JP2002312376A - Electronic commerce support method and system, and business information management system usable for them - Google Patents

Abstract

(57) [Summary] [Problem] Transactions developed on the Web are local activities, and there has been no method of globally utilizing links between data made there. SOLUTION: A plurality of e-customers 12 and e-malls A16 and B18 are connected to the Internet 14. Of the many attributes of the product, the e-customer 12 is of limited interest, but the e-mall A that expresses it.
The attributes on the side of 16 do not always match the e-customer 12. Then, based on the transaction history, e-customer 12 and e-customer
-Record the attribute correspondence in the mall A16 in the attribute correspondence table 32. Global BIM 34 has multiple e
-Monitors the attribute correspondence table 32 in the mall, and provides a useful relationship for transactions to another e-mall.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic commerce support technology and a business information management technology, and more particularly, to a method, an apparatus, a system, a database, and the like for supporting an electronic commerce and other business using a network.

[0002]

BACKGROUND OF THE INVENTION Cyber worlds are being formed on the Web, whether conscious or spontaneous, and with or without a firm design intent. Local activities that are expanding both in mass and mass are fused together on the Web, forming a global cyber world. “E-business” including electronic financing is actually taking place in the cyber world, which is about to reach the order of the national financial scale. In this context, without the proper modeling of electronic activity on the web and the information it uses, cyberworlds will become more chaotic and will no longer be in the near future. It is thought to be.

[0003]

A landing point for eliminating such confusion does not appear at the level of integrating the current spatial database model and time database model. This is because the myriad local activities on the Web are linked by the Internet infrastructure, and none of them can be captured independently. The current mainstream relational model is
Based on the so-called “world model”, it is assumed that there is an administrator who integrally grasps all data or dependencies between attributes. This model is meaningful as long as the information is closed in an independent place, such as within a company, but it provides a single view of the dependencies between data and attributes that appear in unrelated businesses deployed globally on the web. There is no one. In other words, an information model that is based on the concept of a world model cannot be used as a method for converting activities on the Web into a reusable database.

The present invention has been made based on the recognition and analysis of the present state of the art, and its object is to make it possible to reuse multidimensional information in a complex system such as the Web.
To provide a new database or information model.

[0005]

One embodiment of the present invention relates to an electronic commerce support method. This method includes a step of extracting and recording a correspondence between attributes determined from respective viewpoints of a plurality of entities involved in e-commerce by a cell operation method in cell information theory, and a method of extracting the recorded correspondence from e-commerce. Reading and presenting the scene. These steps are realized in electronic form via a network.
"Cell information theory" is a basic theory of an information model obtained by extending the cell structure theory into a form usable for real business by the present inventor.

[0006] The "subject" refers to an "e-shop" that provides an acquisition unit online, an "e-customer" to purchase it online, and others who participate in transactions.
"Each point of view" means, for example, if an e-customer wants to find out which e-shop is the cheapest
Among the attributes related to transactions, those that are determined based on one's own circumstances. In the transaction, the subjective element of each entity is included. In this aspect and the following aspects of the present invention, the correspondence between the attributes is incremental, that is, based on the cell information theory that can be added one after another. The question of whether or not is a subjective factor is not essential.

Another embodiment of the present invention also relates to an electronic commerce support method. This method includes a step of specifying a correspondence between attributes determined from the viewpoints of a plurality of entities involved in e-commerce, using a predetermined equivalence as a clue, and a cell operation in cell information theory using the specified correspondence. Extracting in the data table and recording it in the data table; storing the correspondence in the data table based on an actual example of e-commerce; and referring to the data table and storing the stored correspondence. Reading and presenting the relationship in another e-commerce scene.

[0008] Another embodiment of the present invention also relates to an electronic commerce support method. This method includes a step of extracting an attribute of interest common to a plurality of entities involved in e-commerce by a cell decomposition method in cell information theory based on a predetermined equivalence relationship. Adding to the cell space corresponding to each subject by a cell bonding technique.

The "attribute of interest" is, for example, the price of a product.
It becomes manifest in the form of the condition of "less than a circle". From the point of view of the e-shop, this naturally corresponds to the offered price as a selling price.
As a predetermined equivalence relation, there is a relation of "1000 yen or less for both (as the price of the product)". Because 100
This is because commodities of 0 yen or less satisfy the same rule, symmetry rule, and transition rule described later. Therefore, by setting the equivalence relation, e- customer is considered to be sold at less than 1000 yen, e- customers subspace e ^q made from the commodity considered as may be bought at 1000 yen or ^less, in other words, the transaction A subspace that satisfies one necessary condition for establishment can be separated from a subspace composed of other commodities in the manner of cell decomposition. These subspaces match the original space in the form of a union with no intersection.

[0010] "Cell space corresponding to each subject"
Is the e-customer cell e ^cSay. To this person
Mathematical operation for presenting products under 1,000 yen
Is the e-customer cell e^cSubspace e^q
It is to attach (attach).

[0011] Another embodiment of the present invention also relates to an electronic commerce support method. In this method, a plurality of entities involved in e-commerce each extract and save a correspondence relationship between attributes of interest from an actual scene of e-commerce. Presenting in the actual scene of a commercial transaction. In this manner, the previously formed correspondence can be reused in the future. As an example, suppose that an e-customer purchases a "carpet" on condition that it is "red". On the other hand, the e-shop has no description about the color of the carpet, and merely publishes a photograph. However, if the deal is successful,
Since this carpet is considered to be red, an attribute of “red” that the e-customer is interested in is acquired as a new correspondence relationship with an attribute of a photograph of this carpet in the e-shop. Thereafter, when "red carpet" is searched for in another transaction, the e-shop can present the carpet. That is, the history or achievement in a certain transaction is effectively used in the future, and in a nutshell, it is possible to reduce the waste of human thinking activities that are repeated countlessly in similar situations in the world.

In this embodiment, it is assumed that a blue carpet is actually purchased. However, the carpet is associated with "red" based on the purchase history. This is an incorrect response. Therefore, the correspondence may be appropriately updated and maintained by referring to the correspondence determined from other transactions.

When more correspondences are accumulated, information that can be reused in a certain transaction scene increases. When the transaction is over, the correspondence can be enhanced again. At that time, the correspondence can be corrected. Therefore, in the method of the present invention, the step of storing the correspondence and the step of presenting the correspondence may be cyclically repeated in such a manner that feedback is provided to each other.

[0014] Another embodiment of the present invention relates to an electronic commerce support system. The system includes a shop connected to a network and providing goods to customers through the network, and a business information management agency connected to the network, wherein the shop is configured to execute the transaction in the real scene of electronic commerce. The business information management institution includes a functional block for cross-referencing the table of each of the plurality of shops, the table including a table that records a correspondence relationship between attributes of interest of each of the plurality of subjects. The business information management organization reuses the correspondence between the attributes established in the locally performed transaction at another place, for example, at another e-shop. Therefore, it is possible to effectively link various activities on the Web.

[0015] The business information management institution further includes a function block for detecting a desired correspondence among the correspondences recorded in a table of any of the shops, and an electronic commerce in another shop for the detected desired correspondence. May be provided in a real-world scene.

On the other hand, the shop may be provided with a local business information management block, and this block may manage the table. The local business information management block may include a maintenance function block that examines the accumulated correspondence and appropriately corrects the correspondence. The maintenance function block may detect inconsistencies among the stored correspondences, leave one of the more likely correspondences, and delete the other correspondence from the table.

Still another preferred embodiment according to the present invention relates to a business information management system. This system generalizes the Join operation in the relational model in the form of identification based on equivalence classes, and shows the correspondence between the attributes of interest of multiple entities involved in the business. A functional block that is fixed and recorded by the identification in the local environment where the identification is performed, a functional block that reads a desired one from the recorded correspondence and reuses it in another business scene, and a result of the business And a function block for maintaining or updating the recorded correspondence based on the With this configuration, the correspondence between attributes formed in the local place is added to the global place under the modeling guideline without assuming the presence of an administrator who grasps all the attribute or data dependencies in a unified manner. To go.

It is to be noted that any combination of the above-described components and any conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, and the like are also effective as embodiments of the present invention.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below based on preferred embodiments. First, as a basis of the embodiment, a cell information model (CIM: Cellular Informat
The basic theory of the ion model will be described as a prerequisite technology, and then specific embodiments will be described.

[Base Technology] A new information model called a cell model is proposed. The cell model is an information model that can be applied to the field of irregular data models, and captures specifications over time and space in the form of a situation. Mathematically speaking, the cell model is an extended theory of graph theory based on the cell space structure theory positioned in the theoretical framework for homotopy.

[1] Modeling of cyber world A theory is constructed based on invariants. Think of the cyber world as a type of space that includes time as an irreversible space,
We show that the invariant including the dimension as the degree of freedom and the connectivity indicating how the spaces of different dimensions are connected are appropriate for the expression.

In general terms, modeling a cyber world requires the following four steps.
First, the differences and similarities between the cyber world and the real world must be clarified. The most obvious difference is in the growth rate, and therefore in its complexity. The nature of linking the local world to the world-wide web world at the same time is very special, and its speed is comparable to the speed of light. In fact, in the history of mankind, the speed realized on the Web has given mankind unprecedented abilities. Everyone working on the web is building and destroying the cyber world at the same time.

Second, there is a need to find an appropriate modeling methodology to characterize the clarified differences and commonalities. Due to extreme complexity and rapidity of change, methods for minimizing the scale of modeling should be built on layered concepts. In addition, the hierarchy should be an invariant hierarchy in order to identify universal characteristics in the ever-changing cyber world, and it should be in a form that allows for the addition of concepts in a modular fashion. .

Third, it is necessary to translate the modeling methodology thus constructed into an actual design. In general, design requires an appropriate choice of invariants and specific information structures and operations. For example, a conceptual hierarchy of invariants is designed as a hierarchy that inherits invariants. Previous work has recognized the importance of two invariants, dimensions as degrees of freedom and their connectivity. That is, the cell space structure was considered as the information structure, and the construction (composition) and decomposition (decomposition) of the cell were considered as the operation.

Fourth, the design thus obtained is implemented as an information model named a cell model. The cell model enhances the capabilities of various existing data models, and can guarantee all cell boundaries, cell dimensions, and cell connectivity. Cell models can represent the cyber world in a consistent way and prove its legitimacy. [2] Concept hierarchy of invariants Modeling is a very important step in scientific research. Especially in natural science, theories are built around the concept of invariants in order to model the real world. Objects and phenomena are classified and modeled based on invariants. In physics, energy and mass were invariant until relativity was published. In mathematics, the following steps are taken to model an object. That is, a mathematical object is classified into an equivalence class that can be expressed as an exclusive OR of the subset by the equivalence relation. An example of a conceptual hierarchy of equivalence relations is as follows.

1. 1. Equivalence relations in set theory 2. Extended equivalence relations, with special cases of homotopy equivalence relations. 3. Topological equivalence relations, as a special case of graph theory equivalence relations. 4. Equivalence relation of cell space structure 5. Equivalence relationship on information model Equivalence relations in presentation

In order to realize a design that is modular and can be added incrementally later, ie a continuous hierarchy of invariants of the cyber world, the following classifications are based on the concept hierarchy of equivalence relations in mathematics Hierarchy. 1. Aggregation level 2. 2. Extended level, homotopy level as a special case. 3. Topological level, graph theory level as a special case Cell structure space level 5. Information model level 6. Expression level

[3] Cell Model In order to model the cyber world, an approach based on a cell space structure such as a CW space is much more suitable than that based on graph theory. According to the cell structure space level, the object is bounded,
Alternatively, it can be positioned as a non-existent cell in a recognizable and computable space. Cells with boundaries are "closed" and cells without boundaries are "open". An n-dimensional cell, that is, an “n-cell” is a space in phase with an n-dimensional sphere (n is an integer). Here, notation open n cells and e ^n, close a n
Cell

(Equation 1) Notation. Also, the inside of a closed n-cell

(Equation 2) Notation. Therefore,

(Equation 3) Is the boundary of a closed n-cell, which is (n-1)
Equal to the dimensional sphere ^Sn-1 . According to cell modeling,
The construction and disassembly of a cell can be realized while keeping the dimensions and connectivity of the cell invariant. Therefore, the identification of the object is systematically implemented through a mapping for identification. As will be described later, the construction of the database schema and the decomposition of the schema correspond to a special case of cell construction and cell decomposition.

Here, an example of the dimension will be described. For example, in a cyber world, an object having one attribute cannot move from one attribute to another attribute, and thus has a degree of freedom of 0 and therefore a dimension of 0. Therefore, this can be represented by a "point" at the expression level. An attribute is an independent set for identifying a characteristic or characteristic that an object originally has. In an object having two attributes, it is possible to shift from one attribute to the other attribute, so that the degree of freedom and the dimension are “1”. Therefore, it can be represented as a straight line at the expression level. Similarly, attributes 3 and 4
Objects correspond to two dimensions and three dimensions, respectively, and can be expressed as a curved surface and a sphere. In general, an object with n attributes has (n-1) degrees of freedom,
Its dimension is n-1. This can be represented as a (n-1) -dimensional sphere. In the relational model, an object having n attributes is represented as a relational schema, and is embodied as an n-column table. The relational model is based on the direct product of sets, so it can be said to be an expression at the set theory level.

On the other hand, the connectivity is defined by an adhesive map (attaching map) which is a continuous and surjective map. A mapping f: X → Y is surjective

## EQU4 ## This means (∀y∈Y) (∃x∈X) [f (x) = y]. “Mapping f: X → Y is continuous”
"{F ^-1 (y) | y {A} is open in X, and only if A is a subset of Y is Y
Open in "means.

For phase spaces X and Y having no common part,

(Equation 5) Is a bonding space obtained by bonding X to Y by the bonding map f: X ₀ → Y. here,

(Equation 6) Represents exclusive OR, and is often represented by +.
Indicates an equivalence relationship. The equivalence relation is defined as the same rule "x ~
x ", the target rule" xy if y-x ", and the transitive rule" x-z if x-y and y-z ". There are relationships. Transitivity divides a space into subspaces called equivalence classes that have no common part.

As a basis for more clearly modeling cyberspace, equivalence relations and equivalence classes will be described. The subset of X defined by x / 〜 = {y∈X: x〜y} is called the equivalence class of x. here,
“Class (class)” is actually a set, but since it has long been called a class, the notation is followed. The set X / ~ consisting of all equivalence classes is called the quotient space of X and is written as:

X / 〜 = ｛x / ∈∈2 ^X | x∈X｝ ⊆2 ^X

From the transition law, the following equation holds for each x satisfying x∈X, x / ≠ φ.

(Equation 8) This means that the set X has been divided or decomposed into equivalence classes that are not empty and have no intersection. Here, the equivalent class is expressed as x /／, which has the following meaning.

X / 〜 = {y∈X | x〜y}

A simple example will be described. "Concentration (cardinalit
y) "is an equivalence relation in set theory, and divides the original set into subsets having the same cardinality and having no intersection. As another example, “isomorphism” is an equivalence relation in graph theory, and a set of graphs can be decomposed into a subset of graphs of the same type that have no intersection.

In Euclidean geometry, "congruence" forms one equivalence relation, and all figures can be decomposed into subsets of congruent figures. These subsets have no intersection and their union matches the original set, that is, the set of all figures. This union corresponds to the quotient space. "Similarity" is one equivalent relationship. “Congruence” and “similarity” are both examples of affine transformation.
On the other hand, the relation “symmetric” is an example of an equivalence relation in the group theory, and is decomposed into a union of subsets formed of symmetric figures and having no common part.

The above is an actual example of the adhesive mapping. Here, the general definition of the adhesive map will be described. The set of all equivalence classes is denoted by X / 〜 and is represented by the following equation.

X / 〜 = ｛x / ∈∈2^X| X∈X｝ ⊆2^X  This is also called the quotient space of X. The bonding map f is surjective and
It is a continuous map below.

F: X ₀ → y (X ₀ ⊂Y)

(Equation 12) Is a quotient space and has the following relationship.

(Equation 13) Here, as described later, a special case for integrating information schemas and integrating information by information mining on the web is considered. Now, S ^n-1 is the boundary of a closed n cell,

[Equation 14] Can be written as Here the surjective and continuous glue map f

F: Define as S ^n-1 → X. At this time, the additional space Y is defined as a quotient space as follows.

(Equation 16) Now, the homotopic maps f and g,

F, g: S ⁿ⁻¹ → X is considered. Then

(Equation 18) The homotopy equivalence relation is generated.

According to JHC Whitehead's point, when an arbitrary cyber world X is given as a phase space, a finite or infinite number of ^Xp cells which are subspaces of X indexed by an integer Z from this X are Arrays can be constructed recursively. That is, a space {X ^p | X ^p } X, ^p {Z} called a filtration can be formed as follows. Where X ^p is called the coating of X, the following relationship holds.

X = ∪_p∈ZX^p  Furthermore, X^p-1Is X^pIs a subspace of

X ⁰ ⊆X ¹ ⊆X ² ⊆... −１X ^{p -1} ⊆X ^p
⊆ ... ⊆X Filtration is also called skeleton. A skeleton of at most p dimensions is called a p-skeleton. ^{X 0, X 1, X 2} ··· X p-1 and ^{X p}
Is a partial cyber world of cyber world X.
A space that is topologically equivalent to the filtration is called a filtration space.

There is a practically important cell space. They are CW complexes and manifolds. When the filtration space is finite, it is equivalent to CW space. Furthermore, when the CW space has differentiability, it is equivalent to a manifold.

[4] Modeling of Web Information Through Information Mining by Cell Model, Integration of Inductive Web Information Schema, and Integration of Web Information As a first step in modeling Web information, how the cyber world emerged We characterize the essence of forming a cyberworld, a shared information world on the Web, to identify what it is. Cyberworld X is often formed on the web as a result of local and diverse activities on many websites. Unlike information in a company, it cannot be assumed that there is an information manager who will provide a set of starting schemas. Through the information mining process, it is possible to discover cyberspace X by discovering special information on a plurality of locally existing websites. Of course, information mining is not random. After browsing a website with a browser, one must extract ideas about what should be mined from the information and its integration distributed across multiple websites and what is expected to appear. This type of information mining is commonly referred to as "design-based information mining." This is because there is a predetermined rule to be applied as an “integration guideline” for an object to be mined. This integration guide serves as a design guide on what to integrate and how.

According to information mining on the web based on Whitehead's inductive scheme described above, the integration of the local web world into the global cyber world is fully realized. The specific method of cyberworlds X ⁿ of n dimensions is obtained by recursive integration is described by the following web search and process integration.

Inductive integration consists of two phases. That is, the information integration phase and the information integration phase. The first phase, the integration phase of the information schema, proceeds according to the following procedure. 1. Attributes that are of interest

(Equation 21) Is read out, the following 0-dimensional cyber world X
⁰ is formed.

Equation 22] ^{_{X 0 = {e 0 1,}} e 0 2, e 0 3, ··· e 0 j} to generate a 2.1-dimensional cyberworlds ^{X 1,}
All combinations of two attributes of interest on a website

(Equation 23) Read out. After that, their union union,

(Equation 24) To adhere to the ^{X 0} and by means of an adhesive mapping F. In this way, it is possible to obtain a one-dimensional cyberworld X ¹ below.

(Equation 25)

Here, i = 1, 2,... K, and the bonding map F is

(Equation 26) It is.

3. It is assumed that the (n-1) -dimensional cyber world X ^n-1 is constructed through information mining by repeatedly reading and integrating attributes. Where X
^n-1 has n attributes. In ^order to integrally generate an n-dimensional cyber world Xn having (n + 1) attributes, (n + 1) attributes of interest on the website in the same manner as before.

[Equation 27] Read all combinations of. Followed by their union union,

[Equation 28] Is adhered to the already constructed (n-1) -dimensional cyber world X ^n-1 via the adhesion map G. As a result,
An n-dimensional cyber world ^Xn can be generated as follows.

(Equation 29)

Where i = 1, 2,... K, and the bonding map G is

[Equation 30] It is. With the above process, integration of the information schema is completed.

On the other hand, the information integration phase, which is the second phase, is extremely simple but requires a large amount of calculation. During this phase, all instances of cell bonding performed during schema integration are inspected to determine and determine which instances should be included in the cyber world created by cell bonding based on design guidelines. I do.

A cyber world constructed based on Whitehead's inductive methodology satisfies the following relational expression.

X ⁰ ⊆X ¹ ⊆X ² ⊆... ＮX ^n-1 X ⁿ
⊆ ・ ・ ・ ⊆X From the viewpoint of the validity of the cyber world, this formula means that any valid cyber world includes the lower level cyber worlds and those cyber worlds are valid. .

In the above example, identification is performed by an equivalence class based on an equivalence relationship.
"Identification by equivalence classes" is a generalization of join operations in relational models. This point shows some of the practical capabilities of the cell model. In view of the highly complex and extremely fast change of the cyber world on the web, this integration capability of the cell model provides a real theoretical basis for a web information model.

When the "attribute of interest" is used to execute the design guideline, "interest" means, at least in a partial sense, the selection of an equivalence relation for identification. . That is, "selection of equivalence relations for identification" is a major part of the design guidelines. In web related information systems, design guidelines exist locally to govern local sites as intranets or community nets, or globally to operate in a borderless cyber world. Design guidelines are reusable resources in web-based information systems.

[5] Situation Modeling of Web Information as Integration of Non-Inductive Information Schema and Integration of Information Based on Cell Model On the Web, it is often necessary to create a new cyber world from an arbitrary cyber world .
This is more common than information mining using the inductive approach described in the previous section, and is a common requirement in e-business, including e-commerce on the Web. For example, consider the situation of e-commerce to model the changing state of the web in both space and time. In order to build an e-commerce information system, it is generally necessary to find the structure of commerce on the Web in terms of an information schema. Typical e-commerce situations include the following.

Situation 1. An e-customer who purchases an item browses the web to find an e-shop that sells the item at the lowest price. Situation 2. An e-shop selling goods on the web browses a list of e-customers to expand sales.

In this situation, on the web we are not interested in finding all the details about e-shops, e-customers and e-products. Here, e- shop, e- customer and e- products were s, and cyberworlds of c and m-dimensional, respectively, thus denoted respectively s cell ^e s, and c cells ^{e c} and m cell ^{e m.}

In situation 1, the e-customer has an e-customer
-When the desired e-product is sold at the cheapest price in the shop, identify the product name in the e-shop with interest as a purchaser. This situation can be characterized by a cell decomposition operation and a subsequent identification operation. The cell decomposition operation is represented by a mapping f shown below.
The mapping f is in the form of adhesive mapping g is stored, projects the arbitrary n-dimensional cell e ⁿ to the union of cells without the two common portions of the next.

(Equation 32)

[Equation 33]

As described later, by preserving the adhesion map in each cell decomposition, the cell decomposition can be made a homotopic. The conclusion about Situation 1 is to understand it with the following situation model. 1. S cell ^e s as cell disassembly e- shop, cell degrades m cell ^{e m} as c cells ^{e c} and e- items as e- customer. At this time, the equivalence cell ^eq is separated from other parts in order to specify the attribute related to the electronic commerce.
Examples of attributes, for example if Shimese to simplify e ^q to e ^2, trade name, there is identity and price of goods of goods.

2. The cell construct equivalence cell e ^q by cell adhesion to identify the adhesion mapping. That, m cell as e- product ^{e m,} and e
- bonding the s cell ^{e s} as a shop to the c cell ^{e c} as e- customer.

Similarly, the situation 2 is embodied as the following situation model. 1. Cell disassembly Same as situation 1. 2. The cell construct equivalence cell e ^q by cell adhesion to identify through adhesive mapping. That is, adhering e- product as the m cells ^{e _{m i}} and e- customer as c cells ^{e _{c i}} to s cell ^{e s} as e- shop.

[6] Homotopy as a Theoretical Framework of Cell Model for Space / Time Information and Space / Time Operation Now that the new millennium has begun, we are very fundamental to the real world. It's time to be able to influence in a different way. It must be said that being able to be present at such a moment is extremely lucky. Building science on the web and the cyber world, which is expected to play a major role in the 21st century, will make the greatest contribution to building web-based information technology. The cyber world is the world of information, and in that sense, the web and cyber world information models are key elements. Equally fortunate, we have the mathematical framework needed to create the science described as a cell spatial structure. The same applies to the homotopy theory described below.

The homotopy theory works as a basic theory of the cell space structure. That is, when dealing with changes in the time and space of the cyber world, homotopy theory is used to accommodate space / time information and space / time operations. Now, for example, consider a change in the mapping function f from one phase space X to another phase space Y. After change,
f becomes another mapping function g. Therefore, the following continuous deformation from f to g is designed.

F, g: X → Y This deformation is considered for a normalized section [0, 1]. This section may be temporal or spatial. Now, a portion A of the phase space X that does not change is denoted as A⊂X as a subspace A of X. The homotopy H to be designed is as follows.

H: X × I → Y where ({x}) (H (x, 1) = f (x) and H (x,
1) = g (x)) and (∀a∈A, ∀t∈I) (H (a, t) = f (a) = g
(A)) At this time, f is called a homotopic with g with respect to A, and is represented as follows.

[Equation 36] Here, a new design problem arises. That is, two phase spaces X and Y are homotopy equivalent,

(37) That is, a method of designing them so as to have the same homotopy type. This is solved by the following procedure. That is, f: X → Y and h: Y → X may satisfy the following conditions.

(38) Here, 1 _X and 1 _Y are identity maps and satisfy the following equation.

[Expression 39] 1 _X : X → X and 1 _Y : Y → Y

With the above method, the dimension of the cell can be changed to a homotopic. Homotopy equivalence is a broader concept than topology equivalence. Homotopy equivalence can identify any changes in the cyber world that are no longer topologically equivalent before and after the change. The cyber world continues to evolve through various operations and processes, and the process of the change is specified by homotopy, and the validity is guaranteed by homotopy equivalence. For example, when performing each cell decomposition, one can see why the glue map is preserved. It keeps cell decomposition homotopic and
Therefore, the process of cell disassembly can be reversed.

Research on homotopic information models is an area that should be addressed in the future in order to explore the science of information models. It is a very interesting research topic to consider what information operations result in homotopy equivalence.

[Specific Example] An example in which the business information management system according to the present invention is applied to an electronic commerce support system based on the above prerequisite technology will be described below. This example is a series of transaction systems in which goods are presented from an e-shop to an e-customer via the Internet, and the e-customer purchases a desired one of those goods.

Here, a plurality of e-shops gather to form two e-mall, respectively. In addition, a global business information management system for sharing correspondence information between attributes between these e-malls has been constructed. In this example, the reason for taking an e-mall formed by a plurality of them rather than a single e-shop is that the same product is provided from a plurality of e-shops,
This is because e-customers can more easily search for a product that meets their wishes from among them. That is, the business information management system according to the present invention has a certain volume at least locally even if there are a plurality of pieces of information that can be selected among transaction entities to some extent and they are not managed globally in a unified manner. It is assumed that it can be managed with.

In FIG. 1, on the Internet 14, there are provided e-customers 12, which are a plurality of PC-customers, and e-mall A, each of which is an aggregate of a plurality of e-shops.
16 and e-mall B18 are connected. The e-mall A16 stores products of a plurality of e-shops opened in the mall in a unified or distributed manner.
Product database 30, e-customer 12 and e-customer
When a transaction is concluded between the malls A16, each of the malls A16 has an attribute correspondence table 32 which describes a correspondence relation of an attribute of interest. The e-mall A16 includes therein a search processor 20 for searching the e-product database 30 for product information desired by the e-customer, and an attribute correspondence table based on transaction history or results. 32, and maintain it as necessary (referred to as “business information management organization”
M (abbreviated as M) 22. The above configuration is the same for the e-mall B18.

The global BIM 34 is the e-mall A1
6 and the local BIM 22 of the e-mall B 18 as appropriate, and controls the correspondence relationship to be described in the attribute correspondence table 32. In addition, the global BIM 34 displays the correspondence established in the one mall e-mall A16 at the site of the transaction in the other mall e-mall B18 as appropriate. Is referred to. Global BIM3
Reference numeral 4 denotes, for example, a third party who is requested by a plurality of malls to construct an e-commerce system, and undertakes its business. The structure of the table 32 is designed.

A plurality of e-mall members mutually agree on the exchange of information for facilitating transactions. Once such a cooperative relationship can be established between two e-malls, the efficiency of system operation by sharing the attribute correspondence is a so-called modular property based on the prerequisite technology, that is, a structure in which necessary information can be added as needed. Therefore, acceleration becomes remarkable. Therefore, it is considered that e-malls that did not initially subscribe to the e-commerce support system 10 will be encouraged to subscribe to this system in the future.

[0065] Figure 2 is a cell ^e m indicating the e- products, e- cell ^{e c} represents the customer, and e- shops, that is shown here related cell ^{e s} representing the e- mall. In the figure, a common subspace e ^{m, e} s cell e ^{q occurring,} and the overlapping portions of the e ^c can cause a deal. That is, the area of the cell e ^s representing the first e- malls, and various attributes about the product to which the e- mall provides the attributes representing the e- mall itself is identified. On the other hand, the cell e ^c is e- customer attributes and their e- customer's own product attributes are identified to determine. Therefore, when considering these spaces from the direction of passage of trading now, the intersection of cell e ^s and cell e ^c is
For each attribute that is of interest as the subject of the transaction, it represents the part where supply and demand conditions are balanced.

Further, in this common part, a cell e representing a product
^m regions overlap. Cell e ^m of this product is an attribute of the product itself, but is covered by the product information that is provided in many cases from e- mall, such as information that only knows also the manufacturer of the product in addition to it , Which does not necessarily include information that the e-mall does not have. Therefore,
In electronic commerce, it is possible to form a pattern by overlapping three areas as a general form.

Now, the cell eq, which is a common part of these three areas, is a product that can be provided under the condition that both the e-mall and the e-customer, which are the main ^parties of the transaction, satisfy, for example, “15,000 yen or less”. Represents. Thus, by using the generalized expression in the base technology, cell e ^q is a subspace to hold the transaction is equivalent to the separated equivalence cell e ^q as part satisfying the equivalence relation for example is set with respect to the price of products In order to extract this part, the cell decomposition operation in the base technology is used.

In the base technology, an object having n attributes is defined as a (n-1) -dimensional sphere. this is,
When applying the cell information model to actual database construction, consideration is given to giving each attribute an index using at least one attribute. That is, when the base technology is considered as pure mathematics, an object having n attributes can be simply referred to as an n-dimensional sphere. However, the base technology is intended to be applied to an actual business scene as shown in FIG. 1, and there is a practical consideration of the present inventor in assigning one of n attributes as an index.

FIG. 3 shows the enrichment of the correspondence between attributes in the electronic commerce support system 10 and the feedback relationship of reuse. That is the subject of the transaction cell e ^q is a subspace for a deal cell e ^c
Or adhere to e ^s by calculation of the "cell adhesion 40"
The desired product is presented to the e-customer in a specific form, while the product that the e-customer wants to purchase is specifically presented to the e-mall. Through this presentation process,
"Transaction establishment 42" occurs. At this time, by examining the transaction history, a new correspondence that did not exist before can be found. The correspondence is determined from the relationship between the search keyword or the like given by the e-customer when searching for a desired product and the information used by the e-customer among the information presented from the e-mall as a result, and the like. Is specified under Therefore, according to the system of the present embodiment, a new cell bonding 40 is created by the establishment 42 of the transaction, and the correspondence relationship by the newly bonded cells is presented in a future transaction field, and these are successively formed as a loop. It is responsible for accumulating information necessary for transactions.

FIG. 4 shows a screen 50 for the e-customer to search for a desired product. Here, “tea cup” is specified as the target product 52. Also, e-
As a plurality of conditions 54 for a customer to search for a tea cup, “white or blue”, “less than 15,000 yen”, “round feeling”, and “made in Europe” are input in a free description format. Among those conditions, those specified by the e-customer as essential conditions are checked in the “must” column 56. Therefore, in this example, the e-customer inputs as a required condition that the tea cup is 15,000 yen or less and made in Europe, specifies white or blue as the color, and gives a round impression as a whole impression. specify. The input search condition is notified to the search processor 20 of the e-mall A16, and the search processor 2
0 extracts a desired product from the e-product database 30. Please note that "white or blue" is a color specification.
Generally, it is understood only by human judgment, and this does not always match the information of the product provided from the e-mall, that is, the attribute as it is.

FIG. 5 is a screen 80 showing one of the search results presented by the e-mall under the search conditions input by the e-customer. Here, “22” is described in the result column 82 as a search result. A picture 84 of the product and a product description 86 are described, and a button 88 is provided for the e-customer to press when purchasing the product. It should be noted here that the search results presented from the e-mall do not always completely satisfy the search conditions input by the e-customer. For example, the e-mall may extract 22 search results by referring only to the condition of "15,000 yen or less", which is relatively easy to determine automatically.
Therefore, the search results of the 22 cases may include considerable noise in some cases.

In the example of FIG. 5, the tenth tea cup among the 22 search results is displayed. When the display of this product is advanced, it is assumed that the e-customer has determined that this is the product closest to his / her desire. That is, the price is 13,000 yen, the color is "blue onion" based on white and blue, and the whole impression is round and made in Europe, here made in Germany. Conforms to. Where e-
Assume that the customer has pressed the purchase button 88. This instruction is used for proceeding to settlement at the e-mall, and is notified to the search processor 20 and transferred from the search processor 20 to the local BIM 22, so that a new correspondence relationship is set in the attribute correspondence table as necessary. 32.

FIG. 6 shows the internal configuration of the local BIM 22. The local BIM 22 includes a correspondence generation unit 102, a correspondence presentation unit 104, and a correspondence maintenance unit 106. Correspondence generation unit 10
2 receives the notification from the search processor 20 when the transaction is completed, and records the correspondence between the newly specified attributes in the attribute correspondence table 32. For example, in the case of the example of FIG. 5, any attribute of this tea cup, for example, an attribute such as "white or blue" or "round feeling" for the photograph is added if it does not exist as a correspondence before that,
It is recorded in the attribute correspondence table 32. “Made in Europe” is recorded in a form having a correspondence with any attribute of the product description 86 of the product.

The correspondence presenting unit 104 stores the attribute correspondence table 3
Based on the correspondence between the two, necessary information is provided to the search processor 20 in another transaction. For example, when another e-customer wants to purchase a tea cup at another opportunity and specifies “white” or “blue” as the color, FIG.
Can also be extracted as candidates.

On the other hand, the correspondence maintenance unit 106 selects one of the correspondences stored in the attribute correspondence table 32 one after another.
Modify or delete what is considered to be a wrong response. For example, "red" as the color of the tea cup e
If the e-customer purchases the teacup shown in FIG. 5 despite the customer's designation, "red" may be associated with this teacup attribute. However, if, for example, there are 100 transaction examples relating to this tea cup, and the majority of the remaining cases, for example, 80, suggest that the color of this tea cup is white or blue, the corresponding maintenance unit 106 detects that fact, “Red” associated with the attribute of the tea cup is deleted from the attribute correspondence table 32. As a result, the correspondence is appropriately restored, and smoothness in future transactions is secured.

FIG. 7 shows the internal structure of the global BIM 34. The global BIM 34 includes a local BIM controller 120, a correspondence search unit 122, and a correspondence information providing unit 124. Local BIM controller 12
0 monitors the local BIM 22 of each of the plurality of e-malls, e-mall A16 and e-mall B18, and gives instructions to them as needed. Thus, the local BIM controller 120 may, for example,
A function of upgrading the local BIM 22 installed in the mall, and any local BIM 22
Removes what appears to be an incorrect response from the correspondence,
By notifying this to the other local BIMs 22, the attribute correspondence table 3 in each e-mall
2 can be maintained, or the validity of the attribute correspondence table 32 can be increased.

When no useful correspondence is found in one of the e-malls, the correspondence search unit 122 searches all e-mall attributes in order to detect them from the attribute correspondence table 32 of another e-mall. The correspondence table 32 is monitored. As a result, a useful correspondence is
If found in a mall, the fact is notified to the correspondence information providing unit 124.

The correspondence information providing unit 124 includes the correspondence search unit 1
In response to the notification from the e-mall 22, the local BIM 22 of the necessary e-mall is notified of the useful correspondence. The e-
The local BIM 22 of the mall may newly register this correspondence in the attribute correspondence table 32, or may skip the registration operation and store the correspondence in the search processor 2.
Tell 0. As a result, the correspondence is effectively used thereafter in the transaction scene in the e-mall. In this embodiment, since the global BIM 34 always links the plurality of attribute correspondence tables 32 effectively from a global standpoint, the local BIM 22 associates the valuable correspondence found in different e-mall with the attribute correspondence table. There is no need to record data in the table 32 one by one. That is, according to the present embodiment, the local B of each e-mall
The IM 22 only needs to record in the attribute correspondence table 32 only the correspondence relation relating to transactions that have occurred locally in the e-mall. In this sense, it can be said that the entire system is an autonomous distributed database management mechanism. .

The present invention has been described based on the embodiments. As for this embodiment, naturally, various modifications and combination or modification of elemental technologies that can be understood by those skilled in the art are possible. It is understood by those skilled in the art that such techniques are also included in the scope of the present invention. The following is an example.

In the present embodiment, the cell information theory constructed by the inventor has been applied to the electronic commerce scene. However, the scope of the present invention is not limited to this. For example, if any business is conducted over a network, they are usually done by multiple entities. Therefore, the present invention can be applied to the correspondence of attributes between entities in such a business.

In FIG. 1, the global BIM 34 is provided outside the e-mall A16 and the e-mall B18 as an independent structure. However, the global BIM 34 has its functions as e-mall A16 and e-mall B1, respectively.
8 in which case the local BIM 22
And the global BIM 34 are integrated in each mall. At this time, the correspondence uniquely generated in each e-mall can be used globally by direct communication between BIMs included in each e-mall across the boundaries of the e-mall.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an electronic commerce support system according to an embodiment.

FIG. 2 is a diagram illustrating a relationship between cell spaces of a plurality of entities involved in electronic commerce and a subspace that is a common part thereof.

FIG. 3 is a diagram showing a cycle in which highly reusable information is accumulated according to cell information theory in the present embodiment.

FIG. 4 is a diagram showing conditions input when an e-customer searches for a desired product.

FIG. 5 is a diagram showing products searched in an e-mall based on conditions input by an e-customer.

FIG. 6 is an internal configuration diagram of a local BIM.

FIG. 7 is a diagram showing an internal configuration of a global BIM.

[Explanation of symbols]

12 e-customer, 16 e-mall A, 18
e-mall B, 20 search processor, 22 local BIM, 30 e-product database, 32
Attribute correspondence table, 34 global BIM.

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 17/60 ZEC G06F 17/60 ZEC

Claims (12)

[Claims] A step of extracting and recording a correspondence between attributes determined from viewpoints of a plurality of entities involved in electronic commerce by a cell operation method in cell information theory; E-commerce support method, comprising: reading and presenting the scene. 2. A step of specifying a correspondence between attributes determined from respective viewpoints of a plurality of entities involved in e-commerce using a predetermined equivalence as a clue, and using the specified correspondence as a cell in cell information theory. Extracting by an operation method and recording in a data table; storing the correspondence in the data table based on an actual example of electronic commerce; and referring to the data table and storing the correspondence. Reading and presenting the correspondence in another electronic commerce scene, and an electronic commerce support method. 3. A step of extracting an attribute of interest common to a plurality of entities involved in electronic commerce by a cell decomposition method in cell information theory based on a predetermined equivalence relation; And e. Adding to a cell space corresponding to each subject by a cell bonding method. 4. A process in which a plurality of entities involved in e-commerce each extract and save a correspondence between attributes of interest from an actual scene of e-commerce; Presenting the electronic commerce in a real scene of the electronic commerce. 5. The method of claim 4, wherein said storing and presenting steps are cyclically repeated with feedback from one another. 6. A shop connected to a network and providing goods to customers through the network, and a business information management agency connected to the network, wherein the shop performs the transaction in a real scene of electronic commerce. The business information management institution includes a functional block for cross-referencing the table of each of a plurality of shops. An electronic commerce support system characterized by the following. 7. The system according to claim 6, wherein said business information management institution further has a function block for detecting a desired one from a correspondence recorded in a table of any shop. 8. The system according to claim 7, further comprising a function block for providing the detected desired correspondence to a real scene of e-commerce in another shop. 9. The system according to claim 6, wherein a local business information management block is provided in the shop, and the block manages the table. 10. The system according to claim 9, wherein the local business information management block includes a maintenance function block for examining the accumulated correspondence and appropriately correcting the correspondence. 11. The maintenance function block detects an inconsistent one of the stored correspondences, leaves one of the more likely correspondences, and deletes the other correspondence from the table. System. 12. A generalization of a join operation in a relational model in the form of an identification based on equivalence classes, and a correspondence between attributes of interest of a plurality of entities involved in a business is determined by the business operation. A function block that is fixed and recorded by the identification in the local environment; a function block that reads out a desired one from the recorded correspondence and reuses it in another business scene; And functional blocks for maintaining or updating the recorded correspondences, and these functional blocks are used for modeling guidelines that do not assume the presence of an administrator who centrally grasps all attribute or data dependencies. Originally, the correspondence between attributes formed in a local place was glowed. A business information management system based on cell information theory, characterized by adding it to a balmy place.