JP6192110B2 - Data exchange support method and apparatus - Google Patents

Description

  The present invention relates to a data exchange support method and apparatus for supporting equivalent exchange of possessed data among a large number of entities possessing data, and more particularly to data exchange between entities having a low value of possessed data and entities having a high value. The present invention relates to a data exchange support method and apparatus that enable equivalent exchange.

  In Patent Document 1, as a solution to the inter-subject coordination problem in an ad hoc network, the value of transmission / transfer data is determined by an evaluation formula, and the transfer of data to the main body that has transferred by exchanging data of equal value between adjacent main bodies A method for introducing an incentive function into an ad hoc network by giving consideration has been proposed.

JP 2006-222738 A

  In the conventional data center and data sharing mechanism, data is shared only by simple rules between users, and there is a problem that it is difficult to exchange data when there is no connection between users.

  Moreover, in general data centers and data sharing mechanisms, for example, when sharing marketing data, etc., it is common to exchange data with each other's permission. There is a problem that data exchange cannot be established when there is a difference in quantity and they are not equivalent.

  Furthermore, the value of the data depends on the presence and content of other data that the person who obtains the data already possesses in relation to the data. For example, the value of each data in a set of data handled Is low for those who already have it, but is high for those who do not.

  For example, when a data group can be expressed in a matrix arrangement in which data of the feature amount is arranged in a matrix for each target, the value of data lacking in rows and columns that possess at least one data is the value of other data. Higher than.

  As shown in FIG. 20, the value of the personal information of the customer (user) becomes higher as the data about the feature quantity such as address, name, age, gender, occupation, preference tendency, etc. is collected. Even if you only have data about, the utility value is not so high. Therefore, if you have name, age, gender, occupation, and preference data for User A, and you only have name data for User B, you can obtain data on User A's address if you can obtain this. It can be said that the value is high because all the data on A is available. Similarly, it can be said that the name data of the user 価 値 is highly valuable because the name data of all users can be obtained if this data can be acquired.

That is, when a data group can be expressed in a matrix arrangement in which the feature value data is arranged in a matrix for each target, the value of the data that is deficient in the rows and columns that possess at least one data is the value of the other data. Higher than.
However, until now, the value of each data could not be evaluated based on the circumstances of the side that acquires it, for example, the type and number of data already acquired in relation to the data. It was an obstacle to the spread of equivalent exchange.

  SUMMARY OF THE INVENTION An object of the present invention is to provide a data exchange support method and apparatus that solves the above-described problems of the prior art and realizes equivalent exchange of data between a subject having a low value of possessed data and a subject having a high value. .

  In order to achieve the above object, the present invention is characterized in that a data exchange support apparatus for supporting equivalent exchange of data among a plurality of subjects possessing data has the following configuration.

  (1) A means for evaluating the value of data possessed by each entity, a sharing entity that combines entities that can exchange their possession data with each other and share each possessed data, and a sharing entity An entity capable of equivalently exchanging possessed data with each other is combined with the shared entity to be a shared entity capable of sharing each possessed data, and has means for repeating the same, and an entity capable of exchanging possessed data with each other Added support for data exchange with sharing entities.

  (2) The data exchange support device of the present invention further includes means for converting the data possessed by each subject into a geometric figure based on the row and column elements of the data, the geometric figure of the data possessed by one subject, and the like. And means for calculating the value of the acquired data based on a relative positional relationship with the geometric figure of the data acquired from one subject (for example, a complementary relationship in which the data contact each other).

  (3) The data exchange support device of the present invention further includes a total area of a rectangular shape included in a geometric figure obtained by complementing a geometric figure of data possessed by one subject with a geometric figure of data obtained from another subject. A calculation means is provided, and the value of the acquired data is calculated so as to increase as the total area of the rectangular shape increases.

  (4) The data exchange support device of the present invention further calculates the contact line length between the data geometry possessed by one subject and the data geometry obtained from the other subject, and the value of the obtained data Is calculated based on the contact line length-dependent value, the value of which increases as the contact line length increases.

  (5) The data exchange support device of the present invention further calculates the data size of data acquired from one other subject, the value of the data to be acquired, the size-dependent value and the contact that increases as the data size increases The calculation is based on the sum of line segment length dependent values.

According to the present invention, the following effects are achieved.
(1) Since the entities with low value of possessed data join together to form a shared entity, data can be exchanged with other entities having data equivalent to the total value of the possessed data of each entity, so possessed data It is possible to exchange data equivalently between a low-value subject and a high-value subject.

  (2) Focusing on the fact that the value of the acquired data depends on the relative relationship with the possessed data, the relative relationship of each data that is difficult to quantify was quantified by converting each data into a geometric figure. Since the value of the acquired data is evaluated based on the quantitative relationship between the geometric figures imitating each data, the value of the data can be evaluated quantitatively.

  (3) Since the value of the data is evaluated based on the contact line length-dependent value and the size-dependent value of the geometric data, it is also possible to exchange data with different contact line lengths and sizes. As a result, the object and range of data exchange are expanded and active data exchange becomes possible.

It is the functional block diagram which showed the structure of the principal part of the data exchange assistance apparatus which concerns on one Embodiment of this invention. It is the figure which expressed typically the method of evaluating the value of the data which each subject has as a value for other subjects which acquire the data concerned. It is the figure which showed an example of the concept dictionary. It is the figure which showed the method in which the data exchange order determination part determines the combination pair of data exchange, and its exchange order. It is the figure which showed the procedure which couple | bonds the main bodies which can carry out equivalent exchange of possession data mutually, and makes it a shared subject. It is the figure which showed the example whose value of possession data is relative. It is the figure which showed sharing subjecting in case the value of possessed data is relative. FIG. 4 is a diagram showing an example of a two-dimensional array of data possessed by a subject N1. It is the figure which showed the example of the two-dimensional arrangement | sequence of the data which the main body N2-N5 has. It is the figure which showed the example of the geometricalization of the data which the main body N1 has. It is the figure which showed the example of the geometricalization of the data which the main body N2-N5 has. It is FIG. (1) for demonstrating size dependence value. It is a functional block diagram of 2nd Embodiment of a data exchange assistance apparatus. It is FIG. (2) for demonstrating size dependence value. It is the figure which showed the calculation method (the 1) of a contact line segment length dependence value and a size dependence value. It is the figure which showed the calculation method (the 2) of a contact line segment length dependence value and a size dependence value. It is a figure for demonstrating optimization of a geometric figure. It is FIG. (1) for demonstrating the example of optimization of a geometric figure. It is FIG. (2) for demonstrating the example of optimization of a geometric figure. It is the figure which expressed typically the evaluation method of the data which each subject has.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a functional block diagram showing the configuration of the main part of a data exchange support apparatus 1 according to an embodiment of the present invention, and here, the configuration unnecessary for the description of the present invention is omitted. The data exchange support device 1 of the present invention is connected to a data possessing entity (operator) N that supports equivalent exchange of data via a network NW.

  In the data exchange support device 1, the data management unit 10 manages data (owned data) possessed by the subject N participating in the equivalent exchange of data. The possession data evaluation unit 20 evaluates possession data of each subject N in an absolute or relative manner, and quantifies its value.

  As will be described in detail later, the data exchange order determining means 30 determines the combination pair of the principal N and the common principal M that exchange data with each other based on the value of the possessed data of each principal N and the order thereof. A functional unit that combines entities that can exchange equivalent possession data with each other and share each possessed data, and an entity that exchanges possession data with each other. A sharing entity that can share each possessed data, and a functional unit that repeats this.

  The data exchange support unit 40 supports data exchange between the combination pair of the principal N and the sharing principal M based on the determination result by the data exchange order determination means 30.

  Next, a method for evaluating possessed data by the possessed data evaluating unit 20 will be described by taking an example of direct data exchange between a pair of subject pairs (N1, N2). In the embodiment, when an array of possessed data of each subject N can be expressed by a matrix arrangement of rows (subjects) columns (features), data lacking in rows and columns possessing at least one data among the subjects Are evaluated as equivalent.

  As shown in FIG. 2 (a), the subject N1 possesses height data D11, D21 and weight data D12, D22 as the feature quantities of the two subjects U1, U2, respectively. I do not have it. As shown in FIG. 2 (b), the subject N2 possesses weight data D22 and D32 and age data D23 and D33, respectively, as characteristic quantities of the two subjects U2 and U3, but possesses height data. Not done. In the present embodiment, it is assumed that the data management unit 10 already knows the state of distributed possession of data in each subject N.

  In such a distributed possession situation of data, considering the value for the subject N1 of the data possessed by the subject N2, the subject N1 already possesses the weight data D22 of the subject U2, so the subject N2 The weight data D22 of the subject U2 possessed is not worth (X mark, the same applies hereinafter).

  On the other hand, the subject N1 does not possess the age data D23 of the subject U2, and possesses the height data D21 and weight data D22 of the subject U2, so the subject possessed by the subject N2 for the subject N1 The age data D23 of the person U2 is particularly valuable (circles, the same shall apply hereinafter).

  In addition, since the subject N1 does not have the weight data D32 of the subject U3 and the weight data D12 and D22 of the subjects U1 and U2, the subject N1 owns the subject U3 possessed by the subject N2. The weight data D32 is particularly valuable.

  On the other hand, the age data D33 of the subject U3 possessed by the subject N2 has no value for the subject N1 in the initial stage, but the subject N1 can separately obtain the height data and weight data of the subject U3, or the subject Value can be increased if U1 and U2 age data can be acquired separately. That is, for the subject N1, the age data D33 of the subject U3 possessed by the subject N2 has a potential value (Δ mark, and so on).

  Considering the value for subject N2 of the data held by subject N1 from a different perspective, subject N2 already owns weight data D22 of subject U2, so subject U2 possessed by subject N1 for subject N2 The weight data D22 has no value.

  On the other hand, the subject N2 does not possess the height data D21 of the subject U2, and possesses the weight data D22 and age data D23 of the subject U2, so that the subject N1 possesses the subject N1. The height data D21 of the person U2 is particularly valuable.

  Since the subject N2 does not have the weight data D12 of the subject U1, and the weight data D22, D32 of the subjects U2, U3, the subject N2 has the subject U1 possessed by the subject N1. Weight data D12 is valuable.

  On the other hand, the height data D11 of the subject U1 possessed by the subject N1 has no value for the subject N2 in the initial stage, but the subject N2 can separately acquire the weight data and age data of the subject U1, or the subject U2 , If U3 height data can be acquired separately, the value will be higher. That is, the height data D11 of the subject U1 possessed by the subject N1 has a potential value for the subject N2.

  From the above, the possessed data evaluation unit 20 has the data D12, D21 possessed by the subject N1 and the data D23, D32 possessed by the subject N2 are equivalent in value, and the data D11 possessed by the subject N1 and the subject N2 It is determined that the potential value is equivalent to the data D33 possessed by.

  In addition to the above, the strength of the relevance between the feature amount of the newly acquired data and the feature amount of the data already possessed is reflected in each value. May be determined to be high.

  This can be realized by using a concept dictionary. For example, when there is a concept dictionary as shown in FIG. 3, “sex” and “age” are close concepts [FIG. "Fukuoka" and "Sakura" can be regarded as distant concepts [Fig. (B)]. Counting the steps taken by the concept dictionary at this time, the former is “2” and the latter is “5”. If the value function is set so as to be inversely proportional to the number of steps, the closeness of the concept can be determined. Can be reflected in value. If there are a plurality of paths, these paths may be added up.

  It is also desirable to estimate the reliability of each subject N. The reliability may be estimated by voting by each subject N. Further, in the above-described data worth determination, the data loss ratio may be used.

  Further, in the present embodiment, it is assumed that the authenticity of the data is secured, but the authenticity of the data may be estimated at the same time, and the result may be reflected in the value. The authenticity of data can be calculated by the deviation from the average when the same data is under a plurality of subjects and the values of the data are different.

  The data exchange order determining means 30 determines the data exchange order on the premise that each combination pair of the principal N and the sharing principal M accepts the data exchange in ascending order of the data exchange order. The data exchange support unit 40 provides data exchange support between the new shared entity M, which is a combination of the combination of the principal N and the shared entity M, which has agreed to exchange data, and another entity N or the shared entity M in the data exchange order. Therefore repeat.

  FIG. 4 is a diagram showing a method in which the data exchange order determination means 30 determines a combination pair of data exchanges and their exchange order. Here, focusing on five main bodies N1 to N5, each main body N1 to N5 is shown. An example in which the absolute values of the possessed data are “11”, “6”, “3”, “2”, and “2” will be described.

  First, the total value of possessed data of all subjects N1 to N5 is calculated. In this embodiment, “24” is obtained as the total value. Next, the total value “24” is divided by the number of subjects that can be combined simultaneously. In the present embodiment, the description will be made on the assumption that the number of simultaneously coupled subjects is “2” and the total value is branched into two systems of value “12”.

  Next, an entity N whose value of possessed data is close to the value (system value) “12” is selected from all the entities N. In the present embodiment, the subject N1 whose possessed data value is “11” is selected and distributed to one system. Thereafter, the difference “1” between the system value “12” and the value “11” of the possessed data of the subject N1 is distributed to the other system, and the other system value is updated to “13”.

  Next, the other system value “13” is branched into two systems with the system value “6.5” based on the number of simultaneously connected subjects “2”, and the value of the possessed data is selected from the remaining subjects N. Is the subject close to the system value “6.5”, here, the subject N2 having the value “6” is selected and distributed to one system. Similarly, the distribution of each subject is repeated, and finally a system diagram as shown in FIG. 4 is created.

  When the system diagram is obtained as described above, the two principals N4 and N5 with the value “2” assigned last are made candidates for the data exchange pair as shown in FIG. N4 and N5 are asked whether data exchange is allowed. When the data exchange permission is obtained from the principals N4 and N5, the principals N4 and N5 are made into a data exchange pair and free data exchange is possible. Further, as shown in FIG. 5B, the principals N4 and N5 Becomes a sharing entity, and a sharing entity of value “4” M45 is born.

  Since the value “4” of the possessed data of the shared subject M45 is substantially equivalent to the value “3” of the possessed data of the subject N3, the subject N3 and M45 are now candidates for the data exchange pair, and each subject N3, M45 is asked whether data exchange is allowed. When permission of data exchange is obtained from both parties, the parties N3 and M45 are made a data exchange pair and free data exchange is possible. Furthermore, as shown in FIG. 5 (c), the entities N3 and M45 are shared. A shared entity M345 with a value of “7” is born.

  Since the value “7” of the possessed data of the shared subject M345 is substantially equivalent to the value “6” of the possessed data of the subject N2, the subject N2 and M345 are now candidates for data exchange pairs, and each subject N2, M345 asks whether data exchange is allowed. When the data exchange permission is obtained from both parties, the principals N2 and M345 become a data exchange pair and free data exchange is possible. Further, as shown in FIG. 5 (d), the principals N2 and M345 share the data. The sharing entity M2345 with the value “13” is born.

  Since the value “13” of the possessed data of the shared subject N2345 is substantially equivalent to the value “11” of the possessed data of the subject N1, the subjects N1 and M2345 are now candidates for data exchange pairs, and each subject N1, M2345 asks whether data exchange is allowed. When the permission for data exchange is obtained from both parties, the principals N1 and M2345 are used as a data exchange pair, allowing free data exchange.

  According to the present embodiment, entities having low values of possessed data are combined to form a shared entity, so that data can be exchanged with another entity having data equivalent to the total value of possessed data of each entity. Therefore, it is possible to exchange data equivalently between a subject whose value of possessed data is low and a subject whose value is high.

  In the above embodiment, the case where the value of possessed data of each subject N can be absolutely evaluated has been described as an example, but the present invention is not limited to this, and the possessed data of each subject N is not limited to this. The same applies when the value is a relative value that varies depending on the data exchange partner.

  When the values of possessed data of each subject N are relative, all subject pairs that can exchange data with each other are considered in consideration of restrictions on the physical environment such as a network. Then, each subject may be sorted based on the sum of the relative values of data that can be transmitted and received for each subject pair, and the subjects may be shared sequentially from the subject having the smallest total value of the provided data.

As described above, since it is necessary to cover all subject pairs that can exchange data, an appropriate amount of calculation is required. Next, search for a shared entity that can exchange data relatively equivalent to the entity, and identify an entity that gives large data to other entities among the shared entities,
Next, each shared entity may be constructed by searching for a shared entity that can exchange data that is relatively equivalent to the entity and repeating.

  Taking the data exchange between the three principals Na, Nb and Nc shown in FIG. 6 as an example, the total value of the data provided from the principal Nb to the other principals Na and Nc is relative to the data provided to the principal Na. The sum (10) of the value (value for the subject Na) “5” and the relative value “5” of the data provided to the subject Nc is larger than the total value of the data provided by any other subject. Therefore, in this example, as shown in FIG. 7, when the subjects Na and Nc constitute the sharing subject Mac, data exchange not only between the subjects Na and Nc but also between the subject Nb and the subjects Na and Nc Is also possible.

  On the other hand, there is diversity in the value of data, and the relative value between data can vary depending on the evaluation scale, but there is no effective evaluation scale because it is difficult to express the data quantitatively. It is difficult to quantitatively evaluate the value of data acquired by one subject from another subject. However, focusing on the fact that the value of the acquired data depends on the relative relationship with the possessed data, the relative relationship between the data can be quantified by making each data into a geometric figure.

  Here, first, one entity that owns the data is based on the relative position of the data that the subject obtains from the other entity in relation to the data. The evaluation method will be described.

  8 and 9 are diagrams in which a data group possessed by each of the five main subjects N1 to N5 is represented by a two-dimensional array based on the row elements (users) and column elements (features). ) Means that each subject “has” the corresponding data, and a blank means that it does not.

  As shown in FIG. 8, the subject N1 possesses all the feature quantities other than the gender regarding the user “B”, and if the gender data can be acquired, all the feature quantities of the user “B” are gathered. Therefore, like the possessed data of the subject N2 shown in FIG. 9 (a), the data group including the sex data of the user B is data that can supplement the missing part of the data group that the subject N1 already possesses. High value.

  Similarly, the subject N1 has all the data other than the users “Ding” and “戊” regarding the feature “age”, and if the age data of the users “Ding” and “戊” can be acquired, all the features related to age will be available. become. Therefore, the data of the subject N3 shown in FIG. 9B is also highly valuable as data that can supplement the missing portion of the possessed data for the subject N1.

  Similarly, the subject N1 possesses all data other than the user “丙” regarding the feature amount “name”, and if the name data of the user “丙” can be acquired, all the feature amounts related to the name are obtained. Therefore, the data of the subject N4 shown in FIG. 9C is also highly valuable as data that can supplement the missing portion of the possessed data for the subject N1.

  On the other hand, even if the subject N1 acquires the data possessed by the subject N5 shown in FIG. 9 (d), the complement cannot be realized so that the data is aligned in the row or column direction. Therefore, the data group of the subject N5 is not valuable for the subject N1.

  FIG. 10 (a) is a diagram in which the possessed data (group) Ds of the subject N1 is formed into a geometrical figure assuming that each data is a rectangular area having a predetermined unit area, and there is a missing data portion in the data Ds. It is expressed as “chip”, “dent”, and “missing”.

  Here, considering the relationship between the shape of the data group's geometric figure and the value of the data group, when each data can be expressed in a matrix arrangement based on its matrix elements, as described above, more about rows or columns. The more data, the higher the value of the data (group). This means that the larger the area of the rectangle included in the geometric figure of the data group, the more information is obtained and the value is higher.

  For example, when considering the commonality of a combination of multiple feature values, when the feature values `` address '' and `` age '' are obtained, Fukuoka Prefecture has a high proportion of people over 60 years old, You can get information that Tokyo has a large proportion of people in their 40s and 50s. As the number of data increases, the accuracy of information increases. In addition, as the feature quantity increases, more information can be obtained. This means that the larger the rectangle, the more accurate and versatile information can be obtained from the data.

  Therefore, in the present embodiment, the value of data is evaluated based on the total area of a rectangular shape in which all data is arranged in at least one side in the row or column direction in the geometric figure after data complementation. In the example of FIG. 10B, the total area included in the two rectangles K1 and K2 is determined as the value of the data.

  For example, when the subject N1 acquires the possessed data D2 of the subject N2, the missing portion of the data group possessed by the subject N1 is supplemented in the data viewpoint as described above, and the figure viewpoint shows the data shown in FIG. As described above, the total area of the rectangle is increased by complementing the recessed portion.

  Similarly, when the subject N1 acquires the possessed data D3 of the subject N3, the missing part of the data group possessed by the subject N1 is supplemented as described above in the data viewpoint, and the figure viewpoint also shows in FIG. 11 (b). As described above, the total area of the rectangle is increased by complementing the chipped portion.

  Similarly, when the subject N1 obtains the possessed data D4 of the subject N4, the missing portion of the data group possessed by the subject N1 is supplemented as described above in the data viewpoint, and the figure viewpoint also shows in FIG. 11 (c). As described above, the total area of the rectangle is increased by complementing the missing portion.

  On the other hand, even when the subject N1 acquires the possessed data D5 of the subject N5, the data viewpoint does not have the function of complementing the missing part as described above, and the figure viewpoint also shows the result shown in FIG. 11 (d). As such, the total area of the rectangle does not increase.

  In the example of FIG. 11 (d), the area of the rectangle in which all the data are aligned in the row or column direction does not increase, but as shown in FIG. 11 (e), by acquiring possessed data D5, A nearly rectangular shape close to a rectangular shape in which all data are aligned in the column direction is being completed, and such a substantially rectangular shape is also worthy from the viewpoint of obtaining various information as described above.

  Therefore, even if not all the data is prepared in the row or column direction, if the predetermined ratio or number (for example, 1 or 2) is insufficient with respect to the number of data in the matrix direction, all You may make it recognize the value equivalent to the case where it has prepared.

  Alternatively, a predetermined threshold value is set for the rectangular or substantially rectangular area, and the total area of the rectangular shape or the like that has exceeded the threshold value for the first time after data acquisition is evaluated as the value of the acquired data. May be.

  For example, when the data group possessed by the subject N1 is as shown in FIGS. 12 (a) and 12 (b), when the missing portion is complemented by the possessed data D6 of the other subject, in any case The rectangular shape corresponding to 5 rows and 1 column increases from one to two.

  However, even if the total area of the rectangle is the same, the information obtained from a single large rectangle is often more valuable than the information obtained from a plurality of discrete small rectangles. If the area threshold of the shape is larger than the equivalent of 5 rows and 1 column, for example, and smaller than the equivalent of 5 rows and 2 columns, only data that can contribute to the completion of a larger rectangular shape as shown in FIG. Can be evaluated.

  As described above, according to the analysis results of the present inventors, the value obtained by each subject from the other subject and the total area of the rectangle included in the geometric figure complemented by the obtained data are between Correlation was recognized, and it was found that the acquired data that can increase the total area of the rectangle from the graphic viewpoint tends to be more valuable from the data viewpoint.

  Therefore, in the present invention, the value of the acquired data is quantitatively evaluated using the total area of the rectangular shape included in the geometric figure complemented with the acquired data as an index. In the present embodiment, the total area of the rectangular shape is represented by the sum of the lengths of line segments in contact with geometric figures imitating each data and the area of complementary data, as will be described in detail later.

  FIG. 13 is a block diagram showing a configuration of a main part of a data exchange support apparatus according to the second embodiment of the present invention, where the same reference numerals as those described above represent the same or equivalent parts.

  In the possession data evaluation unit 20, the geometric figure forming unit 201 virtually stores the data possessed by the subject N for each subject N based on the row elements and the column elements as shown in FIGS. As shown in FIGS. 10 and 11, the data is converted into a geometric figure by assuming that each data is a rectangular unit area.

  The contact line segment length calculation unit 202 combines the data Ds geometric figure possessed by one subject Ni with the data Dget geometric figure obtained by the one subject Ni from the other subject Nj. Calculate the length of the line where Ds and Dget touch.

  Hereinafter, a specific method for calculating the contact line segment length will be described. Here, in order to make the explanation easy to understand, assuming that the lengths of the sides of the rectangle representing each data are all “1”, in the example shown in FIG. 11A, the possessed data Ds of the subject N1 is acquired. Since the data D2 is in contact with four sides, the length of the contact line is “4”.

  Further, in the example shown in FIG. 11B, the possessed data Ds of the subject N1 and the acquired data D3 are in contact with each other on three sides, so that the contact line segment length is “3”. Furthermore, in the example shown in FIG. 11 (c), the possessed data Ds of the subject N1 and the acquired data D4 are in contact with each other on four sides, so that the contact line segment length is “4”. On the other hand, in the example shown in FIG. 11 (d), the possessed data Ds of the subject N1 and the acquired data D5 are not in line contact, so the contact line segment length is “0”.

  Thus, in this embodiment, the value of the data acquired by each subject is based on the length of the contact line segment between the data geometric figure possessed by the subject obtaining data and the geometric figure of the acquired data. The longer the length, the higher the rating.

  However, data generally has a quantitative value proportional to the data size. Therefore, as shown in FIG. 14, even if the acquired data has the same contact line length for a subject, the acquired data Dget2 [(b)] having a large data size is higher than the small acquired data Dget1. Need to be evaluated.

  Therefore, in the present embodiment, the data size detection unit 203 is provided to reflect the data amount in the data value, and the data value calculation unit 204 determines the value of the data that each subject acquires from the other subject as the length of the contact line segment. It was calculated as the sum of the value of dependence and the value of size dependence.

  Thus, for example, when considering data exchange among three subjects N1, N2, and N3, even if the amount of data that the subject N1 obtains from N2 is “20”, the value that depends on the contact line length If “80”, the real value of the acquired data is “100”.

  Similarly, if the value that the subject N3 obtains from N1 is “50” depending on the amount of data and the value that is dependent on the contact line length is “50”, the actual value of the obtained data is “100”. . Furthermore, even if the data amount dependent value of the data acquired by N2 from N3 is “80”, if the value dependent on the contact line length is “20”, the real value of the acquired data is “100”. It becomes. As a result, since the values of the acquired data are equivalent among the three entities N1, N2, and N3, an equivalent exchange of data can be established.

  Next, a specific method for obtaining the value of data as the sum of the size-dependent value and the contact line length-dependent value will be described.

  If the data size (area) is x and the contact line segment length is y, the data value V is the sum of the size-dependent value f (x) and the contact line length-dependent value g (y). It can be expressed by the following equation (1) using the increasing functions f and g.

  At this time, if a sigmoid function is adopted as the function f, the size-dependent value f (x) can be expressed by the following equation (2).

  If a monotonically increasing function is employed as the function g, the contact line length dependent value g (y) can be expressed by the following equation (3).

  Here, the geometrical figures of the possessed data Ds and the acquired data Dget of the subject N are as shown in FIG. 15, and the coefficients of the above equations (2) and (3) are k = 200, c = 0.05, and a = 3. If there is, the data amount dependent value f (x) of the acquired data Dget is f (100) = 100, and the contact line segment length dependent value g (y) is g (35) = 105.

  If the two-dimensional shape of the possessed data Ds and the acquired data Dget of the subject N is as shown in FIG. 16, the data amount dependent value f (x) of the acquired data Dget is f (150) = 100, and the length of the contact line segment The dependency value g (y) is g (10) = 30.

  By the way, when the relative positional relationship between the geometrical figures of the possessed data Ds and the acquired data Dget of the subject N is as shown in FIG. 17, the acquired data Dget is compared with (a) in FIG. Even if the size is the same, the length of the contact line lengthens, so the value of the acquired data Dget is highly evaluated. However, when there is no restriction such as time series on the arrangement order of the row elements and the column elements, that is, as described with reference to FIGS. It is desirable to be evaluated.

  Therefore, in the present embodiment, prior to the value calculation, the acquired data having substantially the same value are evaluated equally by optimizing the arrangement of the possessed data Ds. The optimization can be realized by rearranging data in ascending or descending order from one side of a matrix (here, column) as shown in FIG. 18, for example. This is synonymous with minimizing the circumference of the geometric figure of the possessed data Ds.

  However, in the example shown in FIG. 17, as shown in FIG. 19, the perimeter is the same even if the array (in this case, the row) of possessed data Ds is rearranged. Therefore, in such a case, the contact line segment length with the acquired data Dget may be obtained each time while changing the matrix arrangement of the possessed data Ds, and the maximum contact line segment length may be optimized. Alternatively, the acquired data Dget and possessed data Ds may be virtually added to form a single data, and the circumference may be minimized.

  According to the present embodiment, when evaluating the value of acquired data by paying attention to the relative relationship with possessed data, the relative relationship of each data that is difficult to quantify is quantified by graphing each data and acquired. Since the value of the data is evaluated based on the quantitative relationship between the figures simulating each data, the value of the data can be evaluated quantitatively.

  Further, according to the present embodiment, since the value of data is evaluated based on the contact line length-dependent value and the size-dependent value of the graphed data, the contact line length and size Different data can be exchanged, and the object and range of data exchange are expanded to enable active data exchange.

  In the above embodiment, the description has been made assuming that the rectangular shapes when the data are converted into geometric figures are the same shape and the same size. However, the present invention is not limited to this, and when the geometrical figures are formed. Each data may be given a desired weight by changing the shape or size of the data.

  For example, data is more valuable as its granularity is finer (more specific on the concept dictionary). Therefore, for example, in the case of address data, the data described as “Fujimino City in Saitama Prefecture” is more valuable than the data of “Saitama Prefecture” alone. Therefore, the smaller the granularity data, the larger the rectangular shape size when the geometric figure is formed. At this time, the rectangular shape size may be determined by the distance from the root node in the concept dictionary.

  DESCRIPTION OF SYMBOLS 1 ... Data exchange assistance apparatus, 10 ... Data management part, 20 ... Owned data evaluation part, 30 ... Data exchange order determination means, 40 ... Data exchange assistance part, 201 ... Geometric figure formation part, 202 ... Contact line segment length calculation part , 203 ... Data size detection unit, 204 ... Data value calculation unit

Claims (12)

In a data exchange support device that supports the equivalent exchange of data between a plurality of entities possessing data,
Means for evaluating the value of data possessed by each entity as the value for the other entity based on other data possessed by the other entity in relation to the data ;
A shared subjecting means that combines subjects who can exchange possession data with each other and can share each possessed data;
A main body capable of equivalently exchanging possessed data with the shared main body to be combined with the common main body to share each possessed data, and a means for repeating this,
A data exchange support apparatus for supporting data exchange between a sharing entity and a sharing entity that can exchange data possessed with the sharing entity. The means for evaluating the value of the data, when the data possessed by each subject can be expressed by a matrix in which the data of the feature value for each target is arranged in a matrix, the row possessing at least one data between the subjects and The data exchange support apparatus according to claim 1 , wherein data lacking in a column is evaluated as equivalent. The means for evaluating the value of the data is that the data possessed by each subject can be represented by a matrix in which the feature value data is arranged in a matrix for each target. sex and data exchange support apparatus according to claim 1 or 2 distance on the characteristic amounts of the concept dictionary and wherein appreciate closer. The data possessed by each subject can be expressed in a matrix in which the feature value data is arranged in a matrix for each target,
The means for evaluating the value of the data is:
Means for converting the data held by each subject into a geometric figure based on the row and column elements of the data;
And means for calculating the value of the acquired data based on a relative positional relationship between a geometric figure of data possessed by each subject and a geometric figure of data obtained from another subject. 2. The data exchange support device according to 1. Means for calculating a total area of a rectangular shape included in a geometric figure complemented by a geometric figure of data acquired from the other principal with a geometric figure of data possessed by each of the principals ;
5. The data exchange support apparatus according to claim 4 , wherein the value of the acquired data is calculated so as to increase as the total area increases. Calculates the total area of the rectangle whose area exceeds the predetermined threshold due to the complementation of the geometric figure of the data possessed by each principal with the geometric figure of the data acquired from the other principal Further comprising means for
5. The data exchange support apparatus according to claim 4 , wherein the value of the acquired data is calculated so as to increase as the total area increases. The data exchange support device according to claim 5 or 6 , wherein the rectangular shape further includes a substantially rectangular shape that becomes a rectangular shape by complementing acquired data less than a predetermined threshold value. Means for calculating a contact line segment length between the geometric figure of data possessed by each of the principals and the geometric figure of data obtained from the other principals ;
The data exchange support according to any one of claims 4 to 7 , wherein the value of the acquired data is calculated based on a contact line segment length dependent value that increases as the contact line segment length increases. apparatus. Means for calculating a data size of data acquired from the other main body ;
9. The data exchange support according to claim 8 , wherein the value of the acquired data is calculated based on a sum of a size-dependent value that increases as the data size increases and a value that depends on the length of the contact line segment. apparatus. Further comprising means for optimizing data possessed by each of the entities,
The data according to any one of claims 4 to 9 , wherein the optimization means replaces at least one of rows and columns so that an outer peripheral length of a two-dimensional image of the possessed data is minimized. Exchange support device. Further comprising means for optimizing data possessed by each of the entities,
The optimization means replaces at least one of rows and columns of the possessed data so that a contact line segment between the geometric shape of the possessed data and the geometric shape of the acquired data is maximized. A data exchange support device according to any one of claims 4 to 9 . In a data exchange support method for supporting equivalent exchange of data between a plurality of entities possessing data by a computer ,
The data possessed by each entity is managed via the network, and the value of each data is evaluated as the value for the other entity based on the other data possessed by the other entity in relation to the data. Procedure and
Based on the data each principal possessed, the procedure for storing shared principal and to which share the possession data by combining the principal each other can equivalent exchange possession data to each other,
Each entity and based on the data sharing entity owned stores entity capable equivalent exchange possession data to the shared main and cross as a shared entity that can share the possession data linked with the shared mainly repeating this Including
A data exchange support method for supporting data exchange between an entity capable of equivalently exchanging possessed data with the shared entity and the shared entity.