Utility model generation device, utility model generation method, and program

Abstract

To provide a technique that can improve an efficiency of generating a utility model.SOLUTION: A utility model generation apparatus includes a utility model generation unit that generates a utility model based on history data including one or more conditions regarding a past negotiation, a presentation control unit that controls to present a user with information on a utility value based on the utility value output from the utility model corresponding to evaluation data, and a utility model update unit that updates the utility model to obtain an updated utility model based on an input of an update instruction of the utility model from the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a utility model generating device, a utility model generating method, and a program. [Background technology]

[0002] In recent years, various technologies related to automated negotiation systems have become known. In an automated negotiation system, a user negotiating inputs a function or quantitative value called a "utility function" that quantitatively expresses their preferences or values ​​into the system. In an automated negotiation system, an automated negotiation agent performs calculations according to this utility function and searches for an agreeable conclusion.

[0003] For example, it has been disclosed that a user of an automated negotiation system can derive a utility function by using an objective function or parameters in the system that the user manages (see, for example, Patent Document 1). A utility function is a model (hereinafter also referred to as a "utility model") that quantitatively expresses the preferences or values ​​of people, things, companies, organizations, etc. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2018 / 070419 Summary of the Invention [Problem to be solved by the invention]

[0005] However, it would be desirable to provide techniques that can improve the efficiency of generating utility models. [Means for solving the problem]

[0006] In order to solve the above problem, according to one aspect of the present invention, there is provided a utility model generation unit that generates a utility model based on history data including one or more conditions related to past negotiations, a presentation control unit that controls presentation of information related to utility values ​​to a user based on utility values ​​output from the utility model corresponding to evaluation data, and a presentation control unit that controls presentation of information related to the utility values ​​to a user when an instruction to update the utility model is input from the user. In this case, a correction instruction for the utility model input by the user is and a utility model update unit that updates the utility model based on the updated utility model to obtain an updated utility model. the evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions, and the utility model update unit updates the utility model so that the weight of the utility value corresponding to the selection item is changed according to the weight of the utility value for an item other than the selection item; A utility model generator is provided.

[0007] The utility model update unit may update the utility model and obtain an updated utility model every time the update instruction is input.

[0009] before The utility model update unit may update the utility model so that a weight of a utility value corresponding to the selected item is greater than a weight of a utility value for an item other than the selected item.

[0010] The correction instruction may include a correction value for the utility value, and the utility model update unit may update the utility model so that the utility value output from the utility model corresponding to the evaluation data is changed to the correction value.

[0011] There may be multiple evaluation data, the correction instruction may include correction information for the magnitude relationship between utility values ​​corresponding to each of the multiple evaluation data, and the utility model update unit may update the utility model so that the magnitude relationship between utility values ​​output from the utility model corresponding to the evaluation data is changed in accordance with the correction information.

[0012] The presentation control unit may control presentation of the utility value as information related to the utility value.

[0013] The presentation control unit may determine whether the utility value exceeds a threshold value, obtain a determination result, and control presentation of the determination result as information related to the utility value.

[0014] The utility model generation device may include an automatic negotiation unit that performs automatic negotiation based on the updated utility model.

[0015] The utility model update unit may update the updated utility model again based on the fact that the evaluation result of the automated negotiation does not satisfy a predetermined condition.

[0016] The utility model update unit may update the updated utility model again unless an evaluation result of the result of the automated negotiation based on the updated utility model satisfies the predetermined condition.

[0017] The evaluation result may be an approval or rejection of the result of the automatic negotiation input by the user, and the predetermined condition may be a condition that the evaluation result is approval of the result of the automatic negotiation.

[0018] There may be multiple results of the automatic negotiation, the evaluation result may be approval or rejection of each of the multiple results of the automatic negotiation input by the user, and the specified condition may be a condition that the ratio of approvals to rejections is greater than a specified ratio.

[0019] According to another aspect of the present invention, a utility model is generated based on history data including one or more conditions related to past negotiations, and based on utility values ​​output from the utility model corresponding to evaluation data, presentation of information regarding the utility values ​​to a user is controlled. In this case, a correction instruction for the utility model input by the user is and updating the utility model based on the result to obtain an updated utility model. The evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions, and obtaining the updated utility model includes updating the utility model so that weights of utility values ​​corresponding to the selection items are changed in accordance with weights of utility values ​​for items other than the selection items. A method for generating a utility model is provided.

[0020] According to another aspect of the present invention, a computer includes a utility model generation unit that generates a utility model based on history data including one or more conditions related to past negotiations, a presentation control unit that controls presentation of information related to utility values ​​to a user based on utility values ​​output from the utility model corresponding to evaluation data, and a presentation control unit that controls presentation of information related to the utility values ​​to a user when an instruction to update the utility model is input from the user. In this case, a correction instruction for the utility model input by the user is and a utility model update unit that updates the utility model based on the updated utility model to obtain an updated utility model. the evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions, and the utility model update unit updates the utility model so that the weight of the utility value corresponding to the selection item is changed according to the weight of the utility value for an item other than the selection item; A program is provided that functions as a utility model generating device. [Effects of the Invention]

[0021] As described above, the present invention provides a technique that can improve the efficiency of generating utility models. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a diagram illustrating an example of the functional configuration of a utility model generation device according to a first embodiment of the present invention. [Figure 2] 10 is a flowchart showing an example of the operation of the utility model generation device according to the embodiment. [Figure 3] 10 is a flowchart showing an example of the operation of the utility model generation device according to the embodiment. [Figure 4] FIG. 1 illustrates an example of input and output data for a utility model. [Figure 5] FIG. 10 is a diagram illustrating an example of a utility value presentation screen. [Figure 6] FIG. 10 is a diagram showing an example of a correction instruction input screen. [Figure 7] FIG. 10 is a diagram showing an example of a utility value presentation screen corresponding to an updated utility model. [Figure 8] FIG. 10 is a diagram illustrating an example of the configuration of an automated negotiation system according to a second embodiment of the present invention. [Figure 9] FIG. 2 is a block diagram illustrating an example of a functional configuration of a utility model generation device according to the embodiment. [Figure 10] FIG. 2 is a diagram illustrating an example of a functional configuration of a terminal device according to the embodiment. [Figure 11] FIG. 1 is a diagram illustrating a hardware configuration of an information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant explanations will be omitted.

[0024] Furthermore, in this specification and drawings, multiple components having substantially the same functional configuration may be distinguished by adding different numbers after the same reference numeral. However, if there is no particular need to distinguish between multiple components having substantially the same functional configuration, only the same reference numeral will be used. Furthermore, similar components in different embodiments may be distinguished by adding different letters after the same reference numeral. However, if there is no particular need to distinguish between similar components in different embodiments, only the same reference numeral will be used.

[0025] <0. Overview> First, an outline of the embodiment of the present invention will be described.

[0026] In the following, matching users together will simply be referred to as "matching."

[0027] In an embodiment of the present invention, examples of users (requesters) to be combined are described as a user who wants to request collection and delivery of luggage (vehicle requester) and a user who wants to collect and deliver luggage (freight requester). An example of a vehicle requester is a shipper. An example of a freight requester is a logistics company (for example, a transportation company). However, users to be combined are not limited to vehicle requesters and freight requesters.

[0028] Furthermore, in the embodiments of the present invention, delivery is used as a term meaning the transport of luggage. The luggage is also expressed as cargo, and the type of luggage is not particularly limited as long as it is an object that can be transported by a means of transportation. Here, the embodiments of the present invention mainly assume that luggage is transported by truck. However, as the means of transportation for transporting luggage, transportation means other than trucks (for example, railroad cars, ships, aircraft, drones, etc.) may be used instead of trucks.

[0029] In recent years, various technologies related to automated negotiation systems have become known. In the technology described in Patent Document 1, a user of an automated negotiation system or the like must design a utility function before negotiation and input the designed utility function into the system. A utility function is a model (hereinafter also referred to as a "utility model") that quantitatively expresses the preferences or values ​​of a person, object, company, organization, etc. The value returned by a utility function under certain conditions is called a utility value. Here, an example of designing a utility function is described.

[0030] For example, user A designs a simple utility function to be used in automated negotiations for two points: {delivery date and time, transportation fee}. Let us assume that user A's utility value f(t, p) for the pair of delivery date and time t and transportation fee p is defined as shown in the following formula (1).

[0031] f(t,p)=α×g(t)+β×h(p)...Equation (1)

[0032] Here, g(t) is the delivery date utility value calculated for delivery date and time t, and h(p) is the transportation fee utility value calculated for transportation fee p. Furthermore, α and β are weights expressed by real constants. From equation (1), it can be seen that user A's utility function f is expressed as a linear combination of the delivery date utility value g(t) and the transportation fee utility value h(p). As mentioned above, defining the utility function as a specific mathematical formula is referred to here as "designing the utility function."

[0033] Recently, there has been a movement to implement automated negotiation systems as a means of coordination or negotiation between individuals or companies. However, it is not easy for users of such systems to appropriately quantify their true preferences or values ​​and to design a utility function that can be used in the automated negotiation system based on those values. There are three main reasons for this.

[0034] The first reason is that it is difficult to quantitatively evaluate true preferences or values, which are determined by human experience or intuition.

[0035] For example, in the trucking industry, in cargo and truck search services that match cargo with available trucks, logistics companies and shippers often directly negotiate collection and delivery dates and times, transportation conditions, transportation fees, etc. In these cases, logistics company personnel typically propose appropriate conditions or fees by comprehensively considering many parameters, such as other transportation requests, the likelihood of future requests, truck availability, route, traffic conditions, vehicle equipment, and driver skills. It is not easy for personnel themselves to design a quantitative utility function based on preferences or values ​​that are intricately dependent on such a large number of parameters or constraints.

[0036] Therefore, we propose a technology that can design utility functions without requiring users to quantitatively evaluate their true preferences or values, thereby improving the efficiency of utility function design.

[0037] The second reason is that mathematical expertise is required to design utility functions.

[0038] Even when designing the above simple utility function, it is necessary to consider whether f should be a linear or quadratic combination of the delivery date utility value and the transportation fee utility value, how to determine the two term coefficients α and β, and how to calculate the delivery date utility value g(t) for each t. Furthermore, if the utility function is used in automated negotiations with many negotiation points, if the utility value depends on many parameters other than the negotiation points, or if there are dependencies between those parameters, it will be necessary to design a more complex utility function.

[0039] To date, no automated negotiation system or service has been proposed that has the functionality to assist in the design of such utility functions. Therefore, users of the automated negotiation system must rely on personnel with mathematical expertise to use the automated negotiation system.

[0040] Therefore, we propose a technology that enables users to design utility functions even if they do not have mathematical expertise, thereby improving the efficiency of utility function design.

[0041] The third reason is the difficulty of assessing the validity of the designed utility function. For example, a utility function is valid if there is no significant discrepancy or gap between true preferences or values ​​and the utility function.

[0042] However, no method has been proposed to date for evaluating the validity of a utility function. Furthermore, users' preferences or values ​​typically change daily. Therefore, in addition to designing a utility function once and evaluating its validity when first participating in an automated negotiation system, a mechanism is needed for evaluating the validity of the utility function at any time while the automated negotiation system is being used, and for improving the utility function. However, no system or service has been proposed that allows users to perform such tasks themselves.

[0043] Therefore, we propose a technology that allows users to evaluate the validity of a designed utility function and update the utility function based on the user's evaluation, thereby narrowing the gap between true preferences or values ​​and the utility function.

[0044] Furthermore, we propose a technology that allows users to evaluate the results of automated negotiations using utility functions and update the utility functions based on the user evaluations. This makes it possible to provide an appropriate utility function that matches the user's preferences or values, which change daily.

[0045] In the following description, the utility function may be expressed by a utility model, i.e., the design of the utility function may be realized by generating a utility model.

[0046] The outline of the embodiment of the present invention has been described above.

[0047] <1. First embodiment> Next, the first embodiment of the present invention will be described in detail.

[0048] In the first embodiment of the present invention, it is mainly assumed that a logistics company, as an example of a user, uses the utility model generation device. Therefore, a negotiation agent (shipper agent) in which a utility model of a shipper, who is the logistics company's negotiating partner, is set, and shipper request information are prepared in advance in the utility model generation device. Then, the utility model generation device generates a utility model of the logistics company.

[0049] At this time, as will be explained below, automatic negotiation is performed by the utility model generation device based on a pre-prepared shipper agent and shipper request information, and a negotiation agent (logistics company agent) and logistics company request information to which the logistics company utility model generated by the utility model generation device is set. For example, the negotiation agent is an electronic agent that operates within the system and can be realized by, for example, AI (Artificial Intelligence).

[0050] The utility model generation device may also be used by a shipper, another example of a user. In such a case, a negotiation agent (logistics company agent) in which a utility model of a logistics company with which the shipper is negotiating is set and logistics company request information may be prepared in advance within the utility model generation device. Then, the utility model generation device may generate a utility model for the shipper.

[0051] At this time, automatic negotiation may be performed by the utility model generation device based on a logistics company agent and logistics company request information prepared in advance, and a negotiation agent (shipper agent) and shipper request information to which the shipper utility model generated by the utility model generation device is set. Note that, although the utility model generation device is configured to prepare the negotiating partner's utility model and request information in advance in the above, it may also be configured to generate these in accordance with rules given in advance before the start of automatic negotiation.

[0052] (1-1. Example of functional configuration of utility model generation device) Next, an example of the functional configuration of the utility model generation device 11 according to the first embodiment of the present invention will be described. Fig. 1 is a diagram showing an example of the functional configuration of the utility model generation device 11 according to the first embodiment of the present invention. The utility model generation device 11 can be realized by a computer.

[0053] 1, a utility model generation device 11 according to the first embodiment of the present invention includes an input unit 111. The utility model generation device 11 according to the first embodiment of the present invention also includes a UI (User Interface) unit 114, a control unit (not shown), and a storage unit (not shown). The control unit (not shown) includes the input unit 111, a utility model generation unit 112, a utility model evaluation unit 113, an automatic negotiation unit 116, and an evaluation data generation unit 118. The storage unit (not shown) includes a utility model storage unit 115 and a negotiation performance data storage unit 117.

[0054] The UI unit 114 includes an input interface that accepts operations input by the logistics company and an output interface that outputs various information to the logistics company. For example, the input interface may be configured with a mouse, keyboard, or touch panel, but the specific form of the input interface is not limited. The output interface may be configured with a display, but the specific form of the output interface is not limited.

[0055] The control unit (not shown) includes a CPU (Central Processing Unit) and the like, and its functions can be realized by the CPU expanding a program stored in a storage unit (not shown) into RAM (Random Access Memory) and executing it. In this case, a computer-readable recording medium on which the program is recorded can also be provided. Alternatively, the control unit (not shown) can be configured with dedicated hardware or a combination of multiple pieces of hardware.

[0056] The storage unit (not shown) is a storage device capable of storing programs and data for operating the control unit (not shown). For example, the storage device may be a non-volatile storage device. The storage unit (not shown) can also temporarily store various data required in the operation of the control unit (not shown).

[0057] (Negotiation performance data storage unit 117) The negotiation performance data storage unit 117 stores negotiation performance data of a logistics company, which corresponds to an example of history data related to the logistics company's past negotiations. For example, the logistics company's past negotiating partner may be a shipper. The history data includes one or more conditions related to the logistics company's past negotiations. Here, a condition is a combination of items and data desired from a negotiating partner. Note that other history data may be used instead of the negotiation performance data. For example, as will be explained later, transaction performance data may be used as another example of history data related to the logistics company's past negotiations.

[0058] (input unit 111) The input unit 111 acquires negotiation performance data from the negotiation performance data storage unit 117. The input unit 111 outputs the negotiation performance data from the negotiation performance data storage unit 117 to the utility model generation unit 112.

[0059] (Utility model generation unit 112) The utility model generation unit 112 generates a utility model corresponding to the logistics company based on the negotiation performance data output from the input unit 111. As an example, the utility model generation unit 112 may generate a utility model corresponding to the logistics company by performing mathematical optimization or machine learning processing on the negotiation performance data output from the input unit 111.

[0060] (Utility model storage unit 115) The utility model storage unit 115 stores the utility model corresponding to the logistics company generated by the utility model generation unit 112.

[0061] (Evaluation data generation unit 118) The evaluation data generation unit 118 generates one or more evaluation data for a utility model corresponding to a logistics company. Each of the one or more evaluation data may include one or more conditions. Typically, the evaluation data and the negotiation performance data have the same condition items. In such a case, the evaluation data is input into the utility model created by the utility model generation unit 112 based on the negotiation performance data, and a utility value corresponding to the evaluation data is output from the utility model.

[0062] The evaluation data may not be generated by the evaluation data generation unit 118, but may be input to the input unit 111 or may be input from a logistics company via the UI unit 114. In such a case, the utility model generation device 11 does not need to include the evaluation data generation unit 118.

[0063] (Utility model evaluation unit 113) The utility model evaluation unit 113 acquires a utility model corresponding to the logistics company from the utility model storage unit 115. The utility model evaluation unit 113 also acquires evaluation data. Then, based on the utility value output from the utility model corresponding to the logistics company in response to the evaluation data, the utility model evaluation unit 113 outputs information related to the utility value to the UI unit 114, thereby controlling the presentation of information related to the utility value to the logistics company. The utility model evaluation unit 113 can function as an example of a presentation control unit.

[0064] Furthermore, the utility model evaluation unit 113 updates the utility model corresponding to the logistics company to obtain the updated utility model based on a command to update the utility model input from the logistics company via the UI unit 114. The utility model evaluation unit 113 may update the utility model corresponding to the logistics company to obtain the updated utility model every time a command to update the utility model is input from the logistics company via the UI unit 114. The utility model evaluation unit 113 may function as an example of a utility model update unit.

[0065] Furthermore, utility model evaluation unit 113 determines whether the evaluation result of the automatic negotiation satisfies a predetermined condition when automatic negotiation is performed by automatic negotiation unit 116 based on the utility model updated by utility model evaluation unit 113. If the evaluation result of the automatic negotiation does not satisfy the predetermined condition, utility model evaluation unit 113 re-updates the updated utility model stored in utility model storage unit 115.

[0066] In addition, the utility model evaluation unit 113 may obtain another update to the updated utility model stored in the utility model storage unit 115 unless the evaluation result of the automatic negotiation result based on the utility model after being updated again satisfies a predetermined condition.

[0067] (Automatic Negotiation Unit 116) The automatic negotiation unit 116 performs automatic negotiation based on the utility model updated by the utility model evaluation unit 113. More specifically, based on an instruction to start automatic negotiation input from the logistics company via the UI unit 114, the automatic negotiation unit 116 performs automatic negotiation based on the negotiation agent (logistics company-side agent) and logistics company-side request information in which the updated utility model corresponding to the logistics company is set, and the shipper-side agent and shipper-side request information prepared in advance.

[0068] An example of the functional configuration of the utility model generation device 11 according to the first embodiment of the present invention has been described above.

[0069] (1-2. Example of operation) Next, an example of the operation of the utility model generation device 11 according to the first embodiment of the present invention will be described. FIGS. 2 and 3 are flowcharts showing an example of the operation of the utility model generation device 11 according to the first embodiment of the present invention. Note that the example of the operation shown in FIGS. 2 and 3 is merely one example of the operation of the utility model generation device 11 according to the first embodiment of the present invention. Therefore, the operation of the utility model generation device 11 according to the first embodiment of the present invention is not limited to the example of the operation shown in FIGS. 2 and 3.

[0070] (S111) First, when the logistics company selects desired negotiation performance data from one or more negotiation performance data stored in the negotiation performance data storage unit 117, the negotiation performance data selected by the logistics company is input to the input unit 111. Here, as an example, it is assumed that the logistics company's past negotiation performance data shown in Table (1) below is input to the input unit 111.

[0071] [Table 1]

[0072] Referring to Table (1), the combination of the item "Negotiation Date and Time" and the corresponding data "2021 / 4 / 1" is included as a condition. Other items included in the negotiation performance data for each condition include "Pickup Date and Time," "Delivery Date and Time," "Pickup Point," "Delivery Point," "Item," "Temperature Range," "Cargo Weight," and "Fee." However, the number of conditions included in the negotiation performance data is not limited.

[0073] Furthermore, referring to Table (1), the negotiation performance data includes "negotiation results" corresponding to these conditions. The "negotiation results" are set to "agreement" or "disagreement." However, "agreement" may be rephrased as other terms. For example, "agreement" may be rephrased as "order accepted." Similarly, "disagreement" may be rephrased as other terms. For example, "disagreement" may be rephrased as "rejection."

[0074] Although Table (1) shows three specific examples of negotiation performance data, the number of negotiation performance data items may be one or more.

[0075] Furthermore, the "negotiation result" does not have to be included in the negotiation performance data input to the input unit 111. For example, the negotiation performance data input to the input unit 111 may include only one or more conditions. In this case, the "negotiation result" may be input by the logistics company via the UI unit 114.

[0076] Alternatively, it is also conceivable that the "negotiation result" corresponding to the negotiation performance data is determined to be only "agreement." In such cases, the "negotiation result" does not have to be included in the negotiation performance data. Similarly, it is also conceivable that the "negotiation result" corresponding to the negotiation performance data is determined to be only "disagreement." In such cases, the "negotiation result" does not have to be included in the negotiation performance data.

[0077] Furthermore, transaction performance data may be used instead of negotiation performance data. Transaction performance data is data related to transactions that were carried out after negotiations reached an agreement. That is, like the negotiation performance data, the transaction performance data may include multiple conditions, but it does not need to include the negotiation result because the "negotiation result" corresponding to the transaction performance data is always "agreement."

[0078] Constraint conditions for one or more conditions may be input to the input unit 111. Here, it is assumed that one constraint condition is input, but multiple constraint conditions may be input. As an example, it is assumed that the constraint conditions shown in the following formula (2) and table (2) are input for the price p(temp, w) for the temperature range (temp) and cargo weight (w).

[0079] p(temp,w)≧bound(temp,w)...Equation (2)

[0080] [Table 2]

[0081] Here, bound(temp, w) represents the minimum price for cargo in the temperature range (temp) and cargo weight (w). Note that such constraints may be input using a mathematical formula as shown in equation (2). Alternatively, the input unit 111 may present guidance to the logistics company via the UI unit 114 to prompt the logistics company to input information necessary for generating constraints, and generate constraints based on the information input by the logistics company via the UI unit 114.

[0082] (S112) The utility model generation unit 112 generates a utility model corresponding to the logistics company based on the negotiation performance data input to the input unit 111. As an example, the utility model generation unit 112 may generate a utility model corresponding to the logistics company by performing mathematical optimization or machine learning processing on the negotiation performance data output from the input unit 111.

[0083] For example, each condition that corresponds to at least a portion of the negotiation performance data may be used as a variable, and the specific value of the variable may be input to the utility model as input data. For example, if a condition formed by a combination of the item "temperature range" and the data "refrigerated" is used as a variable, the specific value of the variable may be the data "refrigerated." In this case, the utility model may output an evaluation value corresponding to the specific value of the variable as output data.

[0084] Alternatively, a condition obtained by performing preprocessing based on a predetermined rule on each of the conditions included in the negotiation performance data that corresponds to at least a part of the conditions may be treated as a variable.

[0085] Here, it is assumed that the utility model generation unit 112 calculates the required time using route calculation software or the like for the conditions "Pickup point = A" and "Delivery point = B" included in the negotiation performance data. In this case, as shown in Table (3) below, the utility model generation unit 112 may add a combination of the item "required time" and the data corresponding to that item "required time" to the negotiation performance data as a new condition.

[0086] [Table 3]

[0087] 4 is a diagram showing examples of input data and output data for a utility model. As shown in FIG. 4, data corresponding to each of the items "required time," "temperature range," "cargo weight," and "transportation fee" included in the negotiation performance data may be input to the utility model as input data. Then, a utility value corresponding to all of these items may be output from the utility model as output data.

[0088] Although not shown, data (utility values) corresponding to the items "required time," "temperature range," "cargo weight," and "transportation fee" may also be output as output data from the utility model. In this case, the sum of the results of multiplying the utility values ​​of each item by the weights may be the utility value corresponding to all the items.

[0089] Returning to FIG. 2, the description will continue. The form of the utility value output from the utility model may be a class related to the possibility of agreement (for example, "agreement" or "disagreement"). Alternatively, the utility value output from the utility model may be a real value. Here, the method for generating the utility model is not limited to a specific method.

[0090] As an example, suppose that negotiation performance data is input and a utility model is generated that outputs a class related to the likelihood of agreement corresponding to the negotiation performance data. In such a case, a class classification method can be adopted as a method for generating the utility model, in which the "negotiation result" included in the negotiation performance data is used as the objective variable and other conditions included in the negotiation performance data are used as explanatory variables.

[0091] As another example, a utility model may be generated that receives input negotiation performance data and outputs real values ​​corresponding to the negotiation performance data. In such a case, a method for generating the utility model may be adopted in which a probability distribution estimated by a classification model with cross-entropy error as a loss function, or a discriminant score in logistic regression, is used to calculate the probability or evaluation value that the negotiation performance data will be classified as agreement, and the utility model is generated using the probability or evaluation value.

[0092] For example, the probability or the evaluation value that the negotiation performance data will be classified as "agreement" may be used as is as an output from the utility model. Alternatively, the probability or the evaluation value that the negotiation performance data will be classified as "agreement" may not be used as is as an output from the utility model. For example, the probability or the evaluation value that the negotiation performance data will be classified as "agreement" may be converted by a pre-specified method taking into account behavior in automated negotiations and then used as an output from the utility model.

[0093] As an example, we will explain an example in which the conditions corresponding to "transport time," "temperature range," "cargo weight," and "transportation fee" contained in the updated negotiation performance data shown in Table (3) are used as explanatory variables, and "negotiation result" is used as the objective variable to generate a logistic regression model M. Here, the discriminant score of model M represents the probability that the data for the explanatory variables will be agreed upon. However, we assume that a rule is given in advance that "when the discriminant score of model M is less than 0.3, the utility value is -∞."

[0094] In such a case, the utility model pref is generated as shown in the following equation (3).

[0095]

number

[0096] Table (4) below shows an example of the agreement probability estimated by model M and the output of the utility model pref for each combination of conditions corresponding to "travel time," "temperature range," "cargo weight," and "transportation fee."

[0097] [Table 4]

[0098] If there is a bias in the distribution of "cargo weight" or "collection date and time" in the negotiation performance data input to the input unit 111, the utility model generation unit 112 may display a screen prompting the logistics company to input additional data via the UI unit 114. Then, if additional data is input from the logistics company via the UI unit 114, the added data may be added to the negotiation performance data input to the input unit 111.

[0099] (S113) The utility model storage unit 115 stores the utility model pref corresponding to the logistics company generated by the utility model generation unit 112.

[0100] (S114) The utility model evaluation unit 113 acquires a utility model pref corresponding to the logistics company from the utility model storage unit 115. Then, the utility model evaluation unit 113 evaluates the utility model pref. Here, various methods can be adopted as a method for evaluating the utility model pref. For example, the utility model evaluation unit 113 acquires evaluation data.

[0101] The evaluation data may be input by the logistics company. That is, the utility model evaluation unit 113 may display an input form via the UI unit 114 and have the logistics company input the evaluation data into the input form. As an example, the evaluation data may include one or more conditions, similar to the negotiation performance data. In this case, the items corresponding to the conditions included in the evaluation data may be the same as the items corresponding to the conditions included in the negotiation performance data.

[0102] As an example, assume that the evaluation data shown in the following Table (5) is input. Note that in the example shown in Table (5), the number of evaluation data items is two, but the number of evaluation data items is not limited to two, and may be one, or three or more.

[0103] [Table 5]

[0104] The evaluation data may be input to the input unit 111 in a format such as a CSV file. Alternatively, the evaluation data may be generated by the evaluation data generation unit 118. In this case, the evaluation data may be generated based on the constraints or negotiation performance data input to the input unit 111 using a technique such as equivalence partitioning or boundary value analysis.

[0105] The utility model evaluation unit 113 inputs evaluation data to the utility model pref and obtains a utility value output from the utility model pref corresponding to the evaluation data as an evaluation result. At this time, if the utility model pref is a machine learning model, the utility model evaluation unit 113 may evaluate the fitting accuracy of the utility model pref to the constraint conditions, the fitting accuracy of the utility model pref to the negotiation performance data, the absence of overlearning of the utility model pref, the robustness of the utility model pref, etc.

[0106] (S115) The utility model evaluation unit 113 confirms the validity of the utility model pref. More specifically, the utility model evaluation unit 113 controls the presentation of information about the utility value to the logistics company by outputting information about the utility value to the UI unit 114 based on the utility value. For example, the utility model evaluation unit 113 may cause the UI unit 114 to display the information about the utility value on a screen. At this time, the utility model evaluation unit 113 may also display on a screen, in addition to the information about the utility value, the fitting accuracy of the utility model pref with respect to the constraint conditions, the fitting accuracy of the utility model pref with respect to the negotiation performance data, the absence of overlearning of the utility model pref, the robustness of the utility model pref, and the like.

[0107] The utility model evaluation unit 113 may output the utility value to the UI unit 114, thereby controlling the UI unit 114 to present the utility value to the logistics company as information related to the utility value. Alternatively, the utility model evaluation unit 113 may determine whether the utility value exceeds a threshold value, obtain a determination result (for example, agreement or disagreement), and output the determination result to the UI unit 114, thereby controlling the UI unit 114 to present the determination result to the logistics company as information related to the utility value. Here, the case where the utility value is displayed will mainly be described.

[0108] FIG. 5 is a diagram showing an example of a utility value presentation screen. As shown in FIG. 5, the utility model evaluation unit 113 may display a utility value presentation screen G11. The utility value presentation screen G11 includes the first and second evaluation data and their corresponding utility values. The utility value presentation screen G11 also includes a utility model edit button B11 and a utility model save button B12. The operation of pressing the utility model save button B12 corresponds to an example of an instruction to start automatic negotiation.

[0109] (S116) If the logistics company determines that the utility values ​​included in the utility value presentation screen G11 are appropriate, the logistics company inputs an operation to press the utility model save button B12 into the UI unit 114. When the operation to press the utility model save button B12 is input, the utility model storage unit 115 stores the utility model pref. Then, the automatic negotiation operation (the operation from S118 onwards shown in FIG. 3) is started.

[0110] (S117) If the logistics company determines that the utility values ​​included in the utility value presentation screen G11 are inappropriate, the logistics company inputs an operation to press the utility model edit button B11 into the UI unit 114. When the operation to press the utility model edit button B11 is input, the utility model evaluation unit 113 causes the UI unit 114 to display a correction instruction input screen for inputting correction instructions for the utility model. Here, an example of the correction instruction input screen will be described.

[0111] Fig. 6 is a diagram showing an example of a correction instruction input screen. Referring to Fig. 6, a correction instruction input screen G21 is shown. The correction instruction input screen G21 includes a correction instruction input field T21, an update button B21, and a cancel button B22. The operation of pressing the update button B21 corresponds to an example of an instruction to update the utility model.

[0112] When a correction instruction is input into the correction instruction input field T21 and the update button B21 is pressed, the utility model evaluation unit 113 updates the utility model pref based on the correction instruction input into the correction instruction input field T21, thereby obtaining the updated utility model pref'. Here, various input formats for the correction instruction are possible.

[0113] For example, in the example shown in FIG. 6, the evaluation data includes conditions corresponding to "collection date and time," "delivery date and time," "collection point," "delivery point," "item," "temperature range," "cargo weight," and "fee." When the evaluation data includes multiple conditions, as in this example, the logistics company may select an item with an inappropriate utility value from the items corresponding to each of these multiple conditions as a selection item. As an example, the logistics company may enter the selection item "fee" as an item with an inappropriate utility value in the correction instruction input field T21.

[0114] In this case, the utility model evaluation unit 113 may update the utility model pref so that the weight of the utility value corresponding to the selection item "fee" is greater than the weight of the utility values ​​for the items other than the selection item "fee," namely "collection date and time," "delivery date and time," "collection point," "delivery point," "item," "temperature range," and "cargo weight." Note that only the selection item "fee" may be entered in the correction instruction input field T21, or, as shown in FIG. 6, not only the selection item "fee" but also the evaluation data number "No. 1" may be entered in the correction instruction input field T21. Alternatively, instead of "fee," the reason for the inappropriate utility value, "cheap," may be entered in the correction instruction input field T21, such as "No. 1, cheap."

[0115] Alternatively, the correction instruction may include a correction value for an inappropriate utility value. In this case, the utility model evaluation unit 113 may update the utility model pref so that the utility value output from the utility model pref corresponding to the evaluation data is changed to the correction value. As an example, when the logistics operator determines that the utility value "30" corresponding to the evaluation data of "No. 1" is inappropriate and the correction value "27" is appropriate, the logistics operator may cause "No. 1, 27" to be entered in the correction instruction input field T21.

[0116] Alternatively, the correction instruction may include correction information on the magnitude relationship of the utility values corresponding to each of the plurality of evaluation data. In this case, the utility model evaluation unit 113 may update the utility model pref so that the magnitude relationship of the utility values output from the utility model corresponding to the plurality of evaluation data is changed according to the correction information. As an example, when the logistics operator determines that it is appropriate for the utility value corresponding to the evaluation data of "No. ②" to be greater than the utility value corresponding to the evaluation data of "No. ①", the logistics operator may cause "No. 1 < No. 2" or "No. 1 is less desirable than No. 2" to be entered in the correction instruction input field T21.

[0117] [[ID=⑧]]Alternatively, the logistics operator may cause a plurality of correction instructions to be combined and entered in the correction instruction input field T21. For example, the logistics operator may cause a plurality of instructions such as "No. 1 is less desirable than No. 2 because the fee is low" to be entered in the correction instruction input field T21. When an operation of pressing the cancel button B22 is input, the screen may transition from the correction instruction input screen G21 to the utility value presentation screen G11.

[0118] (Repeating S113 to S115, S117) It should be noted that there seems to be an error in the original text where "No. ②" and "No. ①" are used in without prior definition. I translated them as "No. ②" and "No. ①" as they are. Also, there is an "⑧" in which might be a mistake. I translated it as "⑧" as it is. You may want to double-check these in the original context.Once the utility model pref is updated, steps S113 to S115 are executed based on the updated utility model pref'. That is, in S114, the utility model evaluation unit 113 inputs evaluation data to the updated utility model pref' and obtains, as an evaluation result, a utility value output from the utility model pref' corresponding to the evaluation data. In S115, the utility model evaluation unit 113 checks the validity of the utility model pref'.

[0119] More specifically, based on the utility value output from the utility model pref' corresponding to the evaluation data, the utility model evaluation unit 113 outputs information about the utility value to the UI unit 114, thereby controlling the presentation of information about the utility value to the logistics company. For example, the utility model evaluation unit 113 may output the utility value to the UI unit 114, thereby controlling the UI unit 114 to present the utility value as information about the utility value to the logistics company.

[0120] Figure 7 is a diagram showing an example of a utility value presentation screen corresponding to the updated utility model. Referring to Figure 7, the weight of the utility value corresponding to the item "fee" has become larger than the weights of the utility values ​​for the items other than the selected item "fee": "collection date and time," "delivery date and time," "collection point," "delivery point," "item," "temperature range," and "cargo weight." As a result, the utility value "30" corresponding to evaluation data "No. 1" has been changed to "27," and the utility value "25" corresponding to evaluation data "No. 2" has been changed to "30."

[0121] As long as the logistics company determines that the utility values ​​included in the utility value presentation screen G12 are inappropriate, the logistics company may input an operation to press the utility model edit button B11 into the UI unit 114, input a correction instruction into the correction instruction input field T21 on the correction instruction input screen G21, and press the update button B21. In other words, the utility model evaluation unit 113 may update the utility model every time the operation to press the update button B21 is input.

[0122] On the other hand, if the logistics company determines that the utility values ​​included in the utility value presentation screen G12 are appropriate, the logistics company may input an operation to press the utility model save button B12 into the UI unit 114. When the operation to press the utility model save button B12 is input, the utility model storage unit 115 stores the utility model pref. Then, the automatic negotiation operation (the operation from S118 onwards shown in FIG. 3) is started.

[0123] (Variations of S113 to S115 and S117) Returning to Figure 2, the explanation will continue. As a modification of S113 to S115 and S117, the following method can be applied. That is, the utility model evaluation unit 113 acquires evaluation data, similar to S114. Then, the utility model evaluation unit 113 outputs the evaluation data to the UI unit 114, thereby controlling the presentation of the evaluation data to the logistics company.

[0124] The utility model evaluation unit 113 causes the UI unit 114 to display a screen prompting the input of correct data, such as a utility value corresponding to the evaluation data, a magnitude relationship between the utility values ​​corresponding to each of the multiple evaluation data, and a judgment result (e.g., agreement or disagreement) for the utility value corresponding to the evaluation data. The logistics company inputs such correct data into the UI unit 114.

[0125] The utility model evaluation unit 113 inputs evaluation data to the utility model pref and obtains output data from the utility model pref corresponding to the evaluation data. The utility model evaluation unit 113 then evaluates the utility model pref based on the output data and the correct answer data. For example, the utility model evaluation unit 113 may use the utility value, the magnitude relationship between the utility values, or the error between the judgment result for the utility value and the correct answer data as the evaluation index.

[0126] Alternatively, the evaluation index may include the fitting accuracy of the utility model pref to the constraint conditions, the fitting accuracy of the utility model pref to the negotiation performance data, the absence of overlearning of the utility model pref, the robustness of the utility model pref, etc. The utility model storage unit 115 stores the utility model pref when the evaluation index satisfies a predetermined criterion.

[0127] On the other hand, if the evaluation index does not satisfy the predetermined criteria, the utility model evaluation unit 113 causes the UI unit 114 to display a screen prompting the input of additional data (for example, additional constraints or additional negotiation performance data). When the additional data is input from the logistics company, in S112, a utility model is regenerated based on the additional data, and S113 to S115 are executed based on the regenerated utility model.

[0128] At this time, the utility model evaluation unit 113 may identify additional data that is useful for updating the utility model and display it on the UI unit 114. For example, it is assumed that the training data for the utility model pref contains only a small amount of negotiation history data in which the "temperature range" is "refrigerated," and the fitting accuracy of the utility model pref to the evaluation data in which the "temperature range" is "refrigerated" is low. In such a case, the utility model evaluation unit 113 may display information such as "Please add negotiation history data in which the temperature range is refrigerated."

[0129] (S118) The explanation will be continued with reference to Fig. 3. The automatic negotiation unit 116 acquires the utility model pref from the utility model storage unit 115 or the utility model pref' updated by the utility model evaluation unit 113.

[0130] (S119) The automatic negotiation unit 116 performs automatic negotiation based on the acquired utility model. More specifically, the automatic negotiation unit 116 performs automatic negotiation one or more times based on the negotiation agent (logistics company agent) for which the utility model is set and logistics company request information, and the shipper agent and shipper request information prepared in advance. In this case, the utility model of the shipper agent in the multiple automatic negotiations may be the same or different.

[0131] The logistics company request information may include one or more conditions, similar to the negotiation performance data. The shipper request information may include one or more conditions, similar to the negotiation performance data.

[0132] (S120) The utility model evaluation unit 113 evaluates the utility model. More specifically, the utility model evaluation unit 113 obtains the result of the automatic negotiation from the automatic negotiation unit 116. The result of the automatic negotiation may include the result of negotiation (e.g., agreement or disagreement) on the logistics company's request information. Alternatively, the result of the automatic negotiation may include the content of agreement on the conditions included in the logistics company's request information.

[0133] (S121) The utility model evaluation unit 113 confirms the validity of the utility model. More specifically, the utility model evaluation unit 113 displays the results of the automatic negotiation on the UI unit 114, and also displays a screen on the UI unit 114 prompting input of an evaluation result for the results of the automatic negotiation. Here, the input format of the evaluation result is not limited. For example, the evaluation result may be "approval" or "rejection" for the entire results of the automatic negotiation. Alternatively, if there are multiple results of the automatic negotiation, the evaluation result may be "approval" or "rejection" for each of the multiple results of the automatic negotiation.

[0134] (S122) The utility model storage unit 115 stores the utility model when the evaluation result for the automated negotiation results satisfies a predetermined condition. Various conditions are assumed as the predetermined condition. For example, the predetermined condition may be a condition that the evaluation result is "approval" for the entire automated negotiation results. Alternatively, the predetermined condition may be a condition that the ratio of "approval" automated negotiation results to "rejection" automated negotiation results is greater than a predetermined ratio (e.g., 90%).

[0135] (S123) The utility model evaluation unit 113 updates the utility model if the evaluation result of the automatic negotiation does not satisfy a predetermined condition.

[0136] For example, when a logistics company inputs correction information regarding the superiority / inferiority relationship between multiple pieces of request information via the UI unit 114, the utility model evaluation unit 113 may update the utility model so that the magnitude relationship between the utility values ​​output from the utility model corresponding to each of the multiple pieces of request information is changed according to the superiority / inferiority relationship (so that the utility value of the more superior request information becomes larger).

[0137] Alternatively, if a logistics company inputs an item for which the agreement content is inappropriate via the UI unit 114, the utility model evaluation unit 113 may update the utility model so that the weight of the utility value corresponding to the inappropriate item is greater than the weight of the utility value for the other items.

[0138] (Repeat S118-S121, S123 and S124) When the utility model is updated, steps S118 to S121 are executed based on the updated utility model. If it is determined in step S121 that the utility model is invalid, steps S123 and S124 are executed. That is, utility model evaluation unit 113 updates the utility model unless the evaluation result of the automated negotiation based on the utility model satisfies a predetermined condition.

[0139] An example of the operation of the utility model generation device 11 according to the first embodiment of the present invention has been described above.

[0140] (1-3. Effects) As described above, according to the first embodiment of the present invention, even if a user does not have mathematical expertise, a mathematical optimization model or machine learning model that expresses the user's utility can be generated by inputting existing negotiation performance data. Therefore, it is possible to efficiently design a utility function, which is difficult for users to do by themselves when using an automated negotiation service.

[0141] Furthermore, according to the first embodiment of the present invention, the gap between the utility model and the user's true preferences and values ​​can be reduced by having the user evaluate the output of the generated utility model for the evaluation data, updating the utility model, and comparing it with the user's self-reported utility.

[0142] Furthermore, according to the first embodiment of the present invention, the user himself / herself evaluates the results of the automated negotiation using the generated utility model, and the utility model is updated using the evaluation results, thereby making it possible to provide an appropriate utility model that is adapted to the user's preferences or values, which change daily.

[0143] The first embodiment of the present invention has been described above.

[0144] 2. Second embodiment Next, a second embodiment of the present invention will be described.

[0145] (2-1. System configuration example) First, a configuration example of an automated negotiation system according to a second embodiment of the present invention will be described. Fig. 8 is a diagram showing a configuration example of an automated negotiation system according to the second embodiment of the present invention. As shown in Fig. 8, the automated negotiation system 2 according to the second embodiment of the present invention includes a plurality of terminal devices 20 and a utility model generation device 12. The plurality of terminal devices 20 and the utility model generation device 12 are configured to be able to communicate with each other via a network 50.

[0146] Referring to Figure 8, terminal devices 20-A to 20-D are shown as examples of multiple terminal devices 20. Terminal device 20-A is a terminal device used by logistics company A, and terminal device 20-B is a terminal device used by logistics company B. Terminal device 20-C is a terminal device used by shipper C, and terminal device 20-D is a terminal device used by shipper D. However, the number of shippers and logistics companies is not limited as long as there is more than one.

[0147] Each of the multiple terminal devices 20 may be realized by a computer. Typically, each of the multiple terminal devices 20 may be realized by a PC (Personal Computer). However, each of the multiple terminal devices 20 may also be realized by a terminal device other than a PC (e.g., a smartphone or a tablet terminal). The utility model generation device 12 may be realized by a computer. Typically, the utility model generation device 12 may be realized by a computer such as a server.

[0148] 8, the second embodiment of the present invention mainly assumes that a logistics company and a shipper each use terminal device 20. Utility models corresponding to the logistics company and the shipper can be generated by utility model generation device 12. Furthermore, negotiation agents in which the utility models of the logistics company and the shipper are set can be generated by utility model generation device 12.

[0149] The above describes an example of the configuration of the automated negotiation system 2 according to the second embodiment of the present invention.

[0150] (2-2. Example of functional configuration of utility model generation device) Next, an example of the functional configuration of the utility model generation device 12 according to the second embodiment of the present invention will be described with reference to FIG.

[0151] 9 is a block diagram showing an example of the functional configuration of a utility model generation device 12 according to the second embodiment of the present invention. As shown in FIG. 9, the utility model generation device 12 according to the second embodiment of the present invention differs from the utility model generation device 11 according to the first embodiment of the present invention in that it includes a communication unit 121 instead of the UI unit 114.

[0152] The communication unit 121 is a communication interface for transmitting and receiving various types of information to and from other devices via the network 50. For example, the communication unit 121 can function as a transmitting unit that transmits various types of information to the terminal device 20 via the network 50. The communication unit 121 can also function as a receiving unit that receives various types of information from the terminal device 20 via the network 50.

[0153] In the first embodiment of the present invention, data directly output to the UI unit 114 of the utility model generation device 11 is output to the UI unit 114 of the terminal device 20 via the communication unit 121 in the second embodiment of the present invention. Furthermore, in the first embodiment of the present invention, data directly input to the UI unit 114 of the utility model generation device 11 is received by the communication unit 121 via the UI unit 114 of the terminal device 20 in the second embodiment of the present invention.

[0154] An example of the functional configuration of the utility model generation device 12 according to the second embodiment of the present invention has been described above.

[0155] (2-3. Example of terminal device functional configuration) Next, an example of the functional configuration of the terminal device 20 according to the second embodiment of the present invention will be described. Fig. 10 is a diagram showing an example of the functional configuration of the terminal device 20 according to the second embodiment of the present invention. As shown in Fig. 10, the terminal device 20 according to the second embodiment of the present invention includes a control unit 210, a UI unit 114, a storage unit 230, and a communication unit 240. The UI unit 114 has the same functions as the UI unit 114 included in the utility model generation device 11 according to the first embodiment of the present invention.

[0156] The control unit 210 has a function of controlling the overall operation of the terminal device 20. The control unit 210 may be configured by dedicated hardware, or may be realized by a CPU (Central Processing Unit) built into the terminal device 20 expanding a program stored in a ROM (Read Only Memory) into a RAM (Random Access Memory) and executing the program. Such a program may be provided, or a storage medium storing such a program may also be provided.

[0157] The storage unit 230 is a storage device capable of storing programs and data for operating the control unit 210. For example, the storage device may be a non-volatile storage device. The storage unit 230 can also temporarily store various data required in the course of operation of the control unit 210.

[0158] The communication unit 240 is a communication interface for transmitting and receiving various types of information to and from other devices via the network 50. For example, the communication unit 240 can function as a transmitter that transmits various types of information to the utility model generation device 12 via the network 50. The communication unit 240 can also function as a receiver that receives various types of information from the utility model generation device 12 via the network 50.

[0159] An example of the functional configuration of the terminal device 20 according to the second embodiment of the present invention has been described above.

[0160] (2-4. Effects) According to the second embodiment of the present invention, it is possible to enjoy the same effects as those achieved by the first embodiment of the present invention.

[0161] The second embodiment of the present invention has been described above.

[0162] <3. Hardware configuration example> Next, a description will be given of an example of the hardware configuration of the utility model generation device 11 according to the embodiment of the present invention. As an example of the hardware configuration, an example of the hardware configuration of an information processing device 900 will be described below.

[0163] The hardware configuration example of the information processing device 900 described below is merely an example. Therefore, the hardware configuration of the utility model generation device 11 may be modified by deleting unnecessary components from the hardware configuration of the information processing device 900 described below, or by adding new components. The hardware configurations of the utility model generation device 12 and the terminal devices 20A to 20D may also be implemented in the same manner as the hardware configuration of the utility model generation device 11.

[0164] 11 is a diagram showing the hardware configuration of an information processing device 900. The information processing device 900 includes a CPU (Central Processing Unit) 901, a ROM (Read Only Memory) 902, a RAM (Random Access Memory) 903, a host bus 904, a bridge 905, an external bus 906, an interface 907, an input device 908, an output device 909, a storage device 910, and a communication device 911.

[0165] The CPU 901 functions as an arithmetic processing unit and control unit, and controls the overall operation of the information processing device 900 in accordance with various programs. The CPU 901 may also be a microprocessor. The ROM 902 stores programs used by the CPU 901, calculation parameters, etc. The RAM 903 temporarily stores programs used in the execution of the CPU 901, parameters that change as appropriate during the execution, etc. These are interconnected by a host bus 904 that is composed of a CPU bus, etc.

[0166] The host bus 904 is connected to an external bus 906, such as a PCI (Peripheral Component Interconnect / Interface) bus, via a bridge 905. It is not necessary to configure the host bus 904, bridge 905, and external bus 906 separately, and these functions may be implemented on a single bus.

[0167] The input device 908 is composed of input means such as a mouse, keyboard, touch panel, buttons, microphone, switches, and levers that allow the user to input information, and an input control circuit that generates an input signal based on the user's input and outputs it to the CPU 901. By operating this input device 908, the user operating the information processing device 900 can input various data to the information processing device 900 and instruct it to perform processing operations.

[0168] The output device 909 includes, for example, a display device such as a CRT (Cathode Ray Tube) display device, a liquid crystal display (LCD) device, an OLED (Organic Light Emitting Diode) device, or a lamp, and an audio output device such as a speaker.

[0169] The storage device 910 is a device for storing data. The storage device 910 may include a storage medium, a recording device for recording data on the storage medium, a reading device for reading data from the storage medium, and a deletion device for deleting data recorded on the storage medium. The storage device 910 is configured, for example, with an HDD (Hard Disk Drive). This storage device 910 drives a hard disk and stores programs executed by the CPU 901 and various data.

[0170] The communication device 911 is, for example, a communication interface configured with a communication device for connecting to a network, etc. The communication device 911 may be compatible with either wireless communication or wired communication.

[0171] An example of the hardware configuration of the utility model generation device 11 according to the embodiment of the present invention has been described above.

[0172] <4. Summary> Although the preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to these examples. It is clear that a person skilled in the art to which the present invention pertains can conceive of various modifications and alterations within the scope of the technical ideas set forth in the claims, and it is understood that these also naturally fall within the technical scope of the present invention. [Explanation of symbols]

[0173] 11, 12 Utility model generator 111 Input section 112 Utility model generation unit 113 Utility Model Evaluation Unit 114 UI section 115 Utility model memory unit 116 Automatic Negotiation Unit 117 Negotiation performance data storage unit 118 Evaluation data generation unit 121 Communications Department 2. Automated Negotiation System 20 Terminal equipment 210 Control Unit 230 Storage section 240 Communications Department 50 Network

Claims

1. a utility model generation unit that generates a utility model based on history data including one or more conditions related to past negotiations; a presentation control unit that controls presentation of information related to the utility value to a user based on the utility value output from the utility model corresponding to the evaluation data; a utility model update unit that, when an instruction to update the utility model is input from the user, updates the utility model based on an instruction to modify the utility model input from the user to obtain an updated utility model; Equipped with The evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions; the utility model update unit updates the utility model so that a weight of a utility value corresponding to the selected item is changed in accordance with a weight of a utility value for an item other than the selected item. Utility model generator.

2. the utility model update unit updates the utility model to obtain an updated utility model every time the update instruction is input; The utility model generating device according to claim 1 .

3. The utility model update unit updates the utility model so that a weight of a utility value corresponding to the selected item is greater than a weight of a utility value for an item other than the selected item. The utility model generating device according to claim 1 .

4. the modification instruction includes a modification value for the utility value; the utility model update unit updates the utility model so that a utility value output from the utility model corresponding to the evaluation data is changed to the corrected value. The utility model generating device according to claim 1 .

5. There are a plurality of pieces of evaluation data, the correction instruction includes correction information regarding the magnitude relationship of utility values ​​corresponding to the plurality of evaluation data, the utility model update unit updates the utility model so that the magnitude relationship between utility values ​​output from the utility model corresponding to the evaluation data is changed in accordance with the correction information. The utility model generating device according to claim 1 .

6. the presentation control unit controls presentation of the utility value as information about the utility value. The utility model generating device according to any one of claims 1 to 5.

7. the presentation control unit determines whether the utility value exceeds a threshold, acquires a determination result, and controls presentation of the determination result as information related to the utility value. The utility model generating device according to any one of claims 1 to 5.

8. The utility model generation device comprises: an automatic negotiation unit that performs automatic negotiation based on the updated utility model; The utility model generating device according to any one of claims 1 to 7.

9. the utility model update unit updates the updated utility model again based on the evaluation result of the automatic negotiation not satisfying a predetermined condition. The utility model generating device according to claim 8 .

10. the utility model update unit updates the updated utility model again unless an evaluation result of the result of the automatic negotiation based on the updated utility model satisfies the predetermined condition. The utility model generating device according to claim 9 .

11. the evaluation result is an approval or rejection of the result of the automatic negotiation input by the user; the predetermined condition is a condition that the evaluation result is approval of the result of the automatic negotiation; The utility model generating device according to claim 9 or 10.

12. There are a plurality of results of the automatic negotiation; the evaluation results are approval or rejection of each of the plurality of results of the automatic negotiations input by the user; The predetermined condition is that the ratio of the approvals to the rejections is greater than a predetermined ratio. The utility model generating device according to claim 9 or 10.

13. generating a utility model based on historical data including one or more conditions related to past negotiations; Controlling presentation of information relating to the utility value to a user based on the utility value output from the utility model corresponding to the evaluation data; When an instruction to update the utility model is input from the user, updating the utility model based on the instruction to correct the utility model input from the user to obtain an updated utility model; Equipped with The evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions; Obtaining the updated utility model includes updating the utility model so that a weight of a utility value corresponding to the selected item is changed according to a weight of a utility value for an item other than the selected item. A computer-implemented method for generating utility models.

14. Computer, a utility model generation unit that generates a utility model based on history data including one or more conditions related to past negotiations; a presentation control unit that controls presentation of information related to the utility value to a user based on the utility value output from the utility model corresponding to the evaluation data; a utility model update unit that, when an instruction to update the utility model is input from the user, updates the utility model based on an instruction to modify the utility model input from the user to obtain an updated utility model; Equipped with The evaluation data includes a plurality of conditions, the correction instruction includes a selection item selected by the user from items corresponding to each of the plurality of conditions; the utility model update unit updates the utility model so that a weight of a utility value corresponding to the selected item is changed in accordance with a weight of a utility value for an item other than the selected item. A program that functions as a utility model generator.

Images