reused. 2) It suffices to search this cell space. 3) Use the cell space in the middle of deformation for another purpose
I can do things. It is possible to use it. For each usage
explain about. 1) The cell space can be reused. If another customer searches for products under the same conditions,
Then you don't have to go through the same steps again,
By storing the mapping used for the shape, each cell space B_a ^P
, The function used for the above transformation, f_mDue to store information only
Cell space B_a ^P-rAnd product information cell space B_a ^rGet
You can, product information cell space B_a ^rCustomer information cell space
B_c ^qTherefore, the mapping used for transforming the cell space
From what I have, As shown in Fig. 9, the Integrate cell
B can easily follow, so customers from B like the most
All you have to do is select a product. This is a cell you use frequently
Very effective for space, for example the cyber world
In the above transaction, the store must deliver more products to customers.
I have to. At this time, in the cyber world, the store is a freight forwarder
Order and deliver the product to the customer. The store is e-bus
In order to do iness, you have to carry out such procedures many times.
Can be considered. However, decomposition
Save maps and attaching maps
Used by the store when first placing an order with the carrier
You can reuse the cell space to get the cheapest shipping cost from that cell space.
You can see that you can place an order with a shipping company. That is, in the condition
Therefore, the relationship between the derived cell space and the original cell space
Can be held, so
Use information effectively by using quivalence
It can be used. 2) It suffices to search this cell space.
Yes. The customer searches for the condition that the product category is TV and the price is 30,000.
I'm looking for a TV within the yen, then Integra
Let the aged cell be B. Customer E also does the same
Suppose you want to search for Bi within 20,000 yen. At this time, Integra
aged cell, B_JThen, this cell space B
_JIs the cell space B_I, Becomes a subset. B_J⊆ B Therefore, the customer D meets his condition from the cell space B.
You can find out what you need by looking for the product. Customer C
Relationship between Cell Space and Integrate Cell B
Since it is retained, it is easy to integrate cells for the above mentioned
Space B can be derived. Therefore, the inspection by customer E
The rope derives the relationship shown in FIG. 9 from the stored map.
It is possible to do things, and it is only necessary to search the cell space B.
Light 3) Another use of the cell space in the middle of transformation
Can be used for Even in the middle stage of these transformations, cell decomposition
Even if you take one operation, you used it for each,
Keep the decomposition map
By nature homotopy equivalence
Therefore, the cell space obtained by the decomposition is
It can also be used for various purposes. The assumption here is also homo
Prove that it is topy equivalence
It F now_c, F_t, F_aWas defined as follows. One of these mappings will be described as an example. Customer cell space
, The mapping F used in the integrated cell space_cIs fc / p / m / f
_m ^-1, Obtained by combining each mapping
Is a map and F_cBut, homotopy equiva
fc · p · m ·
f_m ^-1, The existence of the inverse map of each map is proven
So Then, from the continuity of the mapping sequence, the inverse map F_c1 ^-1Also
Because it exists Then, F_c1And H_c1Have the following relationship. Therefore, this modification is based on homophone equiva
It can be said that it is a lens. Similarly, the maps m, f_m
^-1The transformation of the cell space due to
It's proven that it is ivalence.
This map is a homotopy equivalenc
It can be said that it is e. That is, each cell space deformation
All of the intermediate stages of are similar in terms of "hotomopy".
Map F_t, F_aIs also the same. [2-6] Conclusion In the base technology, Cellular Model is h
Since it is built on the omotopy theory,
By storing the mapping for the transformation between
It is possible to dynamically form a relationship between cell spaces with respect to
The relationship between cell spaces is guaranteed by mathematical support
To be done. This has traditionally been fixed in the relational model.
The relationship between data that could only be treated qualitatively
Is the mathematical structure of the theory that
It becomes possible to relate dynamically, and in the cyber world
And consider the information after various transformations as equivalence
Flexibility, efficiency in your data model
I showed that I can have it. Effective use of information
Also in this sense, it is possible to deal with various cell space deformations in this way.
And give efficiency, such as guaranteeing the relevance of the cell space
Can be said. In conclusion, Cellular Mo
By using del, it can be changed with extreme complexity and speed.
Data model capable of responding to the cyber world
It has been shown that it is possible to provide [Third prerequisite technology] [3-0] Summary It is said that various transactions are currently taking place in the cyber world.
Sometimes the situation is constantly changing over time. Sermo
Dell has the flexibility to adapt to the sudden changes in the cyber world
Can be modeled as This is based on its underlying invariant
The equivalence relation makes it possible to combine cells dynamically.
It When combining cells, in each cell space,
Need to be identified. Therefore, the cell ID is used.
When designing this cell ID, the outline of the scope is important.
Just in case. Find the cell scope (effective range) and
It is possible to specify the cell by adding it to the ID
Become. This uniquely identifies the cell in the cyber world,
It is possible to dynamically perform cell combination. [3-1] Introduction A database is a real-world database built in a computer
It is a model. It may be called a map of the real world. Yo
Therefore, in order to build a database, the real world is modeled.
Need to be converted. Currently, most DBMS are relay
It is operated by a relational DBMS. Relationer
The data model was developed by E. F. By Codd
This is the data model proposed by Cartesian product of set theory in mathematics
And all data are mathematical relations
Corresponding to is represented in the table. This relational data
Models allow database administrators to recognize data dependencies
Model the data assuming that you are doing This
This causes the following problems. ・ A space where the situation changes suddenly like a web
Cannot be expressed. ・ Because only closed space can be expressed, each data
There are differences in how to manage data on a base,
Data management and data manipulation between different databases
Yes. Due to the above problems, relational DBMS
Now, whenever the situation changes
It must be redesigned. Therefore, the situation changes
Every time it happens, it costs a lot of money. Effective data that can solve such problems
The data model is the cellular data model. Cellular day
The data model is a cell model with various equivalence relationships based on invariants.
Can be done dynamically. Furthermore, the change is called a homotopy invariant.
By saving it, you can always change the situation like on the web.
You can model the space in which the transformation occurs. With this prerequisite technology
Is used to identify cells when performing cell merging.
Discuss Le ID and scope. [3-2] Cell space theory Cell space theory incorporates the concept of abstract hierarchy
By defining an invariant inherited in the hierarchy
Therefore, various situations can be modeled. In this cell space theory
The outline is shown below. [3-2-1] Cell space structure About Cellular Spaces
Will be outlined. First, the cell is topologically n-th
Original Open Ball IntBⁿTopology equivalent to
-In the space, n-cell eⁿIt is written as. X?
Et al., A finite or infinite cell row X by cell joining^p
Can be configured to be inductive. X
^pIs constructed to be a subspace of X and is searched for by the integer Z.
Attract. Thus obtained {X^p| P ∈ Z},
It is called filtration. X^p  covers X
(Or X^p  is a covering of
X), that is, X = ∪_{p ∈ z}X^p X^p-1Is X^pA subset of, that is, X⁰⊆ X¹⊆ X^Two⊆ ... ⊆X^p-1X^p⊆ ... ⊆X In this way, the cell structure space obtained from X {X; X^p｜
pεZ} is called a filtration space. [3-2-2] Cell combination (cell at)
cell space structure Open n-cellⁿThe already built topology
In the interval X, both surviving and continuous
The cell structure space Y is
Can be built. Map f: X → Y is a surjective
Is (∀y∈Y) (∃x∈X) [f (x) = y] Means Mapping f is continues
That means "a subset A ⊂ Y is open"
  in Y if and only if {f
^-1(Y) | y∈A} is open in X ”
means. Is a tra called an attaching space
space from nsitity by equivalence relation
Divide into exclusive sum of uivalence classes
it can. One equivalence class
Notated as x / ~. Then x / ~ = {yεX | x ~ y} Is. All equivalence classes
If the set of X is written as X / ~, it means that
nt space or identificati
called on space X / ~ = {x / ~ ε2^x│x∈X} ⊆2^x Is. [3-3] Cellular data model A cellular data model is a data model that applies the cell space theory.
Data model. Cellular model is a hierarchy of abstractions
Based on the nature of various existing data models
Is included. For the sake of clarity, the relationer
It will be explained while comparing with the data model. Relationer
Relation in the data model is a set
Therefore, the cell hierarchy is abstractly inherited. Servant
So, at least the cell can be treated as a table. N
Since the table with attributes has n degrees of freedom,
Represented as n-cell. Relational data model
Depending on the degree of the relation schema,
(N-ary relation)
There is. The cell space theory has the concept of a boundary.
An n-dimensional close-cell is a cell with a boundary.
Ri BⁿAnd the n-dimensional open-cell is the boundary
Not having Is the boundary of And is an n-1 dimensional space. It becomes). (When n = 1, the boundary is B⁰Next to the genus
It has no sex. ) Cellular databases are open
Attribute is the primary key in the relational data model
Design to play the role of. In addition, identify the cell
In order to manage the cell, a cell ID described later is attached. [3-4] Definition of scope Without knowing it, people can
It incorporates the concept of co-op. In general, scope
Is a tree structure or graph, as seen in XML.
Take a structure. For example, identify the person "Taro Yamada"
In order to do so, “Yamada Futoshi of the Department of Electrical and Electronic Engineering, Faculty of Engineering, Hosei University
"Horo University, Faculty of Engineering, Department of Electrical and Electronic Engineering"
Add a scope (effective range). What is a scope?
Since it is a range to determine the object uniquely,
Two people in the Department of Electrical and Electronic Engineering, Faculty of Engineering, Hosei University "Taro Yamada"
If there is such a person, this scope is inappropriate
I can not specify "Taro Yamada" because I am sorry
It In such a case, "Hosei University Faculty of Engineering Department of Electrical and Electronic Engineering
"Taro Yamada for 3 years"
I have to start over. The scope in this case is shown in FIG.
It has a tree structure like. And this tree
"Taro Yamada" by adding all the scopes from parents
Can be specified. Also, this is the path from the parent of this tree
You can also see that. [3-4-1] In programming language
Scope This concept of scope is widely adopted in computers.
Has been Example of scope in programming language
Show. To identify variables in programming languages,
There is a variable scope rule. When defining a variable
Needs to know to what extent the variable is effective
And when referencing a variable, where is it defined
Need to know if the variable is The solution to this is
This is a variable scope rule. Taking the JAVA (registered trademark) language as an example,
Scope rule when referencing variables in JAVA language
Is defined as follows. 1. Look for blocks at the same level as the one you are referring to. 2. Find the block outside it in the method. 3. Find the variable in the class to which the method belongs. 4. Find superclass variables. 5. If not found so far, an error occurs at compile time.
Become The variables in the program are searched in this order, and the first one found.
Variable is referenced. Variables suitable for the following conditions
But not referenced. The scope rule is shown in FIG. [3-4-2] e-mail address
Kicking scope Next, consider the scope on the web. below
Take e-mail address as an example.tashiro @k. hosei. ac. jp The e-mail address can be uniquely specified on the web.
It is a thing. Looking at its structure, it looks like this:
@The previous name (tashiro) is the domain name after @
Must be unique among (k.hosei.ac.jp)
I have to. Therefore, the scope of the name before @ is after @
It is a domain name. Also, the domain name is on the web
Must be unique, so the domain name after @
Is on the web. Email address
The tree structure of the scope of the table is shown. The e-mail address has the following structure
And the domain name whose scope is the name before @
To add uniqueness on the web.
You can see that Name @ scope (domain name) Thus, the scope uniquely identifies the object
Sometimes it plays an important role. And the object
Object by adding scope to
Can be uniquely identified. [3-5] Cell ID Cellular databases allow equivalence relationships based on invariants.
Therefore, it is possible to combine cells (tables) on the web.
It At this time, it is necessary to specify which cell to combine.
It So, in the cellular database, specify the cell
The cell ID is used for this purpose. Relational database
Is the equivalent role of the table ID in. In the cellular database, the cell ID is the cell
It is designed to write in 0-cell of. 0-cel
l is a cell that has no independent variable. Sanawa
Well, it's not an attribute. ∂B¹= B⁰= K = cellID (k is a constant) 0-cell always contains every cell, so here
The cell ID can be written. In addition, the cell ID
The cell is managed by writing in the cell ID table. Plays an important role in assigning cell ID
The above is the scope mentioned above. Database on the Web
The definition of the scope in the table is as follows. Database
The device name is unique within the computer, so
The database scope is host (domain). SE
The cell name is unique within the database, so
Corp is a database. The attribute name is
Attribute, the scope of the attribute is table. This
Therefore, databases, cells, and attributes have scopes.
By doing so, each can be uniquely decided on the web.
it can. For example, the computer "db.k.hose"
i. ac. Database named "student DB" in "jp"
If you want to identify cells called “student cells”,
Add the scope of “” to the student cell and express it as shown in FIG.
It By expressing it like this,
The "student cell" can be uniquely specified. This is
And manage it by writing it in the cell ID table.
Then, it becomes possible to specify the cell on the web. [3-6] Case Study A company that sells products on the web adds customers
Compare the case with a relational data model. W
Product sales on eb is a map of customer data and product data.
It is ching. [3-6-1] Relational data model
In case of As an example, sell products using the following database
Suppose Own database Different for relational databases
Operations such as joining tables between databases cannot be performed. So
Therefore, in this case the product list of the new
From the store table,
It is necessary to add product data to the cable. Product data in the database of new customers
Is managed as follows. New customer product table At this time, the product
While managing the studio by dividing it into three tables,
In the database of new business partners, one product master
Because it is managed by a bullion, the product
The product data in the database of new business
I cannot insert it. This is a relational data
The database is managed by the database administrator for each organization.
Recognize data dependencies and design your database
This is because it is done. In other words, relational
The database projects a closed space only for that organization.
Data can be managed only in that closed space.
Is. This is also a relational database
The maximum scope of
Cannot be scoped larger than the
Tables and attributes cannot be specified between databases
It can be said that it cannot be operated. In this example, the product data of a new customer is checked.
Export the product data according to your database.
Data and import it into your database
Although it can be better resolved, a database merger
Need to be redesigned, huge cost, development time
This is unavoidable. [3-6-2] Cellular database
Combined Cellular databases will incorporate abstract hierarchies
Dynamic equivalence based on invariants
It can be carried out. Therefore, in this case,
Use the product master as it is and combine it with the customer cell
You can Specifically, a cell is set as shown in FIG.
It is assumed that the cell ID table is used for management. In-house database
The cells at the source are unique within our database.
Therefore, the cell ID should be written and managed without the scope.
Be present. In addition, the cell ID table is actually
Information such as permissions is also described, but here it is easy
Therefore, we will not discuss that much. New transaction here
Add the previous product cell so that it can be managed. Above
As described above, add a scope to the product cell ID of the new customer.
In addition, the product cell of the new customer can be
To be able to identify. Here, according to the web rules
A certain @ is used to describe the cell ID. When selling an existing product, the existing product
Merge cells with customer cells. If you sell a new business partner product, the new business partner product
Merging cells and customer cells. In this way, the cellular data model dynamically connects cells.
Therefore, you can flexibly respond to such changes in the situation.
Compared to the relational databases currently in use
It can be seen that the total cost can be reduced. [3-7] Conclusion In this base technology, in the cellular data model,
Cell ID and scope for identifying cells when combining
Was discussed. Cell ID is attached to the scope
Therefore, the cell can be specified on the web, so
It shows that cell merging is possible even between different databases.
did. In conclusion, the database administrator
Assuming that you are aware of data dependencies
In the relational data model, each time a situation change occurs
It is necessary to redesign the database for
However, the cellular data model is built on cell theory.
Therefore, the cell space can be dynamically combined, and various
Can flexibly respond to changes in the situation. Finally, this time mainly
Comparison with the Rational Data Model
The comparison with the data model of will be a future subject. [Fourth prerequisite technology] [4-0] Summary In this base technology, the WWW server is applied as an application of the cell model.
User's eyes from the information accumulated in the database on the
A method to efficiently retrieve patent information that is consistent with
It [4-1] Transition of protection of intellectual property rights Traditionally, intellectual property rights are protected against new technologies.
And specialized in industrial fields such as machinery, electricity, and electronics.
It has been considered as something. Therefore, prior art search and related techniques
The target of the surgical investigation is also specified in that field, and the trend of peers
Search or search for limited materials such as publications.
With, it was possible to collect information to some extent. However, in recent years, technology has become more complex,
Only limited material surveys such as those mentioned
There is a fear of conflicting with profits. Especially in financial relations,
Services on the Internet, trading network,
In addition to using computer-related technology,
Services are also diversifying. In addition, the business
Each model's unique service called Smodel is developed, and
Now that new technologies are combined, it becomes possible to obtain patents.
Later, an environment in which we have no choice but to pay attention to trends in the rights of others
Is coming. Therefore, the cell model which is the information combination model
To enable technological research across multiple fields,
Efficiently searching for technologies related to each other
Create a database from the result, and enable update and change
This is the purpose of this base technology. [4-2] Transition of technical survey method Publications are the only conventional method of technical research.
It was general. Also, the subject of novelty search by the JPO
Publications that have been distributed and whose publication date is
The abstracts of surgical papers were often used. this
Is a means available to examiners of the Patent Office, and
As a material for determining whether or not to approve the patented invention
And because it is useful. However, in recent years, the Internet has exploded.
It became popular and its value as a source of information could not be ignored.
There is. In particular, compared to publications, research is cheaper and requires less labor.
Since it is possible to announce the research results, each company has its own
The number of cases of posting company technology is increasing. In recent years,
The number of visits to the site has increased and
That the eligibility for
Have come to be used as a target for novelty research.
There is. In such situations, prior art searches and
Use of the Internet as an information source for research on trends in rights
In addition, a database is constructed based on that data.
The demand for building is on the rise. The conventional database is a relational
This database is often used
Data attributes and their relationships are predetermined
And it is difficult to change them later.
A new model database that facilitates work
Creation is required. [4-3] Database using cell model
Build The cellular model was created by Kunii, one of the inventors.
The model was introduced in 1999, and the actual local information
It is possible to combine information and global information,
Information space construction model that can project all phenomena in the bar world
It is Le. (1) Cell model As a cyber world model, “cell structure space level
"Is based on the cell space structure, but
Iverworld space is a conceptual and data modeling
Than those based on graph theory levels common to
It is a general foundation and has an information model with or without boundaries.
As a missing cell, by recognition and by calculation
Can be identified. Bordered cells are “closed
Cells that have no boundaries are "open".
n-dimensional cell, n-cell is an n-dimensional sphere that is an integer,
It is a pology equivalent space. Open n cell eⁿAnd
B closed n cell^{n Is B n}, Which is the (n-1) th dimension.
The cell model allows the cell dimensions and connectivity to be universal quantities.
By maintaining it, cell combination and cell disassembly are possible. of
Object identification is often referred to as identification mapping (often a quotient map
Called system). here
Then, dimension means the degree of freedom. Later, we
Database schema join and schema decomposition of
Demonstrate a special case of cell merging and cell disassembly
You Let us give an example of dimensions. For example, Cyberwa
Objects that do not have an attribute
It doesn't have the freedom to change,
The dimension in the object is zero. we,
This is achieved at the presentation level. attribute
Is a particular property or property that is unique to the object
, Which are mutually independent sets. It has one attribute,
Objects that can change the value of the attribute of
Can be expressed. Similarly, 2 and 3 degrees of freedom
Objects with attributes 2 and 3 have
It can be represented as a surface and a sphere. n degrees of freedom
The mochi attribute is n, and therefore its dimension is n.
The object is to express this as an n-dimensional sphere.
You can The relational model is relational
Expressed as an object with n attributes as a keema
Can be embodied as a table with n columns. The connectivity is called the attaching map.
Defined by continuity and surjective mapping. Function f
Is a mapping in which X → Y is surjective, (∀y∈Y) (∃x∈X) [f (x) = y] It means that. X → Y is continuous due to the function f
Mapping means that A, which is a subset of Y, is {f^-1
(Y) | y∈A} is open at X and then
Only means open in Y. Exclusive sum of topological space X and Y
Get by attaching It is written in. ~ Are equivalence relations. The equivalence relation is
, Symmetric, reflective, transitive
Relationship. It may be a set theory equivalence
Yes, it may be topological equivalence or geographical
It may be an equivalence relation, or homotopy
It may be an equivalence relation. Its transitivity is empty
Is divided into exclusive sums of subspaces called equivalence classes
It Here, a general definition of the attaching map is given.
Describe righteousness. Let the set of all equivalence classes be X / ~
Known as X equalization space. X / ~ = {x / ~ ε2^x│x∈X} ⊆2^x The attaching map f is a surjective and continuous map.
, This is because when embedding information in the web,
Later used for information schema integration and information integration.
Another case. S^n-1Closed n-cell BⁿBoundaries of
That is, ∂BⁿAnd The following equation holds there. S^n-1= ∂Bⁿ= Bⁿ-IntBⁿ= Bⁿ-Eⁿ The attaching map f is targeted and continuous.
If you say f: S^n-1→ X The additional space Y is defined as the equalization space as follows.
Be done. The two homotopy maps f and g are If the cyber world is a topological space, X
We are J. H. C. Adjust according to whitehead
Cell X, a subset of X indexed by the number Z^pof
Finite or infinite sequences can be constructed recursively
It Applications have significant cell space. So
These include cyberworld complexity and manifolds.
I'm out. (2) Web information modeling What We Must Do for Web Information Modeling
The first thing that does not happen is how the cyber world originated.
To understand what has happened and what they are,
As an information world shared with Web
To characterize the essence of form. Often a cyber
World X is local and diverse on many websites.
There are no cases of being created on the web as a result of dynamic creation
It Unlike the company information,
It is assumed that there is an information manager who provides a set of
You cannot do it. Through embedding information,
This is a special indication to the local website what X is
Information can be embedded. Of course we
Does not embed information on its own. Information is
After browsing the website, the embedded information
Information and the integration of them
Gather the information we want in the news. This
This is commonly referred to as "information embedding by design".
It is the embedding of the information that is stored. Because embedded
As information about what is or is not embedded
This is because there is a regular set that applies. That
Integration guides such as what and how to integrate
It will be a design guide. Information embedding on the Web is described in the above-mentioned Whi.
According to the scheme of tehead, the local
From the web world to the global cyber world,
It How the embodied integration is done and the n-dimensional support
Everworld XⁿIs described in more detail.
Therefore, the web search and integration process will be described. Concrete integration consists of two phases.
In the information schema integration phase and the information integration phase
It The first phase, the information schema integration phase,
It is processed like. 1) Zero-dimensional cyber world X⁰To create
All cells B that do not have the attribute of Busite⁰ _iSearch for
It X⁰= {B⁰ ₁, B⁰ _Two, B⁰ _Three,…, B⁰ _j} 2) One-dimensional cyber world X¹Create
One of the websites to do Join through the attaching map F. By that
And various one-dimensional cyber world X¹Can get
it can. 3) By repeating the search and integration,
(N-1) dimensional cyber world
You can get it. X^n-1Has n-1 attributes
One. n-dimensional cyber world X with n attributesⁿTo
N cells B with n-1 attributes for integrationⁿ _iof
Search for all combinations. [0131] n cell Bⁿ _iBoundary element of each of n attributes of
To match each of the (n-1) attributes
The (n-1) th order already created by the
Original Cyber World X^n-1Attach to. More than
Through processing, various n-dimensional cyber world XⁿBut
can get. The above is the details of the information schema integration. Next, in the information integration phase,
Is pretty simple. This is based on the design guide
And attach cells when integrating schemas
In these steps, all instances are
Included in the cyber world created by the attachment
It is to check whether it is OK or not. In the range described above, identification is an equivalence relation.
It is done by the equivalence class based on the engagement. Cell model book
This strength is seen in this respect. Pretty complex on the web
Although it changes rapidly due to miscellaneous, due to the integrated power of the cell model,
It provides a web information model with the theoretical basis of.
For web-based information systems, design guides
It can be local, such as a runnet, or
A global world like Dulles Cyber World
It is either. Design guides provide information based on cells.
It is a reusable resource of the information system. (3) Concrete case From now on, information construction and search with information on the web
Through the patent information search on the Web,
I will discuss about it. Information Schema Integration Phase In the information schema integration phase, first from the web
Cell X with no attributes⁰To search. Next, one-dimensional service
One of the websites to build Iverworld
Find all combinations of cells with the attribute. To this attribute
Is the technical field to be searched, for example, the network
Use information such as network, computer, and finance. Them
The exclusive sum is calculated by identifying the individual boundary elements as X⁰
Join to. Finally, repeat these searches and integrations
As a result, the (n-1) -dimensional cyber world
It can be obtained through inference. This repeated line
The specific search target attribute is the specific key to be searched.
Use words. For example, to search for related technologies,
For network search, application layer, TCP
If you are searching for financial targets such as IP / IP technology,
Information such as service names for online trading, securities, stock trading, etc.
To use. Then, the exclusive sum of them isⁿN of
(N-1) of the boundary elements of the
Attaching maps that match each of the attributes of
The already created (n-1) -dimensional cyber world
X^n-1Attach to. Through the above processing, various
n-dimensional cyber world XⁿIs obtained. The advantage of this model is that it is equivalent to cell search.
By joining by the clerk, the joining relationship of the information you want
In this respect, conventional data
It is different from the database. Traditional database
In, do not define the schema when building the data.
And any changes to that schema will come later.
Re-normalization is required even when it is necessary, and it is easy
There is no. In addition, conventional databases and relational data
Query must be optimized in the database
Also, in XML, a style sheet is created separately.
To obtain the information on the style you need.
Becomes However, in the cell database,
Introducing the concept of building a new cyber world
It is necessary to create a new style as described above.
It does not become. Moreover, specific attributes from various technical fields
It is suitable for searching the data
Multiple technologies, such as realizing work through a network
In today's operating environment, the
The lit is big. Information integration phase It is in the information integration phase, but based on the design guide,
When integrating the keema, attach each cell to the cell.
Cell attach all instances in step
Whether it should be included in the cyber world created by
Check by checking. The cyber word obtained by the above-mentioned processing
Ludo is a (n-1) -dimensional cyber world
Attack to match each of the attributes that are
Since it is attached by the teaching map,
If there is a large amount of information in the technical field
Waste considerable time and effort in building a field. But,
If you narrow it down in the information integration phase, each of the attributes
Prioritize them or set required equivalence relationships
Equalization by setting the equivalent attributes
Cell joining by function, thereby reducing the amount of information
And the desired cyber world can be easily constructed.
It becomes Noh. Also, in practice, the processing depends on the application.
In this case, instead of rebuilding the whole,
The world can be constructed and optimized by attaching maps.
Becomes (I) Web state model Next, from the existing cyber world, a new cyber
Consider the case of building a field. In this case, information skis
Are formed by the cell disassembly process according to the identification process.
It When the cell disassembly process is performed with the map f, This map is an exclusive sum of existing n-dimensional cells Bn
Map to. The attaching map g_iIs given
Has been. The attaching map g_iIs surjective
, A discriminant that indicates continuity Meet Here, ∂ is repeated n-ui times. This state model is as follows. (A) Cell disassembly: s cell B as shown in FIG._s ^pTo
Network company S site, c-cell Bc^pFinancial
Related company c site, m cell Bm^pComputer related
Let's say it is the site of company m. These cells are the network
Web-based services based on equivalence relationships for
It was identified by Ito. Processing that cell disassembly
Is Is. Here, g, h, and k are the following discriminant expressions.
is there. g: ∂ ... ∂B_s ^p  → B_s ^t  (∂ is repeated pt times
Be done) h: ∂ ... ∂B_c ^q  → B_c ^t  (∂ repeats qt times
Be done) k: ∂ ... ∂B_m ^r→ B_m ^t  (∂ is repeated rt times
) (B) Cell combination: Cell decomposed in (a)
According to the attachment map below.
It p_m: ∂ ... ∂B_c ^q→ B_m ^t  (∂ is repeated q-t times
) p_s: ∂… ∂B_c ^q→ B_s ^t  (∂ repeats qt times
Be done) Due to the above, equivalence relationships are established across multiple technical fields.
More cell disassembly and cell joining can be achieved.
The information can be saved in the database. Also,
This data is the attachment map that defines the equivalence relation.
Optimization is possible. (Ii) Implementation In order to implement this model, a special interface
The provided XML structure applies mutatis mutandis. As mentioned above, XML
Is a database containing XML data in an XML file.
Form the source. But extract specific information from it
When doing, use a style sheet such as XSLT,
The data format needs to be redefined. This is its raw
The reason for this is that it is based on document information management. Utilizing this XML for the cell database
In order to do this, you can flexibly define the style of XML data.
Just do it. To do this, information extracted from the web
To use XML tags for the search information
, A tool that can convert to XML format file
Just do it. Today, with technologies such as Xquery,
Efficiently processing various XML format data on the Internet
It is possible to build a technology to select and convert to a unique data structure.
It is recommended. However, with this method alone, XML
Extracting unformatted information from the web
Is impossible. Therefore, with the base technology,
In order to solve the problem, data that is not described in XML format is detected.
Proposal of a method that enables searching according to the cell model described above
To do. (1) Cell disassembly   Search for the "technical field" you want from the web information
Identify the target area and add the tag for that "technical field".
The XML data hierarchical structure (library.xm
l) is created. The contents are shown in FIG. (2) Cell combination   The part that has the desired “keyword” from the target
Specify the minute and add the tag of that "keyword"
Append on the file. The contents are shown in FIG. (3) Example of information retrieval (using Xquery
for)   Created with. Hierarchical structure of XML data (li
barry. of the Xquery inquiries
Identify the target part and store it in a variable for further processing.
To end. Thailand for papers whose author is "TLKunii"
When querying for tor and embedding the result in the result element
In this case, the description in Xquery is as shown in FIG.   Filter the results of according to the query conditions.   The result of re-transforming the structure of the filtering result
Create a tree. Through such processing, the desired service
It is possible to build the Iberworld. [Fifth prerequisite technology] [5-0] Summary The educational process is a cellular model based on educational data
Recording and accumulating in cyber space by database
In the trial and error process during learning, homotopy reproducibility
Can have As a result, effective practical education
Show that the model can be built. [5-1] Current State of Education in Japan Education in Japan today is due to post-war reform of school system.
Japan's society and economy since the 0's (Showa 35)
The rapid development of
It has a base. Compulsory education has taken root, so go to high school
The enrollment rate is 97.0% in 2000, and it is globally
Even when compared, it shows a very high value. Education based on such high growth period
However, the surrounding environment has changed drastically now, and the turning point is approaching.
I am. In the early stages it was done using textbooks
However, with the rapid growth of television, OHP,
However, it has been performed using a computer. In fact
In addition, the Ministry of Education has
Education due to intensification and "informatization" such as the Internet
We have begun to “reform” the system and content. Specifically
Has reviewed the Basic Education Law and School Education Law since 2000.
First, we are making revisions. From 2002, the new learning guidelines will be implemented.
Is given. Among them, one of the aims of the Ministry of Education
And the following are listed. "Find your own problem,
Manabi from the water, thinking from the water, proactively thinking, proactive
To develop the qualities and abilities to resolve problems better
It In order to realize this aim, the current education is
What do you need? [5-2] Teaching that you can find tasks yourself
Rearing method It is often said that the educational method up to now is a stuffed type.
It Without being able to see what the final goal is, as it is taught
It has become. However, “I think myself
In order to achieve the goal,
Education is needed. Individual goals clearly
If you know, naturally think what you need and what you don't
To be Then, in the limited time of education, the final goal
What kind of method is necessary for education with a focus on the goal?
Educators need to know the number of educational elements until reaching the final goal.
Grasp and clearly identify what you need and what you don't
And by arranging it efficiently
It [5-3] Adoption of top-down method How to efficiently conduct such education with an eye toward the final goal
The top-down method shows whether it is possible. This
Until now, education has accumulated educational elements from nothing
It was to go. Here, the top-down method
It will be compared and it will be a stacking method. The following is one of the authors (Country
Well). In the stacking method, when there are n educational elements,
When stacking each element,
Depending on the number of elements)
And then consider which one to do next.
It was a state. If this method is used, the final result
Consider two things n times, which is the number of elements, until
I have to do 2ⁿOnly the process is required
It FIG. 21 shows an image of the stacking method.
As you can see, the actual education is lined up as shown in the figure.
Note that this is not always the case. Not necessarily next
The number of items that can be stacked is not limited to one.
There are many. Choose the most efficient one
There must be. On the contrary, what is piled up and completed (final
Goal), and then associate educational elements with each other,
How many processes do you need to take apart? As well
If there are n elements in, the last corresponding to each element
There are two ways to think about whether the elements on the goal should be exposed or not.
To do, n^TwoOnly the process of will be good. Figure
22 shows an image of the top-down method. Then, an efficient top-down method is performed.
What do you need to do? It is a database of educational elements.
Data and connect the data using a cell model.
By combining them, an optimized teaching method can be obtained. Also,
Educational elements and their curriculum are converted into data using a cell model.
By being based, homotopy reproducibility in the educational process
It becomes possible to have it. Cell model and homotopy
This will be described below. [5-4] Outline of cell model [5-4-1] Cyber space and cell model The cell model is a model announced by Kunii,
Originally the UNIX theory of pipe
It is an extension. A pipe is two processes (commands)
Can be connected by one pipe. Process (comman
Pipes connect things of different dimensions. It
Is the theory that can be applied to various things
It's Dell. In addition, the cyber world is the real space and
Ibar space and the fusion space of them. The cell model is based on the characteristics of the cyber world.
Then, after identifying the equivalence relation, the fusion space is created in the cell structure space.
Stipulate. As a result, the real space and cyber space are integrated.
Defining a cell adhesion model to match, and the amount of change
Can be stored as a homotopy invariant
Is the model that discussed. That is, in the cell model,
Can map events in the bar and real worlds
It is a theory. [5-4-2] What is the homotopy theory? Homotopy is an example of extension of a continuous function. Topolo
When we consider a continuous function in the Gee space, the existence of an inverse function
Argue that it will always be possible to return to the original state with
It is a theory. Below is a homotopy from Kunii's paper.
I will cite a concrete way of thinking. It is a continuous function in the topology space.
Consider X and Y. f and g are continuous functions of X → Y
It And if I = [1,0], then homotopy
(H) can be expressed as follows. H: X × I → Y In tεI, H = f when t = 0, and H when t = 1.
= G Becomes Furthermore, homotopy is an extension of continuous function
From the following can be said. i₀= X × {0} → X and i₁= X × {1} → X If it is, H | X × {0} = f_i0  And H | X × {1} = g
_i0 Is. [0164] It becomes possible to write Y, and the following situation holds. two
Discrimination function 1 for two functions f and h_xAnd 1_yBut 1_x: X → X and 1_Y: Y → Y Can be expressed as Becomes Homotopy equivalence is the ratio of topological equivalence to
It can be said that it is more general when it is solid. Homotopy equivalence
Is a topological change, even if the changes are not topologically equivalent.
The changes can be the same. The shape is completely different
Even if it becomes a thing, the process becomes homotopy
Therefore, it is possible to specify the homotopy equivalence.
It is possible to prove that. [5-4-3] Cellular using cell model
-What is a database? Traditional relational databases utilize keys
It was managed by doing. Key management is immediate
It is connected to data management, and when data is input / output
Has very high restrictions such as administrator authority. In contrast, the cell
In the database, the cell model is based on the cell model
Management is performed using information and homotopy. By key
Different from the relational database that manages
It is flexible and does not require an administrator, so
This is a database with a high degree of freedom. In addition,
The reversibility of homotopy is
Yes, it is safe to say. Cellular database
Negotiation system, cell information, cell definition information, cell
Has operation information. Cell information is a relational data
It corresponds to the database table. The cell definition information is
Record the open cell and border information of the cell. By that
Therefore, it is possible to manage data in object units.
In the cell operation information, operations related to homotopy of cell deformation
Record the work information. This makes the theory of homotopy alive.
The cell can be reused. [5-5] Consideration of specific example Here, using a concrete example, the top-down method is used.
Consider the educational method. The example examined this time is one PC
Windows (registered trademark) 2000 and Linux
Listed how to install and configure dual boot
It [5-5-1] Understanding of educational stage Windows 2000 and RedHat Linux 7.
The procedure for dual booting 1 is
It is divided into the following 5 stages. In addition, these
The tage is also an educational element. PC settings Windows installation Creating D drive Linux installation Startup settings and confirmation [5-5-2] Educational requirements at each stage
Understanding Making these five major stages more educational elements
It is divided into the following. PC settings ・ Understand the current situation (PC specifications). ・ Determine how to divide the hard disk capacity
It -Clean the hard disk. -Set the boot order using the BIOS setup utility.
Set. (FD → CD → HD) Installation of Windows -Start from the Windows2000 CD. -Determine the type of keyboard. (Japanese 106 keyboard
De) ・ Define a partition for Windows 2000
Cut -Format the partition with NTFS
It -Install Windows 2000. -Remove the CD and restart to start Windows 2000.
Confirm the movement. Creating a D drive (Windows and L
inux shared space) -Start "Administrative Tools" from "Control Panel". -Select "Disk Management" from "Computer Management".
Selected. ・ Right-click the unallocated disk and select
Select "Create". -"Primary" in "Create Partition Wizard"
choose. -Add "D" to the drive name. -Format with FAT32. Installing Linux -Start from the Linux CD. -Install graphically. ・ Select "Japanese" for language selection. ・ Select 106 key on the keyboard. -For mouse selection, use a PS / 2 standard 2-button mouse.
Choice. -Select a custom system for installation. ・ "Set partition manually using Fdisk"
To do. "choose. -Select "Hda" and partition the partition with the command.
Cut. (Minimum Swap and /) -Select "Format HD". -Change the ID of the area system. ・ Select "/" as the mount point with Disk Druid
Bu -Create a boot disk. Install Lilo first in the sector. -Set up the network. ・ Set the Firewall to "medium". ・ Select "Japanese" for the language. ・ Determine the root password. -Add general users. ・ Set X. -Set the monitor. -Select graphical on the login screen. ・ Installation of Linux by CD. ・ Insert the FD when the creation of the startup disk appears. ・ Linux startup confirmation by startup FD. Startup settings and confirmation ・ Enter into Linux by root, lilo. conf
put out. ・ Change the automatic start time to 15 seconds. ・ Add Windows to the startup. ・ Set to automatically start Windows. -Lilo. Save conf and exit. ・ Make sure that the lilo screen appears when you start up the PC.
Acknowledge. ・ Check that Windows starts up automatically. ・ Check that Linux can be selected and started with lilo
It FIG. 23 shows the number of educational elements at each stage. [5-6] Stacking method of education As you saw in [5-5-1], when you put it all together, you get 5 on stage.
There are 47 elements, concrete educational elements.
As you saw in “[5-3] Adopting the top-down method”,
As the number of elements to be weighed increases, the number of processes increases.
You can see that the two elements are compared here.
Only consider the case. [5-6-1] Stage Education Process First, start by deciding the order of the five stages.
If you take two out of five elements and compare⁵= 32
Process is required. [5-6-2] Educational Engineers at Each Stage
Degree Next, we will examine each stage. In the setting of PC, there are four possible educational factors. these
Similarly, comparing two by two, two^Four= 16 steps are required
Becomes The same applies to the following stages. Windows installation Number of elements: 6 2⁶= 64 Creating D drive Number of elements: 6 2⁶= 64 Linux installation Number of elements: 232²³= 8,388,608 Startup settings and confirmation Number of elements: 8 2⁸= 256 The total number of steps in each stage is 16 + 64 + 64 + 8,388,608 + 256 = 8,
389,008 About 8,400,000 steps are required. [5-6-3] Time required for education By 5-6-2, you can see the number of steps required for education so far.
I was able to grasp it. Then, if one step takes 1 second
And how long does it take to build these educational methods
Do you think it is necessary? Structure of educational method on stage
It takes 32 seconds to build, and each stage has 8,38
It takes 9,008 seconds. With these totals, the final educational work
How many days do you need to get the work process?     32 seconds +8,389,008 seconds = 8,389,040 seconds       = 1398177.3 minutes = 2330.3 hours = 97.1 days       = 3.2 months   It takes about 3 months. [5-7] Education method using top-down Use top-down to set final goals and
Consider the case of considering growth factors. Final goal
Since the process has been completed,
Correlate each element of education with "can be removed" and "wa" in the final goal
There are two ways to think about. [5-7-1] Stage Education Process Just like when you don't use top-down,
Understand the number of processes to be decided. There are 5 stages, each one
Regarding this, there are two ways to examine after associating it with the final goal.
I will do 5^Two= 25 steps are required. [5-7-2] Educational Engineers at Each Stage
Degree Consider each stage as well as the stacking method
Go. In the setting of PC, there are four possible educational factors. these
The same examination for 4^Two= 16 steps are required
Becomes The same applies to the following stages. Windows installation Number of elements: 6 6^Two= 36 Creating D drive Number of elements: 6 6^Two= 36 Linux installation Number of elements: 23 23^Two= 529 Startup settings and confirmation Number of elements: 8 8^Two= 64 The total number of steps in each stage is 16 + 36 + 36 + 529 + 64 = 681 About 700 steps are required. [5-7-3] Time required for education Up to 5-7-2, the number of educational processes using top-down
Figured out. Similar to the stacking method, one process
If it takes 1 second to
How much time do you think you need? On stage
It takes 25 seconds to build the educational method in each
Ji takes 681 seconds. With these totals, the final teaching
How many days do you need to ask for a work process?
Or? 25 seconds + 681 seconds = 706 seconds = 11.8 minutes It will be finished in about 12 minutes. [5-8] When not using top-down
Comparison when used If we build an educational process without using top-down, [5-
As we saw in section 6], we need about 3 months.
It If you use top-down, it will only take 12 minutes.
Do you get it. When not using top-down and when it is used
Figure 24 shows a comparison of the number of steps and the time with respect to the number of elements.
Meru. As you can see in Figure 24, the top-down method
If you use it, you can see how much labor and time can be saved.
It is. The greater the number of elements, the more obvious the difference
Become. In building dual boot education, the final
If the top-down method is used for
Compared to the case, it is one-hundredth of the time. [5-9] Efficient use of cell model
education Use a top-down method to create educational procedures
How efficient is the number of processes and time
I've seen the law. Now, let's save these educational element data.
Saved in bar space and cellular data by cell model
By building an educational database as a base,
What kind of advantage can be gained over education up to? In the conventional teaching methods, the teacher has made
Repeated trial and error, and tried several times the learned education method
Therefore, they were looking for the optimal education method. But Sermode
By creating a database of educational elements using
By storing the discriminant function by Topy theory,
It is possible to return to the state. In other words, so far trial and error
It can be said that the process that was being done is theoretically gone. To these
Then, the "dual boot creation" that was the previous example
Let us give a detailed explanation using an example. [5-9-1] Relation of Educational Element In “Dual boot creation” mentioned in Chapter [5-5]
There are five educational elements on each stage,
There were 47 elements and 52 as educational elements.
Become. The stages are arranged in order from top to bottom.
However, the actual elements from the stage are all mutually
It is related. For example, each part decided on the stage
The size of the station is not limited to the stage
Of course, it will be involved in the page. That is, each
FIG. 25 shows the image of the relationship of tages. Decompose these into the levels of actual educational elements
Then it has a more complicated entanglement. This is shown in Figure 26.
You To manage elements with such complex relationships
A database that uses a cell model is most suitable
Is a "cellular database". [5-9-2] Utility of using cell model Cell models, homotopy and cellular in chapter [5-4]
The outline of the database has been explained. So in this paper
Of the educational elements we have seen into a cellular database
Consider its usefulness. Cellular database
Is superior to relational databases in
It is said that 1) Data management is not limited to the main administrator. 2) It is easy to extract each data. 3) It is easy to change (additional) data. 4) Since the homotopy equivalence relationship holds,
It will be easier to re-consider and re-examine. Education is something that changes year by year.
Therefore, the installation procedure of the PC given in this example is
It is an exaggeration to say that it is changing daily with the progress of
Wouldn't. Also, even if you take only school education, the same
The number of teachers who teach subjects is not limited to one. Same by year
Even in the same subject, the teachers often change.
That is. However, the educational objectives for the same subject vary from year to year.
Should not have changed. In other words, such an intention
In terms of taste, although the final goal is set,
As the number of members changes, so does its ultimate goal.
It is also true. So I used a cellular database
That data management is limited to the primary administrator.
As there is no such method, any teacher can write a method that is more efficient.
Similar level of education for efficiency that can be replaced and extracted
Can be applied. [0189] Also, since education is originally a failure
Do not. In the cellular database, homotopy
The theory of values is valid, and it is possible to redo from the middle and
It has the reproducibility to return to. Flexible from these things
Cellular database with high reliability is a top-down method
Think of it as the best way to save your data
available. [5-9-3] Educational database model
Long If you do top-down educational elements, it is shown in Figure 26.
As such, those elements are intricately intertwined. Individual
The element is joined to several other elements and its form is
It depends on the element. With these in mind
At that time, as an example from "Dual boot setting",
Consider creating a cellular database. First, as an example,
Let's consider the tage "PC settings". The four elements of the stage are
As shown in Figure 27, if you drop it in a database cell
Correspond to. Based on the cell model, these decompositions
The image and discriminant map can be written as follows. With these mappings, each cell can be recorded by recording the cell information.
It becomes easier to handle the educational elements stored in the cell. Tsuma
Data can be rearranged, extracted, added, etc.
Become. For example, in the education from the middle,
By extracting the element and comparing it with the element of the final goal,
It becomes clear where it is located with respect to the mark. Also, consider the connection between cells.
It In the example, there are four elements, so these combined cells
(X) is as follows. Saving the information in the merged cell allows the current position of the educational element to be
Once you know it, the next step will be clear. Since
As I've seen briefly above, the educational element is the cellular database.
You can see that it is possible to put it in the space. [5-9-4] Use of homotopy theory Educational data can be dropped into a cellular database
Now that it is possible, the homotopy theory can be utilized. Teaching
As for education, what is always said to be the best is changing. new
Theories are created in minute increments, which also contributes to education.
It changes. In addition, the educational procedure changes due to changes in teachers.
I will do it. An educational model that can flexibly respond to these changes
This homotopy theory is important when creating
is there. By the establishment of the homotopy theory,
Follow the inverse function even if the number of elements increases by one
You can get back exactly to that place, and try again from there.
It can be fixed. Education that has always involved trial and error
In, the fact that homotopy can be used means that
Is also important. [5-10] Summary The example of "creating a dual boot" that I gave you this time is educational
Seen from the side, it belongs to a category with very few elements.
will do. In actual education, how many times these, or
We provide education with tens of times the educational elements. In this paper
Even in the procedure of only 52 educational elements I saw, the top
1 depending on whether down or stacking is used
The difference between 2 minutes and 3 months became clear. Simply top da
Eun system requires 1/100 the time of stacking system
If you think about it, Japan's compulsory education is for 9 years,
That education will be completed in about 33 days.
It In addition, the cellular database
All teachers can easily reach the same level of education and
Even if you make a mistake, you can easily redo the education.
Become. In fact, not only school education but also other education
On the side, this top-down method and cellular data
If you manage the educational database based on Japan,
The educational model itself is dramatically efficient both in terms of time and ability
Is expected to improve. [Sixth prerequisite technology] [6-0] Summary An existing general-purpose DBMS, a relational DBMS, is a
Assuming that there is a database administrator as a default model
On the WEB where the data dependency is easy to change
It is not suitable for information management in situations such as the data management of.
Cellular model, which is an excellent information space construction model
We have developed a cellular DBMS using. relational
The main advantages of data output compared to DBMS are (1)
Data output in object units, (2) homotopy
Reusable cells (tables) that can be saved,
Is. In addition, it greatly reduces the effort of data output.
It was shown by the case that it was possible. [6-1] Background and purpose Various computer devices are growing on the Internet.
Entering the era of constant connection to Bar, enormous amount of information is constantly processed.
It came to be understood. Among them,
Customer management at convenience stores, convenience stores
Inventory management of goods at the construction site, management of materials and equipment at construction sites, etc.
The condition that data management is based on is constantly changing.
There is a need for an information management system that suits the situation.
It was. However, at present, it is most commonly used.
Looking back at relational DBMS, which is a database
Then, the data dependency does not change and the data
A static situation where the presence of a database administrator is required
It is assumed that there is plenty of room for substitution in the future.
To be Therefore, a good model of information combination,
That data dependency changes dynamically
Database logic that can flexibly respond to the current situation
Is the purpose of this research. [6-2] New needs for database Number of companies to understand new needs for DBMS
We conducted an interview survey with the information system department of the company. That
As a result, the needs from the user's point of view revealed
Shown below. A) It is easy for the user to output data.
When The larger the database, the entire database
It becomes difficult to understand the design of the output target
It becomes difficult to specify the storage location of the data. Yo
To find the storage location of the data that the user needs.
Output even if all attributes of data are not specified
Data attributes can be specified for each object and
Data that can also specify linked objects
Ease of output is required. B) A high degree of freedom for the user to input data
thing The data by which a user recognizes a real-life event may vary depending on the situation.
Since there are various types of data, many restrictions should be set when entering data.
Not. That is, the attributes of the data to be entered and their relation
The data entered by the user is not set in advance.
Data entry to recognize and determine data and their attributes.
A high degree of freedom is required. C) It is easy to change the database design
To be Due to the integration of multiple databases due to corporate integration, changes in corporate strategy, etc.
Database design, that is, data attributes, attribute tables
The dependency relationship between the cables will change even after the DBMS is operated.
There are many. Therefore, there was a change in the database design.
In this case, only the changed parts should be repaired without reviewing the overall design.
It is necessary to have the easiness and flexibility to deal with the work required for regular work.
It [6-3] Problems of Relational DB Normalize and relay the problems of existing relational DB
Consider from the viewpoint of the optional model. [6-3-1] Normalization In relational DB design, information is stored on the database
In general, normalization is performed to manage efficiently. Normalization
Then, the precise definition of the relationship between attributes, the minimum of data duplication
In order to perform consistent, addition / deletion / update operations,
Reduces data redundancy and collects attributes with strong relationships
Therefore, the dependency between attributes is minimized. However, relay
For normalization in the regional DB, the new needs mentioned above
On the other hand, there are the following problems. A) Difficult to deal with changes in database design
I In normalization, determination and management of the entire database design
It is assumed that there is a database administrator who
Changes in some data dependencies affect the overall design
There are many possibilities. Also, the database design change
The database administrator must handle all the adjustment work.
No For example, when multiple companies integrate
It will also be integrated, but the coordination work of the DBMS integration will be done by each company.
Business database manager discussing details
This is a huge amount of work
Can be said. In this way, the response to changes in the database design
A database administrator who understands the overall design
Must be dealt with and can often be a daunting task
Absent. B) A method of specifying a table and performing JOIN join
A long time As the degree of normalization increases, the number of tables inevitably increases.
It Therefore, users try to utilize the database.
Table and data to be processed when requested
Of work to find JOIN attributes from the whole and join them
The number increases. As a result, the burden on the user is increased.
It [6-3-2] Relational model The relational model is E. F. 1 by Codd
A model published in 970, with set theory as a mathematical basis
deep. There are relational operations and set operations in operations.
In arithmetic, data are considered as two-dimensional tables and
Pseudo-join table relationships with what is called a key
Data management. This relational D
The relational model, which is B's data model,
Considered the following problems for the new needs mentioned above
available. A) A great deal of effort is required for the user to output data.
I In set theory, which is the mathematical basis of the relational model,
When multiple sets overlap, the separation of elements for each set is expressed.
Neither is it, nor is the map of that change expressed. Therefore,
Rational DB has multiple tables for data output.
After joining the joins, the attributes of each table before joining
Sex information is not recorded, and information about changes before and after binding is also recorded
Not done. Because of this, users need from the database
When outputting various data, the data from the combined table
All attributes must be specified, and refer to the data before combining.
When you illuminate, you have to specify the original table again.
Absent. These efforts are not required when the database size is large.
Inevitably, the burden on the user for data output will inevitably increase.
It B) There is a great deal of trouble when the user inputs data.
I In the relational model, the map of decomposition of overlapping sets
Is not expressed. So with relational DB,
Set the combined table from
It cannot be disassembled. Therefore, the user
Input the information you have recognized to the database as it is
Cannot be done by the database administrator in advance
Must be entered according to the specified design. this is,
It can be said that the degree of freedom in data entry is low. [6-4] Cellular model The cellular model was created by one of the inventors (Kunii).
With the model announced in 1999, the actual local information and
Global information can be combined in the real world / cyber
An extremely excellent information space structure that can project all phenomena in the world
It is a built model. Absoluteness of time and space in physics
The relationship between spatiotemporal space and classical dynamics
It is just an approximation of the theory of relativity
The relational model for the static world is dynamic
Thought to be a cross section of a cellular model that can represent the world
You can [6-5] Definition of Cellular DBMS Cellular DBMS based on cellular model
Are three types of information: cell information, cell definition information, and cell operation information.
Composed of information. [6-5-1] Cell information In the cell information, record the cell ID and cell attributes / data.
It A cell is equivalent to a relational DB table
It In addition, cell attributes and data are related to variables and values.
It is represented as a two-dimensional table. Also, the cell ID is
Table ID, which is the information for uniquely identifying each cell.
Information, and the actual squat that guarantees the uniqueness of the cell.
It is decided at the time. Also, the DBMS can
A management cell is provided to manage the cell ID.
Is defined. [6-5-2] Cell definition information The cell definition information includes open cell and boundary information for each cell.
To record. In the cell model, cells with boundaries are closed.
A closed cell and a cell without boundaries are called open cells. sand
Wachi, closed cell BⁿOpen cell eⁿ, The border
∂BⁿAnd ∂ Bⁿ= Bⁿ-Eⁿ Open Cell in Cellular DBMS
Is at the same time an attribute contained in a closed cell
Also, the open cell itself is treated as a cell with one attribute.
Be seen. Also, the primary key of the relational database
It plays a role equivalent to
One attribute that is arbitrarily identified is selected. In addition, the open
The closed cell is the actual reality that the closed cell is going to project.
The body itself, the boundary representing the environment attached to the entity
It is meant to be information. As described above, the open cell / boundary information is
Recording makes cell data management much easier.
For example, open cells in each closed cell
Data in ascending / descending sort order and data output
Thus, it becomes possible to efficiently search the data. Well
Also, even if you do not specify all attributes of the data when outputting the data,
If you specify the open cell / boundary of the target cell,
You can output the data you want, and the object unit
You can manage data with. About this [6
-6-4-1]. [6-5-3] Cell operation information The cell operation information includes operations related to the homotopy of cell deformation.
Record the work information. By recording this operation information,
Since the shaped cell can be returned to the state before deformation,
It is possible to utilize the cell in the form as a resource. That is,
The cell can be reused. About this [6-6-
4-2]. [6-6] Cellular DBMS in Case
Comparison of relational DBMS and relational DBMS Cellular and relational DBMS practical
Set up one situation to compare convenience
Then, the search man-hours for the data output request are compared. [6-6-1] Situation setting Multiple users from different positions can access the database
Repair / repair, etc., as an example of various data output requests
One situation related to construction small lot construction of
And consider the information management. Situation: “Customer wants to build a small-scale construction project for a construction company
The order is placed and the manager arranges the workers and equipment.
Then, the worker makes full use of mobile devices on site and
Performing construction while exchanging information with contractors and managers
U In addition, a salesperson can provide services that meet individual customer needs.
Establish sales measures to provide screws. " [6-6-2] Use of cellular DBMS This section describes the basic cell design and security of a cellular DBMS.
The operation of disassembling and joining is shown below. <Step 1 Closed cell open
Cell design> Customer cell: Bc ^ 6 Closed cell: (code, name, age, sex, address, ma
Manager) Open cell: (code) Manager cell: Bm ^ 7 Closed cell: (code, name, age, sex, college student)
(Academic, affiliation, extension number) Open cell: (code) Worker cell: Bw ^ 9 Closed cell: (code, name, age, sex, occupation, product
(Business site, telephone, qualification, qualification name) Open cell: (code) Material cell: Bb ^ 5 Closed cell: (number, type, name, weight, manufacturer) Open cell: (number) Project cell: Bp ^ 9 Closed cell: (name, manager, customer code,
Worker code, materials and equipment, date, time, work content, work
Work site) Open cell: (name) <Step 2 Cell decomposition by equivalence relation
Joining> Customer cell, Manager cell, Worker cell, Equipment
The 4 cells of the material cell are based on the equivalence relation with the project cell.
The cells are disassembled and joined. These transformation maps are saved
Therefore, the cell can be reused. Cell decomposition map c1: Bc ^ 6 → Bc ^ 1 (code) ∪g1 Bc ^ 5 Discrimination map g1: ∂… ∂Bc ^ 6 → Bc ^ 1 (∂ repeated 5 times) Cell decomposition map p1: Bp ^ 9 → Bp ^ 1 (customer code) ∪ h1 Bp
^ 8 Identification map h1: ∂ ... ∂Bp ^ 9 → Bp ^ 1 (∂ repeated 8 times) Attaching map k1: Bc ^ 1 (code) → Bp ^ 1 (customer code) Cell decomposition map m1: Bm ^ 7 → Bm ^ 1 (name) ∪g2 Bm ^ 6 Identification map g2: ∂ ... ∂Bm ^ 7 → Bm ^ 1 (∂ repeated 6 times) Cell decomposition map p2: Bp ^ 9 → Bp ^ 1 (manager) ∪ h2 Bp
^ 8 Identification map h2: ∂ ... ∂Bp ^ 9 → Bp ^ 1 (∂ repeated 8 times) Attaching map k2: Bm ^ 1 (name) → Bp ^ 1 (manager) Cell decomposition map w1: Bw ^ 9 → Bw ^ 1 (code) ∪g3 Bw ^ 8 Identification map g3: ∂… ∂Bw ^ 9 → Bw ^ 1 (∂ repeated 8 times) Cell decomposition map p3: Bp ^ 9 → Bp ^ 1 (worker code) ∪ h3 B
p ^ 8 Identification map h3: ∂ ... ∂Bp ^ 9 → Bp ^ 1 (∂ repeated 8 times) Attaching map k3: Bw ^ 1 (code) → Bp ^ 1 (worker code) Cell decomposition map b1: Bb ^ 5 → Bb ^ 1 (number) ∪g4 Bb ^ 4 Identification map g4: ∂ ... ∂Bb ^ 5 → Bb ^ 1 (∂ repeated 4 times) Cell decomposition map p4: Bp ^ 9 → Bp ^ 1 (materials and equipment of use) ∪h
4 Bp ^ 8 Identification map h4: ∂… ∂Bp ^ 9 → Bp ^ 1 (∂ repeated 4 times) Attaching map k4: Bb ^ 1 (number) → Bp ^ 1 (materials and equipment of us
e) <Step 3 of integrated cell by cell joining
Creation> Integrated cell (X) can be expressed by the following formula. X = (∪i = 1-4 Bi) ∪Bp / ~ B1 = Bc, B2 = Bw, B3 = Bm, B4 = Bb Where ∪ represents the exclusive sum of cells and ~ represents the equivalence relation.
You [6-6-3] Use of relational DB In this section, relational DB master tables
Design and normalize. <Step 1 Table Design> Customer master table (C): (code, name, age,
Sex, address, manager) Manager master table (M): (code, name
(Previous, age, sex, university of origin, affiliation, extension number) Worker master table (W): (code, name, year
Age, sex, occupation, work site, telephone, qualification, qualification name) Equipment master table (B): (Number, type, name, weight
Quantity, manufacturer) Project transaction table (P): (name
Previous, manager, customer code, worker code, resources
(Equipment, date, time, work content, work site) <Step 2 Table Normalization> C → C1, C2 C1: (code, name, age, sex, address) C2: (Address, Manager) M → M1, M2 M1: (code, name, age, sex, university of origin, affiliation) M2: (affiliation, extension number) W → W1, W2, W3, W4 W1: (code, name, age, sex, work site) W2: (Work site, telephone) W3: (Qualification, Qualification name) W4: (code, qualification) B → B1, B2, B3 B1: (Number, type) B2: (type, name, weight) B3: (Name, manufacturer) [6-6-4] Relational DBMS
Comparison of search man-hours of cellular DBMS In the above example
For the types of data output in 1. Data related to construction work
Search for data, 2. Examine the original resource data for construction work
The search man-hours when outputting the data are divided into two
To do. [6-6-4-1] Day related to construction work
Search for data <Cellular DBMS> User once establishes equivalence relation
If you create an integrated cell with
Therefore, it is not necessary to create an integrated cell for each request. Also,
In the integrated cell, the cell to be output from the cell definition information
Requested by specifying open cell / boundary information
Data can be obtained. In the example, the customer cell
Equivalence relationship between manager cell, worker cell, and equipment cell
To join the project cell to create an integrated cell.
And save the map. Then, the cell definition operation is performed for each request.
Information about any open cell or boundary of 4 cells from the product
Specify and search for data in object units. Two
After the first time, the integrated cell is restored by the saved map.
It does not require the labor of cell joining. <Relational DBMS> The user
Find the necessary table for each request and join it with JOIN.
Specify the required data attributes to obtain the required data.
You can In the example, each request is normalized for each request.
Customer master table by JOIN connection on request
Table, manager master table, worker master
-Create a table and equipment / material master table. That
Then, join the project table and JOIN to request data.
Data is specified by specifying the data attribute. <Comparison between general formula of search man-hours and man-hours> Relay
Represents the man-hours for searching for the optional DBMS and the cellular DBMS
To create a general formula and compare the man-hours, set each variable as follows.
To define n: Number of data output requests m: number of requested data attributes C (n, m): Search time in cellular DBMS R (n, m): Search engine in relational DBMS
number c: Find and combine two target cells (tables)
The time it takes f: The time and effort to specify one attribute According to the above consideration, two DBMSs are detected.
The general formula for man-hours is as follows. C (n, m) = 4e + n^*f ・ 4e is a process for joining Bc, Bm, Bw, and Bb to Bp.
number ・ N^*f specifies n closed cells to be requested
Man-hour R (n, m) = n^*3e + Σ_i ⁿ(Mi^*f) ・ 3e creates each table by JOIN join as required.
Efforts to complete, 3 is the average number of joins ・ Σ_i ⁿ(Mi^*f) is the number of different data attributes for each request
Man-hour to specify n mi Therefore, if these differences are taken, Therefore, the number of cases (n), the number of attributes of the requested data (m)
The more there are, the more clearly R (n, m)> C (n, m).
Cellular DBM over existing relational DBMS
It can be seen that S requires less search man-hours. <Comparison of search man-hours in actual example> Here,
In the set situation, data related to construction work
Data output required for frequent data searches
The following is a practical example of the request, and two DBMSs are added as described above.
The search procedure and search man-hours in this section are described. "A certain worker does a certain construction (name =" A
Manager A (code =
Check the affiliation and extension of "a") and contact them. " Cellular DBMS: ・ After joining Bp and Bm, Bm as a closed cell
Is specified. ・ From n = 1 and m = 1, C (1,1) = e + f Relational DBMS: ・ After joining tables P, M1 and M2 by JOIN, 2 groups
Specify the gender (affiliation, extension). ・ From n = 1, m = 2 and JOIN coupling twice, R (1,
2) = 2e + 2f [A certain manager has a certain construction (name
= "CD") worker (code = "b") age and resources
Check case and name of qualification. And later on that
Worker B (code = “b”) of construction (name = “CD”)
Check their name and phone to get in touch. " Cellular DBMS: ・ After joining Bp and Bw cells, as a closed cell
Specify Bw. The attaching map at this time is saved.
Therefore, in the second request, there is no need to join cells.
Only specify Bw as a closed cell. ・ From n = 2, m1 = m2 = 1, C (1,1) = e + 2
f Relational DBMS: ・ Join tables P, W1, W2, W3 and W4
After that, specify 3 attributes (age, qualification, name of qualification)
It At the second request, the tables P, W1, W2 and W are again displayed.
Join W3 and W4 together and specify 2 attributes (name, phone)
Set. ・ N = 2, m1 = 3, m2 = 2, JOIN combination 8 times
, R (2, (3,2)) = 8e + 5f [A certain manager has a certain construction (name
= "EF") Equipment B (No. = "3")
Check the car. In addition, another manager
Equipment C (No. = "4") for a certain construction (name = "GH")
Check the type, weight, and manufacturer. " Cellular DBMS: ・ After joining cells Bp and Bb,
And specify Bb. The attaching map at this time is preserved.
The second request requires cell bonding.
No, just specify Bb as a closed cell. ・ From n = 2, m1 = m2 = 1, C (1,1) = e + 2
f Relational DBMS: -Joining tables P, B1, B2 and B3
After that, specify 3 attributes (type, weight, manufacturer). Second time
Request, the tables P, B1, B2, and B3 are again JOI
After N-joining, the same 3 attributes are designated. ・ N = 2, m1 = m2 = 3, JOIN combination from 6 times, R (2, (3,3)) = 6e + 6f [A certain sales person sets up a sales strategy.
Customer D (customer) who requested a certain construction (name = “GH”)
Check the gender, age, and address of the customer code = “d”). Also,
Another sales person requests a certain construction (name = “IJ”)
Check the gender, age, and address of customer E (customer code = “e”)
Bell. " Cellular DBMS: ・ After joining cells Bp and Bc,
And specify Bc. The attaching map at this time is preserved.
The second request requires cell bonding.
No, just specify Bc as a closed cell. ・ From n = 2, m1 = m2 = 1, C (1,1) = e + 2
f Relational DBMS: ・ After joining tables P, C1, C2 by JOIN, 3 groups
Specify gender (sex, age, address). At the second request,
After joining the degree tables P, C1 and C2 by JOIN, the same
Specify 3 attributes. ・ From n = 2, m1 = m2 = 3, JOIN combination total 4 times, R (2, (3,3)) = 4e + 6f Putting these together, Therefore, it is clear that the existing relational DBMS does not
Ruler DBMS may require less search man-hours
Recognize. [6-6-4-2] Original regarding construction work
Retrieval of resource data <Cellular DBMS> User once establishes equivalence relation
If you join the cells for the first time, the homotopy will be preserved
Then, from the cell that is focused on it based on
Resource data cells can be referenced. In the example,
Customer cell, manager cell, worker cell, equipment cell
The cell is joined to the project cell by the equivalence relation,
If you save the homotopy, you can request it from the project cell
Depending on the homotopy, the cell before joining (in this case, the customer
Cell, manager cell, worker cell, equipment cell, etc.
Search for resource data by specifying (or). <Relational DBMS> Construction work
Find the relevant table for the resource data request
Create a resource data table by joining and joining JOIN.
Search the requested resource data by specifying the table of
You can In the example, the project table
Users who are paying attention to the transaction table)
Resource data table (customer, manager
Related to each master table of workers, workers, and equipment)
Find a table that is JOIN and join the table
Resource data is searched by specifying a resource. <Comparison of search man-hours> Relational DBM
Create a general formula that expresses the search man-hours of S and the cellular DBMS.
In order to compare the man-hours, each variable is defined as follows. n: Number of data output requests C (n): Search time in cellular DBMS R (n): Man-hour for searching in relational DBMS g: Search for one target resource data cell (table)
Labor h: Hassle when specifying one cell (table) According to the above consideration, two DBMSs are detected.
The general formula for man-hours is as follows. C (n) = 4g + n^*h ・ 4g is man-hour to search for Bc, Bm, Bb, Bp ・ N^*h specifies n closed cells to be requested
Man-hour R (n) = n^*3g + n^*h = n^*(3g + h) 3g has a table related to the resource table for each request
Man-hours to look for (3 = average number of tables) ・ N^*h specifies n JOIN-joined tables
If you take these differences according to the number of man-hours, R (n) -C (n) = n^*3g-4g = (3n-4)^*
g Therefore, as the number of cases (n) increases, it becomes clear that R (n)> C (n) And cellular D from existing relational DBMS
It can be seen that BMS requires less search man-hours. <Comparison of search man-hours in actual example> Here,
Materials related to construction in the set situation
Data that can occur frequently as a source data search
Below is an example of output request and two DBs
The search procedure and search man-hours in MS are described. [The manager has a certain construction (name =
"AB") worker A
Data. After that, there is another construction (name = “C
D ”) to search for worker B instead of worker s data
refer. " Cellular DBMS: ・ Find Bw, join cells with Bp in equivalence relation, and then homo
Bm is designated by the topy. In the second request,
Bw is designated by the existing homotopy. ・ From n = 2, C (2) = g + 2h Relational DBMS: ・ Look for M1 and M2 and join them by joining them to the master master
Create a Bull and specify its table. Second request
But do the same. ・ From n = 2 and 2 tables, R (2) = 8g + 2h [A manager has a certain construction (name =
Machine made by the same manufacturer as the "CD") equipment (number = "2")
Examine the material. Machine for construction (name = “EF”) with workers
The same type as the material (number = "3") but with a smaller weight
Find out. " Cellular DBMS: ・ Find Bb, join cells with Bp in an equivalence relationship, and
Bb is specified by Topy, and the same manufacturer as "Code = d"
Search is performed under the condition of "equipment". Second request
Then, specify Bb by the saved homotopy,
No. = 2 and the type is the same, but the weight is smaller ”
And search. ・ From n = 2, C (2) = g + 2h Relational DBMS: ・ Find B1, B2, B3, join JOIN and
Create a table, specify the table,
= "Equipment of the same manufacturer as 2"
U The same applies to the second request, and the type is the same as "Number = 2".
Searching is performed under the condition that "the weight is smaller than". ・ From n = 2 and the number of tables 3, R (2) = 6g + 2h [A certain sales person has a certain construction (name =
Customers who live near “GH” customers (code = “e”)
List the data. There is another salesman
Customer (code = “f”) and year of construction (name = “IJ”)
List customer data of near age. " Cellular DBMS: ・ Find Bc, join cells with Bp in an equivalence relationship, and then
Specify Bc by Topy, and specify "Code = e and neighbor
The search is performed under the condition of "customer." In the second request,
Bc is specified by the saved homotopy, and "Code =
The search is performed under the condition of "customer whose age is close to f". ・ From n = 2, C (2) = g + 2h Relational DBMS: ・ Look for C1 and C2, join JOIN,
Create a Bull, specify the table, and specify "Code = e
Perform a search under the condition of "neighboring customers."
In the same way as for the job, say "Customer whose age is close to code = f"
Search under the specified conditions. ・ From n = 2 and the number of tables is 2, R (2) = 4g + 2h Putting these together, Therefore, it is clear that the existing relational DBMS does not
Ruler DBMS may require less search man-hours
Recognize. [6-7] Conclusion Relational D, the most general-purpose database in existence
BMS presupposes the existence of a database administrator.
Situations where data dependencies frequently change, such as on the WEB
It is not suitable for information management. Therefore, an excellent information space structure
A cellular model, which is a built model, is used as a mathematical model.
Proposed a ruler DBMS. In a cellular DBMS,
In terms of data output, the advantages are: 1) By defining the concept of boundaries in cells,
Data output is possible 2) Disassemble and join according to the equivalence relation between cells, and copy the
Cell reuse is possible by saving the image And the situation of data management in construction small-lot construction
Options and compare the search man-hours when outputting data.
And, cellular DB is more data than relational DB
It was revealed that the output effort was very small. [6-8] Future tasks From the user's perspective, the following points in the cellular DBMS
Consider it while comparing with relational DBMS. 1) Hassle when inputting data 2) Database design effort 3) Hassle when changing the design [7-0] Summary While the course of study changes frequently, ironically
The number of voices pointing out the decline of students' academic ability is increasing year by year. There
For each person, work place, final educational background, university / high school / junior
Educational day that covers the names and grades of courses taken at school and elementary schools
Assume a database. By utilizing this database
And verify how courses will affect human resource development
Education reform that presents what can be done and is an urgent national imperative
The purpose of this research is to assist human resource development. [7-1] Objective of research Since the Meiji era, Japan has been a human resource that has supported the nation with pillars of public education
Have been cultivated and have attached importance to the world. However, long
The educational level that has been said to be the top of the world remains
The educational guidelines without a vision, behind the established educational system and reform
With the revision of the
I have to say that the Western countries are compelled to
Yes. Under the Reagan administration, the United States of America
National Committee on Educational Excellence "
Report by "A Nation AT
RISK) "while seeing the coming information society
, "Reflected biased elite education, mathematics and English (national
Review the importance of basic academic ability called "words"
Aim to raise your academic ability. " Declined in the late 1990s
Returning Japan as a driving force of the world economy
bloomed. [0251] In contrast to this, Japan
The main theme of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) with the banner of "leisure learning"
By the guidance, the course of study was changed repeatedly. However
However, how will this change result in
There is no verification as to what has affected students
It was Therefore, the changes have led to improved quality of education.
If there is no, or if it is directly linked to the decline in academic ability, leave it alone
It is the actual situation that has been done. The academics of students by academics
It has been urgently necessary to verify the quality of education, as weakness is pointed out.
It [0252] Therefore, what kind of learning history a particular individual has
Database to find out if you got the profession
Build a space. For example, I became an engineer at Hitachi
What courses do individuals take in elective courses in high school and junior high school?
Learn what you learned in elementary school, what kind of understanding
Database). This allows
Courses set by elementary, junior high and high schools or colleges for students
So that you can see what impact it has on
It What is the impact of deleting courses and reducing hours?
A guideline for learning so that it can be inferred whether the sound
It will be used as a reference when making further changes. like this
The aim is to propose a database and promote the quality of education.
Is. This database contains one of the inventors
Cell, a model announced in 1999 by
A model (Cellular Model) is applied.
The homotopy (h
omotopy), equalization function (identificat)
dynamically associated with (ion function)
Modelers that accurately characterize common parts of things to be attached
To present. In addition, currently available relay shows
Compared to the null database, the data using the cell model
The database shows that man-hours at the time of search can be reduced. [7-2] Assumed "educational database"
Features of The “education database” includes individual names, workplaces, occupations,
Qualifications, TOEIC scores such as TOEIC, Center test
Exam scores, final educational background, take courses at university / high school / junior high school / elementary school
The name of the course you have taken and the study guidance that is the basis for the content of the course you took.
The year of revision of the territory and the results of each subject are accumulated. Students' teaching
When students enter data such as grades, individual students
I have no choice but to disclose the name of the person
Considering this, a third party using the database can search for information.
If you do, the individual's name will be hidden. Database users are via the Internet
Access the database at, "(
Search conditions such as "Course guidelines) High school mathematics grade 5"
At the same time, enter as a search result such as "job type" "final educational background"
By entering the information you want to handle, the corresponding person (anonymous)
Get the information you need. In addition, "doctor" "lawyer" such as jobs
Obtain the grades of the target person using the seed as the search condition
You can also [0256] Furthermore, for example, the learning guidance of 1982
Those who excel in a specific subject based on the territory are not
The course of those with excellent results in the same subject based on the learning guidelines
And the difference in scores such as English test
To be able to How is primary and secondary education the path of a person?
And how they influence human resource development
Propose a mechanism that can verify [7-3] Cell model and relational model
Comparison with Dell Data in the currently popular relational model
The relevance of is represented by a table. Table, table
The relationships within the rules are normalized based on the data
Is determined. On the other hand, the cell model uses homotopy theory.
Table in the relational model
In terms of dynamically configuring the "cell space" equivalent to
It is flexible. Moreover, in the relational model,
Required from table due to shadow
After the attributes are retrieved, the retrieved attributes and the original table
The relationship with Le cannot be maintained. On the other hand,
At Dell, decomposition when projecting (cell decomposition) required attributes
Save the decomposition map f
Then, by applying the homotopy theory, the cut-out cell
Maintaining a relationship between the space and the original cell space (table)
Will be possible. Here, a relational model will be given with a specific example.
Data manipulation by Dell and data manipulation by cell model
I'll see the difference. Master table "person information"
From, "name" and "junior high school grade" corresponding to occupation "doctor"
Assume a data operation that seeks. Relational model
The procedure in is selection → projection.
Table "person information" attribute "professional" to "doctor"
Select a tuple (FIG. 29). After that, tap the "doctor"
The necessary information is projected from the file and the result (FIG. 30) is obtained. On the other hand, in the case of the cell model, the projection
(Cell disassembly) → Obtain information by the procedure of selection. That is,
Cell decomposition (cell decomposition)
To refer to only tuples for job attributes and
To obtain (FIG. 31). Furthermore, the decomposition map f for cell decomposition is
By saving and applying the homotopy theory,
Homotopy equivalence function between the cell space of the
Keep the clerk. By this work, I corresponded to the occupation
You can get your name and junior high school record (Fig. 3
2). This operation is an empty cell that contains the attributes required by the projection.
Original cell space and cut cell even if only the space is divided
To relate the relationship with space by homotopy theory
Therefore, it becomes possible. In addition, the map of the homotopy theory
Save is other than the function that guarantees the relationship between cell spaces.
Also, in the sense that the cell space can be reused,
It also helps improve efficiency. For example, for the above cell space, "defense
"Person" corresponding to "name" and "junior high school grade"
And The operation at this time is to project "job type" as an attribute
Select “Name” and “Junior high school grade” of “Lawyer”.
In other words, check all attributes and find out the attribute of "job type".
There is no need to cast a shadow, and by tracing the decomposition map f
You can derive the relationship between them. On the other hand, the relational model
In the case of
And every time there is a request, the data manipulation must be repeated.
Yes. By applying the cell model,
By tracing, the cell space used once can be reused,
Useful for efficient request processing and effective use of information. [7-4] Cell to "education database"
Application example of model Here, the cell model is applied to the above-mentioned “education database”.
A typical example is shown below. Users can access this database
The case of acquiring personal information based on certain conditions.
Suppose. Specifically, the user is "(
Course of study) High school 3rd year English performance 5 ”(Same as 1st year middle school)
Persons who meet the criteria based on the search condition of "English grade 5"
The TOEIC score of is searched. At this time,
Names and places of work of many people, TOEIC scores, final studies
History and learning history
"The cyber world of the whole" is a p-dimensional cell space B_DB ^P,
The cell space of the learning history that accumulated the grades of elementary, junior high, and high schools
q-dimensional cyber world B_{Learning history} ^q, Specified by the user
The cyber world of the user including the search condition is the r-dimensional cell space
Interval B_user ^rIt is defined as The dimensions mentioned here are
Indicates the number of attributes contained in each cell space. "Education database
The learning history cell space and the learning history cell space are both
It is thought that there are 100 items, but in the typical example, for convenience, dimension
The number is 10 items in the "education database" cell space (p = 1
0), 4 items (q = 4) in the learning history cell space. [Education database] Cell space: B_DB ^P
(P = 10) = B_DB ¹⁰= {Name, age, address, work
Previously, TOEIC score, educational background,
Course of study) High school third grade English grade, (1989 revised learning
Guidance) High school third grade English grade, (Showa 52 revision study)
Guidance) Grades in English for the 1st grade of junior high school
Guideline) Grades in English in the 1st grade of junior high school} Learning history cell space: B_{Learning history} ^q(Q =
4) = B_{Learning history} ^Four= {(Showa 52 revised learning guidance required
Ryo) High school third grade English grade, (Heisei 1 1989 revised learning guidance)
Ryo) High school 3rd grade in English, (Showa 52 revised learning guidance)
Ryo) 1st year junior high school English score, (Heisei 1 year revised learning guidance)
Territory) 1st year junior high school English score} User cell space: B_user ^r(R = 3) =
B_user ^Three= {User name, search condition, search condition} At this time, the learning history cell space B_{Learning history} ^Four"Educational data
Database ”cell space B_DB ¹⁰And has the following relationship. B_{Learning history} ^Four⊂B_DB ¹⁰ Based on this relationship, "education database"
Cell space consists of learning history cell space and other attributes
The cell is decomposed into the cell space to be used. m: ∂. ． ． ∂B_DB ¹⁰→ B_{Learning history} ^Four(∂ is repeated 6 times
Return) B_DB ⁶= {Name, age, address, work place, TOEIC
Score, final educational background} B_{Learning history} ^Four= {(Showa 52 revised course of study) High school
Grade 3 in English, (Heisei 1 year revised course of study) High school 3
Grades in English Year, (Showa 52 revised course of study) Middle school 1
Year English grade, (1st year revised learning guidelines) Middle school 1st year
English grades} As described above, the learning history cell space B_{Learning history}
^FourIs the "educational database" cell space B_DB ¹⁰Included in
It is defined as a cell space consisting of learning history information.
Here, f used in the cell disassembly_mIs the decomposition map and m is
It is an equalization function. Save the mapping for this transformation
This allows homotopy between the cell spaces before and after deformation.
Equivalence can be maintained. The mapping f used to transform this cell space_mAnd m
Since is a continuous transformation, there is an inverse mapping. Where h = f_m ^-1And (1_X1_Y: Identity maps) 1_X: B_DB ¹⁰→ B_DB ^{10 From this, the deformation of the above cell space is the homotopy equivalence
It can be seen that the sex is maintained. Next, the learning history cell space is obtained by cell decomposition.
And the user cell space, the common part is extracted and cell integration (
An integrated cell is constructed by the cell attachment).
I will decide. First, user cell space B} user ^ThreeWhen
Learning history cell space B_{Learning history} ^FourFor both
ivalent cells B_e ^t(E = user, academic
Learning history) and the cell space containing other attributes.
Understand. At this time, the user asks "(
High school 3rd year English grade 5 ”
Search course based on "5th grade in junior high school 1st grade English"
I will do this. Therefore, the equivalent
  cells B_e ^tIs B_e ^TwoBecomes B_e ^Two= {(Showa 52 revised course of study) High school 3rd grade English
Grade 5 (Revised Course of Study 1972) 1st year of middle school
English grade 5} B_user ¹= {User name} B =_{Learning history} ^Two{(Heisei 1 year revised learning guidelines) High school 3
Year English grade, (1st year revised learning guidelines) Middle school 1st year
English grades} F used here_user, F_{Learning history}Is a decomposition map.
Further, g, k, are equalization functions, defined as
To be done. g: ∂. ． ． ∂B_user ^Three→ B_user ^Two(∂ is once) k: ∂. ． ． ∂B_{Learning history} ^Four→ B_{Learning history} ^Two(∂ is twice
repeat) Next, an integrated cell is constructed by cell integration.
It User cell space B_user ^ThreeAnd learning history cell space B
_{Learning history} ^FourTo B_e ^TwoThe common part (quantities
  by attaching map as
The integrated cell is constructed. p: ∂. ． ． ∂B_user ^Three→ B_{Learning history} ^Two(∂ is once) And an integrated cell is constructed. Furthermore, this integrated cell and the "education database" cell are empty.
The remaining cell space B excluding the learning history cell space from the space_DB
⁶The relation with Becomes Furthermore, the "educational database" cell space
Learning history cell space and cell space consisting of other attributes
Decomposition map f used for cell decomposition_mUsing the inverse mapping of
hand, If you use the relationship And the final integrated cell space B is As described above, the user is "
) High school 3rd grade English grade 5 ”
The TOEIC score of the target person and the final study
Information such as history can be acquired. Furthermore, by the same procedure, "(
Year revised course of study) High school 3rd grade English score 5 "(Same)
TOEIC score of "5th grade in English in middle school"
If you get the points,
Compared with the case, English due to differences in the course of study
The impact on education can be inferred. The sources discussed here
In the type example, "educational database" cell space and learning
The relation with the history cell space is based on the relational model.
It has the same role as the association between tables. Just this
In the typical example taken up here, the user can freely set search conditions.
At the same time, the learning history cell can be used to condition the user.
Therefore, it is dynamically constructed. That is, the relational model
Table must be fixedly associated with each other
On the other hand, the association between cell spaces is dynamic.
There is an advantage in terms. [7-5] Cell model and relational model
Dell Search Effort Comparison Application of cell model to database is relational
Reduction of man-hours required for information retrieval compared to the model
To help. The more information in the database, the more
However, if the number of data output requests increases, it will increase.
However, the cell model is expected to increase its superiority. here
Using the above typical example, relay man-hours at the time of data output.
Compare with the case of the local model. First, the cell model and the relational model
A general formula representing the search man-hour of is presented. n: Number of data output requests m: number of requested data attributes C (n, m): Cell model search man-hours R (n, m): Relational model search man-hours e: Effort to combine cell spaces (tables) f: time and effort to specify one attribute Then, C (n, m) = 2e + nf * 2e is B_user ^ThreeAnd B_{Learning history} ^Four, B_user ^ThreeIntegrated
Man-hours to integrate into cell space B * Nf is the number of man-hours for specifying n requested cell spaces R (n, m) = n × xe + Σ_i ⁿ(M_iXf) * Xe creates each table by JOIN join upon request
The number of man-hours to be performed and x vary depending on the design of each master table.
X ≧ 2 * Σ_i ⁿ(M_iXf) is the number of data attributes that differ for each request
m_iMan-hours to specify n Taking these differences, Becomes That is, the number of data output requests (n), the request
The larger the number of attributes (m) of the data that is created, R (n, m)> C (n, m) And the cell model is better than the relational model
It can be said that the man-hours at the time of data output are small. [7-6] Conclusion High school, junior high school, and academic records of elementary school days are used as search conditions.
Acquire personal information such as work place and final educational background of the person
The “educational database” that is
What are the changes based on the course of study for human resource development?
Presented to be useful in finding out if they are affecting
did. In addition, the cell model is the most popular relay model today.
The relationship between the tables is
It is superior in that it can be dynamically associated by topy or an equivalence function.
It was shown that there is a position. Furthermore, the equivalence relation between cells
The number of search steps can be reduced by disassembling and joining cells.
I showed that. The "educational database" is a database of people
Because the large amount of information is accumulated, the internal information changes constantly,
It can be said that it is just one huge cyber world. cell
Is the application of the model to the "educational database" such a point?
Are also considered useful. [0276] According to the present invention, highly convenient information management
Technology can be provided.

[Brief description of drawings]

FIG. 1 is a diagram showing an image of situation 1 in a first prerequisite technique.

FIG. 2 is a diagram showing an image of situation 2 in the first base technology.

FIG. 3 is a diagram for explaining a procedure of data operation by a relational model in the second base technology.

FIG. 4 is a diagram for explaining a procedure of data operation by a cellular model in the second base technology.

FIG. 5 is a diagram showing how the cell space of a store is disassembled into cells.

FIG. 6 is a diagram showing how a customer cell space and a product information cell space are decomposed into cells.

FIG. 7 is a diagram showing how a cell space is formed by cell joining.

FIG. 8 is a diagram showing how an integrated cell space is formed.

FIG. 9 is a diagram for explaining that the deformation of the cell space is homotopy equivalent.

FIG. 10 is a diagram for explaining that the halfway stage of transformation of the cell space is homotopy equivalent.

FIG. 11 is a diagram showing a procedure for operating a cell space.

FIG. 12 is a diagram showing a tree structure of a scope used for explaining a third prerequisite technique.

FIG. 13 is a diagram showing a scope rule in the JAVA language.

FIG. 14 is a diagram showing a scope of an e-mail address.

FIG. 15 is a diagram showing a state in which a scope is added to a cell for expression.

FIG. 16 is a diagram showing internal data of a cell ID table.

FIG. 17 is a diagram showing a relationship between sites on the web used for explaining a fourth prerequisite technique.

FIG. 18 is a diagram showing a state in which a technical field tag is added to web information.

FIG. 19 is a diagram showing a state in which a keyword tag is added to web information.

FIG. 20 is a diagram showing a state in which inquiry results are embedded.

[Fig. 21] Fig. 21 is a diagram for explaining the outline of a stacking-type education method related to the fifth prerequisite technology.

FIG. 22 is a diagram for explaining the outline of a top-down method of education according to the fifth prerequisite technology.

FIG. 23 is a diagram showing the number of educational elements in each stage.

FIG. 24 is a diagram showing a comparison of the number of processes and time in the stacking method and the top-down method.

FIG. 25 is a diagram for explaining the relationship between the stages.

[Fig. 26] Fig. 26 is a diagram for describing the relationship between the stages when decomposed into educational elements.

FIG. 27 is a diagram showing correspondence between educational elements and cell names.

FIG. 28 is a diagram showing how an integrated cell is configured in the sixth prerequisite technique.

FIG. 29 is a diagram showing how necessary information is selected in the relational model in the seventh prerequisite technique.

FIG. 30 is a diagram showing how necessary information is projected in the relational model.

FIG. 31 is a diagram showing how necessary information is decomposed into cells in the cell model.

FIG. 32 is a diagram showing how necessary information is selected in the cell model.

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G06F 17/60 504 G06F 17/60 504 ZEC ZEC (72) Inventor Sachiko Yonezawa 2-17 Fujimi, Chiyoda-ku, Tokyo -1 (72) Inventor Niki Suzuta 2-17-1 Fujimi, Chiyoda-ku, Tokyo (72) Inventor Toshio Kodama 2-17-1 Fujimi, Chiyoda-ku, Tokyo (72) Inventor Yoshito Funahashi 2 Fujimi, Chiyoda-ku, Tokyo -17-1 (72) Inventor Toshi Kunii ▲ Yasu ▼ 1-25-21 Hongo, Bunkyo-ku, Tokyo F-term (reference) 5B075 KK07 ND20 NR10 NR20 QT10 5B082 GA08

Claims (5)

[Claims] 1. Prior to searching, a judging unit for judging a situation, a specifying unit for specifying an attribute of a cell to be searched according to the situation, and a cell including the attribute are searched from a data space. A computer processing device comprising: a search unit; and a processing unit that constructs a cell space including a desired attribute based on the search result. 2. The apparatus according to claim 1, wherein the search unit decomposes the data space into cells in terms of presence or absence of the attribute. 3. The apparatus according to claim 1, further comprising a holding unit that holds a cell operation executed by the search unit or the processing unit. 4. A database that stores information in the form of cells in cell information theory, and a management unit that manages the database, wherein the cell holds a cell ID in its zero cell, and the management unit A computer processing device characterized by adding an effective range to a cell and managing the cell so as to uniquely identify the cell. 5. A computer processing device comprising: a database storing educational elements; and a management unit that grasps the educational elements as cells in the cell information theory and manages a curriculum by cell calculation.