JP6813328B2 - Predictors, prediction methods, and prediction programs - Google Patents

Description

The present invention relates to a prediction device, a prediction method, and a prediction program.

Conventionally, there are known techniques for providing various services based on user evaluation. For example, a technique for determining (predicting) the priority of providing a predicted time based on the evaluation of a participant (user) is provided.

Japanese Unexamined Patent Publication No. 2011-198358

However, in the above-mentioned prior art, it is not possible to appropriately provide a predetermined service without information on the evaluation of the user in the predetermined service, and it is desired to appropriately predict the evaluation of the user. However, it can be difficult to properly predict user ratings for a given service. For example, it is difficult to properly predict a user's evaluation of a service that the user is not using.

The present application has been made in view of the above, and an object of the present application is to provide a prediction device, a prediction method, and a prediction program for appropriately predicting a user's evaluation in a predetermined service.

The prediction device according to the present application is a model generated by an acquisition unit that acquires behavioral information about one user on the Internet, behavioral information about another user on the Internet, and evaluation of the other user in a predetermined service. It is characterized by including a prediction unit that predicts the evaluation of the one user in the predetermined service based on the behavior information about the one user acquired by the acquisition unit.

According to one aspect of the embodiment, there is an effect that the evaluation of the user in a predetermined service can be appropriately predicted.

FIG. 1 is a diagram showing an example of a generation process according to an embodiment. FIG. 2 is a diagram showing an example of prediction processing according to the embodiment. FIG. 3 is a diagram showing a configuration example of the prediction device according to the embodiment. FIG. 4 is a diagram showing an example of a user information storage unit according to the embodiment. FIG. 5 is a diagram showing an example of the behavior information storage unit according to the embodiment. FIG. 6 is a diagram showing an example of a model information storage unit according to the embodiment. FIG. 7 is a diagram showing an example of a certainty information storage unit according to the embodiment. FIG. 8 is a flowchart showing an example of the generation process according to the embodiment. FIG. 9 is a flowchart showing an example of the prediction process according to the embodiment. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the function of the prediction device.

Hereinafter, a prediction device, a prediction method, and a mode for carrying out the prediction program (hereinafter referred to as “the embodiment”) according to the present application will be described in detail with reference to the drawings. The prediction device, prediction method, and prediction program according to the present application are not limited by this embodiment. Further, in each of the following embodiments, the same parts are designated by the same reference numerals, and duplicate description is omitted.

(Embodiment)
[1. Generation process]
First, an example of the generation process according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram showing an example of a generation process according to an embodiment. In the example shown in FIG. 1, the prediction device 100 generates a model for predicting the evaluation in the service X based on the evaluation in the service X of the plurality of users and the behavior information of the plurality of users. Further, the prediction device 100 predicts the evaluation of the user in the service X using the generated model and estimates the certainty of the predicted evaluation, which will be described with reference to FIG. Further, the evaluation in the service X referred to here may be an evaluation based on an evaluation (score) or the like in which each user who exchanges between the users who use the service X scores another user. For example, the evaluation in the service X may be an evaluation based on the average of the evaluations (scores) given by one user to the other user among the users who have made a transaction in the electronic commerce. 1 and 2 show a case where users U1 to U4 and U11 use terminal devices 10-1 to 10-4 and 10-11, respectively. When the terminal devices 10-1 to 10-4 and 10-11 are described without distinction, they are collectively referred to as the terminal device 10.

As shown in FIG. 1, the prediction system 1 includes a terminal device 10, a service providing device 200 (see FIG. 2), and a prediction device 100. The terminal device 10 and the prediction device 100 are communicably connected by wire or wirelessly via a predetermined communication network (not shown). The prediction system 1 shown in FIG. 1 may include a plurality of terminal devices 10, a plurality of service providing devices 200, and a plurality of prediction devices 100.

The terminal device 10 is an information processing device used by the user. The terminal device 10 is realized by, for example, a smartphone, a tablet terminal, a notebook PC (Personal Computer), a desktop PC, a mobile phone, a PDA (Personal Digital Assistant), or the like. FIG. 1 shows a case where the terminal device 10 is a smartphone.

In addition, the terminal device 10 accepts operations by the user. In the example shown in FIG. 1, the terminal device 10 displays the content according to the operation of the user, and displays the search result according to the search query (hereinafter, also simply referred to as “query”) input by the user. .. In the following, the terminal device 10 may be referred to as a user. That is, in the following, the user can be read as the terminal device 10.

Specifically, FIG. 1 shows a case where the terminal device 10-1 is a smartphone used by a user identified by the user ID “U1” (hereinafter, may be referred to as “user U1”). As described above, when "user U * (* is an arbitrary numerical value)" is described, it means that the user is a user identified by the user ID "U *". For example, when described as "user U1", it indicates that the user is a user identified by the user ID "U1".

The prediction device 100 is an information processing device that generates a model that predicts the evaluation of a predetermined service based on the behavior information about another user on the Internet and the evaluation of the other user in the predetermined service. Further, the prediction device 100 is an information processing device that predicts the evaluation of one user in a predetermined service based on the generated model and the acquired behavioral information about one user.

First, in the example of FIG. 1, the prediction device 100 acquires the evaluation in the service X of the user U1 who uses the terminal device 10-1 (step S11-1). For example, the prediction device 100 acquires the evaluation of the user U1 in the service X from the terminal device 10-1. In the example of FIG. 1, the prediction device 100 acquires information indicating that the evaluation of the user U1 in the service X is “3”. Further, the prediction device 100 acquires the behavior information of the user U1 on the Internet (step S12-1). For example, the prediction device 100 acquires behavior information of the user U1 other than the service X from the terminal device 10-1.

Further, in the example of FIG. 1, the prediction device 100 acquires the evaluation in the service X of the user U2 who uses the terminal device 10-2 (step S11-2). For example, the prediction device 100 acquires the evaluation of the user U2 in the service X from the terminal device 10-2. In the example of FIG. 1, the prediction device 100 acquires information indicating that the evaluation of the user U2 in the service X is “5”. Further, the prediction device 100 acquires the behavior information of the user U2 on the Internet (step S12-2). For example, the prediction device 100 acquires behavior information of the user U2 other than the service X from the terminal device 10-2.

Further, in the example of FIG. 1, the prediction device 100 acquires the evaluation in the service X of the user U3 who uses the terminal device 10-3 (step S11-3). For example, the prediction device 100 acquires the evaluation of the user U3 in the service X from the terminal device 10-3. In the example of FIG. 1, the prediction device 100 acquires information indicating that the evaluation of the user U3 in the service X is “2”. Further, the prediction device 100 acquires the behavior information of the user U3 on the Internet (step S12-3). For example, the prediction device 100 acquires behavior information of the user U3 other than the service X from the terminal device 10-3.

Further, in the example of FIG. 1, the prediction device 100 acquires the evaluation in the service X of the user U4 who uses the terminal device 10-4 (step S11-4). For example, the prediction device 100 acquires the evaluation of the user U4 in the service X from the terminal device 10-4. In the example of FIG. 1, the prediction device 100 acquires information indicating that the evaluation of the user U4 in the service X is “4”. Further, the prediction device 100 acquires the behavior information of the user U4 on the Internet (step S12-4). For example, the prediction device 100 acquires behavior information of the user U4 other than the service X from the terminal device 10-4.

It should be noted that steps S11-1 to S11-4 are for explaining the process, and any of steps S11-1 to S11-4 may be performed first, and each step S11-1 to S11-4 may be performed first. May be performed multiple times. Hereinafter, when steps S11-1 to S11-4 will be described without distinction, they are collectively referred to as step S11. Further, steps S12-1 to S12-4 are for explaining the process, and any of steps S12-1 to S12-4 may be performed first, and each step S12-1 to S12-4 may be performed first. May be performed multiple times. Hereinafter, when steps S12-1 to S12-4 will be described without distinction, they are collectively referred to as step S12. Further, for example, each step S11 may be performed a plurality of times, and step S12 may be performed at a predetermined timing to acquire action information. For example, the prediction device 100 stores the evaluations of the services X such as the users U1 to U4 acquired in step S11 in the user information storage unit 121 in association with each user. Further, for example, the prediction device 100 stores the action information of each user U1 to U4 acquired in step S12 in association with each user in the action information storage unit 122. The prediction device 100 may acquire the evaluation in the service X of the users U1 to U4 and the behavior information of the users U1 to U4 and the like from an external device other than the terminal device 10.

Then, the prediction device 100 generates a model based on the evaluation in the service X of each user U1 to U4 or the like and the action information of each user U1 to U4 or the like (step S13). For example, the prediction device 100 can use the evaluation in the service X stored in the user information storage unit 121 in association with each user and the action information of each user stored in the action information storage unit 122 in association with each user. Based on this, a model that predicts the evaluation in service X is generated. For example, the prediction device 100 generates a model that predicts the evaluation of a user in the service X when the behavior information of the user is input.

In the example of FIG. 1, the prediction device 100 generates the model M11 as shown in the model information storage unit 123. Further, as shown in the model information storage unit 123, in the model information regarding the model M11, the weight of the feature 1 is "0.5", the weight of the feature 2 is "-0.4", and the weight of the feature 3 is "0. 2 ”etc. For example, the prediction device 100 predicts (calculates) the evaluation of the user in the service X by inputting the behavior information of the user.

Further, the prediction device 100 generates certainty information based on the evaluation in the service X of each user U1 to U4 and the like and the action information of each user U1 to U4 and the like (step S14). For example, the prediction device 100 may determine the degree of certainty associated with each action information based on the fluctuation range of the evaluation of each user when each action information is included.

For example, the behavior of "page A browsing", which is the target identified by the target ID "TG10" in the certainty information storage unit 124 shown in FIG. 1, is associated with the certainty "10%". In this case, when the evaluation is predicted only by the behavior of "page A browsing", it indicates that the probability that the predicted evaluation matches the actual evaluation is "10%". That is, when the behavior of "page A browsing" is included in the behavior information, it indicates that the predicted evaluation varies widely.

For example, the search behavior using the target "query A" identified by the target ID "TG11" in the certainty information storage unit 124 shown in FIG. 1 is associated with the certainty "80%". In this case, when the evaluation is predicted only by the search behavior using "query A", it indicates that the probability that the predicted evaluation matches the actual evaluation is "80%". That is, when the search behavior using "query A" is included in the behavior information, the predicted evaluation shows that the variation is small.

As described above, the prediction device 100 provides certainty information indicating the correspondence between the certainty and the target (behavior) based on the evaluation in the service X of each user U1 to U4 and the action information of each user U1 to U4 and the like. Generate. The prediction device 100 estimates the certainty of the predicted evaluation based on the certainty information and various information associated with each object (behavior), and this point will be described with reference to FIG.

As described above, the prediction device 100 appropriately predicts the evaluation of the user in the predetermined service by learning based on the evaluation in the service X of each user U1 to U4 and the behavior information of each user U1 to U4 and the like. The model used to do this can be generated. Further, the prediction device 100 generates confidence level information indicating the correspondence between the certainty level and the target (behavior) based on the evaluation in the service X of each user U1 to U4 and the like and the action information of each user U1 to U4 and the like. This makes it possible to estimate the certainty of the predicted evaluation, which will be described in detail in FIG. The prediction device 100 predicts the user's evaluation as a seller in the electronic commerce service and the buyer when the user's evaluation in the electronic commerce service is divided into an evaluation as a buyer and an evaluation as a seller. A model that predicts the evaluation as a model may be generated separately.

[2. Prediction processing]
An example of the prediction process according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram showing an example of prediction processing according to the embodiment. FIG. 2 shows a case where the prediction device 100 predicts an evaluation of a user in service X based on the model M11 generated in FIG. 1, certainty information, and the like, and estimates the certainty of the evaluation. Further, FIG. 2 shows a case where the prediction device 100 provides the predicted evaluation in the service X and information on the certainty of the evaluation to the service providing device 200 that provides the service X. In the example shown in FIG. 2, the user U11 shows a case where the user U11 is a non-user who does not use the service X. It is assumed that the service X is a predetermined electronic commerce service.

First, the prediction device 100 acquires the behavior information of the user U11 on the Internet (step S21). For example, the prediction device 100 acquires behavior information of the user U11 other than the service X from the terminal device 10-11. For example, the prediction device 100 acquires behavior information of the user U11 other than the service X in advance.

Then, the user U11 requests the service providing device 200 to use the service X by operating the terminal device 10-11 (step S22). For example, the user U11 transmits a request for using the service X to the service providing device 200 by performing a predetermined operation on the shop site A page W11 displayed on the terminal device 10-11.

Upon receiving the request for using the service X from the terminal device 10, the service providing device 200 requests the prediction device 100 to provide the information (step S23). For example, the service providing device 200 requests the prediction device 100 to predict the evaluation of the user U11 in the service X.

The prediction device 100, which has acquired the prediction request from the service providing device 200, predicts the evaluation based on the behavior information list AL11 of the user U11 and the model M11 (step S24). For example, the prediction device 100 predicts the evaluation based on the behavior information included in the behavior information list AL11 of the user U11 acquired in step S21 and the model M11 stored in the model information storage unit 123. In the example of FIG. 2, the prediction device 100 predicts the evaluation of the user U11 in the service X as "3.2" based on the model M11 stored in the model information storage unit 123, as shown in the predicted value SC11. ..

Further, the prediction device 100 estimates the certainty of the evaluation predicted based on the behavior information list AL11 of the user U11 and the certainty information (step S25). For example, the prediction device 100 determines the certainty of the evaluation predicted based on the action information included in the action information list AL11 of the user U11 acquired in step S21 and the certainty information stored in the certainty information storage unit 124. presume. In the example of FIG. 2, the prediction device 100 evaluates the user U11 in the service X based on the certainty associated with each action stored in the certainty information storage unit 124, as shown in the estimated value AD11. The certainty of "3.2" is estimated to be "75%". For example, since the action information list AL11 of the user U11 includes a search action using "query A" having a high certainty of "80%", the prediction device 100 is evaluated by the service X of the user U11 as "". The certainty of "3.2" is highly estimated to be "75%". For example, the prediction device 100 may use the average value of the conviction of each action as the conviction of the predicted evaluation, or predict the conviction of each action using a value weighted based on the number of times of each action. The confidence level of the evaluation may be estimated. For example, in the prediction device 100, when the search action using "query A" is "3" times and the number of views of page A is "1" times, the certainty level regarding the search action using "query A". May be three times greater than the weight of confidence regarding the number of views of page A. The above-mentioned process is an example, and the prediction device 100 may estimate the certainty of the predicted evaluation by appropriately using various information.

After that, the prediction device 100 provides information that combines the evaluation "3.2" in the service X of the user U11 predicted in step S24 and the certainty "75%" of the predicted evaluation "3.2". Provided to 200 (step S26). In the example of FIG. 2, the prediction device 100 provides the service providing device 200 with the evaluation “3.2” shown in the predicted value SC11 and the certainty “75%” shown in the estimated value AD11.

The service providing device 200 provided by the prediction device 100 with the evaluation "3.2" shown in the predicted value SC11 and the certainty "75%" shown in the estimated value AD11 determines whether or not the service X can be provided to the user U11 ( Step S27). For example, in the service providing device 200, when the certainty of the evaluation predicted by the predicting device 100 is "50%" or more and the evaluation in the service X of the user is the threshold value "3.0", the service X is given to the user. Determined to provide. In the example of FIG. 2, the evaluation of the user U11 in the service X is "3.2", the certainty of the predicted evaluation "3.2" is "75%", and both conditions are satisfied, so that the service providing device 200 provides service X to user U11 (step S28).

As described above, the prediction device 100 predicts the evaluation of the non-user in the predetermined service based on the behavior information about the non-user and the model, and thereby determines the user who is not using the predetermined service. Can properly predict the evaluation of the service. In this way, the prediction device 100 can appropriately predict the user's evaluation in a predetermined service. Further, the prediction device 100 estimates the certainty of the evaluation in the predetermined service predicted by the model based on the certainty assigned to each of the behavioral information about the other user, and thereby the predicted evaluation confidence. Information about the degree can be estimated appropriately.

For example, since the user's evaluation in a predetermined service is an evaluation value associated with the individual user, if the predicted evaluation is lower than the evaluation given when the user actually uses the predetermined service, the user is notified. The disadvantage is great. For example, the disadvantage to the user is large as compared with the evaluation value of the user predicted when the advertisement is delivered. Further, for example, when the service X is a service that can be used only by a user having an evaluation equal to or higher than a predetermined threshold value, the service of the service X is provided to determine whether the user who has not yet used the service X is a user suitable for using the service X. It can be difficult for a person to judge. In such a case, the evaluation of the user in the predetermined service is a material for determining whether or not the predetermined service allows the user to use the service individually. Therefore, when the predicted evaluation actually uses the predetermined service, the user actually uses the service. If the evaluation is higher than the given evaluation, the service will be provided to users who should not originally provide the service, which is a great disadvantage for the service provider. However, based on the evaluation of the service X provided by the prediction device 100 and the certainty thereof, the service provider of the service X appropriately determines whether or not the non-user is a user suitable for using the service X. Can be done.

The prediction device 100 may estimate the certainty of the predicted evaluation by appropriately using various information, not limited to the above-mentioned information. For example, the prediction device 100 may use a word having a low number of searches as a whole as a word having a low certainty. Further, for example, the prediction device 100 may change the degree of certainty according to the amount of behavioral information. For example, the prediction device 100 may increase the certainty of the predicted evaluation when the amount of behavior information is equal to or more than a predetermined threshold value. Further, for example, the prediction device 100 may lower the certainty of the predicted evaluation when the amount of behavior information is less than a predetermined threshold value. For example, the prediction device 100 may estimate the certainty of the predicted evaluation based on the number of action histories and the total number of searches of the query used by the user. For example, the prediction device 100 approaches 0 as the number of action histories and the total number of searches of the query used by the user decreases, and the amount increases and gradually fluctuates to approach 1. Based on this, the certainty of the predicted evaluation may be estimated. For example, when the prediction device 100 includes a query used by a predetermined ratio or more of users or an action performed by a predetermined ratio or more of the users in the behavior information of the prediction target user, the prediction device 100 predicts the prediction target user. The certainty may be estimated so that the certainty of the evaluation is high.

As described above, when the model for predicting the user's evaluation as a buyer and the model for predicting the evaluation as a seller in the electronic commerce service are individually generated, the prediction device 100 appropriately generates each model. It may be used to estimate a user's reputation as a seller or as a buyer. In this case, the prediction device 100 may estimate the certainty of the predicted evaluation as described above. Further, the prediction device 100 determines the evaluation in the predetermined service of the user by using both the evaluation in the predetermined service predicted for the user and the evaluation in the predetermined service actually scored and given by the user. May be good. For example, the prediction device 100 considers that when the behavior in a predetermined service for a certain user, the number of times scored by another user, or the like does not reach a predetermined threshold value, the information regarding the evaluation in the predetermined service of the user is insufficient. The evaluation in the predetermined service predicted for may be added. For example, when the prediction device 100 lacks information on the evaluation of the user's predetermined service, the prediction device 100 is a value obtained by multiplying the predicted evaluation by a predetermined coefficient (for example, a coefficient "α". α is an arbitrary value of 0 to 1). And the value obtained by multiplying the evaluation actually given to the user by a predetermined coefficient (for example, the coefficient "1-α") may be used as the evaluation in the user's predetermined service. For example, the prediction device 100 may reduce the influence of the predicted evaluation by reducing the value of the coefficient “α” as the information regarding the evaluation in the predetermined service of the user increases. As a result, the prediction device 100 gradually increases the influence of the evaluation actually given to the user as the information about the evaluation in the predetermined service of the user increases, so that the evaluation given to the actual user is gradually increased. The influence of can be increased. Therefore, the prediction device 100 can appropriately estimate the evaluation of the user's predetermined service by adding the predicted evaluation even when the information regarding the evaluation of the user's predetermined service is insufficient.

[3. Predictor configuration]
Next, the configuration of the prediction device 100 according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram showing a configuration example of the prediction device according to the embodiment. As shown in FIG. 3, the prediction device 100 includes a communication unit 110, a storage unit 120, and a control unit 130. The prediction device 100 has an input unit (for example, a keyboard, a mouse, etc.) that receives various operations from the administrator of the prediction device 100, and a display unit (for example, a liquid crystal display, etc.) for displaying various information. You may.

(Communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card) or the like. Then, the communication unit 110 is connected to the network by wire or wirelessly, and transmits / receives information to / from the terminal device 10.

(Memory unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory (Flash Memory), or a storage device such as a hard disk or an optical disk. As shown in FIG. 3, the storage unit 120 according to the embodiment includes a user information storage unit 121, an action information storage unit 122, a model information storage unit 123, and a certainty information storage unit 124.

(User information storage unit 121)
The user information storage unit 121 according to the embodiment stores various information about the user. For example, the user information storage unit 121 stores various information related to user attributes. FIG. 4 is a diagram showing an example of a user information storage unit according to the embodiment. The user information storage unit 121 shown in FIG. 4 includes items such as "user ID", "age", "gender", and "evaluation (service X)".

The "user ID" indicates identification information for identifying a user. For example, the user identified by the user ID "U1" corresponds to the user U1 shown in the example of FIG. Further, "age" indicates the age of the user identified by the user ID. The "age" may be the specific age of the user identified by the user ID, such as 35 years old. Further, "gender" indicates the gender of the user identified by the user ID. Further, "evaluation (service X)" indicates the evaluation in service X of the user identified by the user ID.

For example, in the example shown in FIG. 4, the age of the user identified by the user ID "U1" is "30's", and the gender is "male". Further, the evaluation of the user identified by the user ID "U1" in the service X is "3". Further, for example, in the example shown in FIG. 4, the age of the user identified by the user ID "U2" is "20's", and the gender is "female". Further, the evaluation of the user identified by the user ID "U2" in the service X is "5".

The user information storage unit 121 is not limited to the above, and may store various information depending on the purpose. For example, the user information storage unit 121 may store other demographic attribute information and psychographic attribute information in addition to age and gender. For example, the user information storage unit 121 may store information such as name, home, work location, interests, family structure, income, and lifestyle.

(Behavior information storage unit 122)
The behavior information storage unit 122 according to the embodiment stores various information related to the user's behavior. FIG. 5 is a diagram showing an example of the behavior information storage unit according to the embodiment. The action information storage unit 122 shown in FIG. 5 includes items such as "user ID", "action ID", "type", "content", and "date and time".

The "user ID" indicates identification information for identifying a user. Further, the "behavior ID" indicates information that identifies the user's behavior. Further, "type" indicates information regarding the type of action of the corresponding user. In addition, "content" indicates the content targeted in the corresponding user's behavior. The "date and time" indicates the date and time when the corresponding user action was performed. In the example of FIG. 5, the information stored in the "date and time" is illustrated by a code such as "dt10", but the "date and time" includes "September 21, 2016 22:01:43" or the like. It is assumed that the specific date and time are stored.

For example, in the example shown in FIG. 5, the user (user U1) identified by the user ID "U1" has an action identified by the action ID "AC10", an action identified by the action ID "AC11", or an action. It indicates that the action identified by the ID "AC12", the action identified by the action ID "AC13", or the like has been performed. In the example shown in FIG. 5, it is shown that the user U1 browses the content of the page A (action AC10) at the date and time dt10. Further, in the example shown in FIG. 5, it is shown that the user U1 has performed a search (action AC11) using the query "query A" on the date and time dt11. Further, in the example shown in FIG. 5, it is shown that the user U1 has performed a search (action AC12) using the query “query B” at the date and time dt12. Further, in the example shown in FIG. 5, it is shown that the user U1 has purchased the product A (behavior AC13) at the date and time dt13.

The behavior information storage unit 122 is not limited to the above, and may store various information depending on the purpose. Further, although FIG. 5 shows a case where the action information is stored in the action information storage unit 122 for each user ID, the action information is not limited to each user ID and may be stored, for example, in chronological order.

(Model information storage unit 123)
The model information storage unit 123 according to the embodiment stores information about the model. For example, the model information storage unit 123 stores the model information generated by the generation process. FIG. 6 is a diagram showing an example of a model information storage unit according to the embodiment. The model information storage unit 123 shown in FIG. 6 has items such as "feature 1" to "feature 3" corresponding to each model M11 and the like as model information. Although only the model M11 is shown in FIG. 6, a plurality of model information such as the models M2 and M3 may be stored.

For example, in the example shown in FIG. 6, the model information regarding the model M11 is such that the weight of the feature 1 is "0.5", the weight of the feature 2 is "-0.4", the weight of the feature 3 is "0.2", and the like. Indicates that. For example, when the features (features) of the model are represented by an m-dimensional vector, the number of features is m, and the weights of features 1 to m are stored.

The model information storage unit 123 is not limited to the above, and may store various model information depending on the purpose.

(Confidence information storage unit 124)
The certainty information storage unit 124 according to the embodiment stores various information regarding the certainty. FIG. 7 shows an example of the certainty information storage unit 124 according to the embodiment. The certainty information storage unit 124 shown in FIG. 7 has items such as "target ID", "target", and "confidence". In the example of FIG. 7, a case where a certainty level is associated with each action is shown.

The "object ID" indicates identification information for identifying an object to which the certainty is associated. “Target” indicates the content of the target, that is, the content of the action to which the certainty is associated. “Confidence” indicates the certainty associated with the object. For example, "confidence" indicates the confidence of the evaluation predicted when the corresponding object is included in the behavioral information.

For example, in the example shown in FIG. 7, the behavior of "page A browsing", which is the target identified by the target ID "TG10", is associated with a certainty degree of "10%". In this case, when the evaluation is predicted only by the behavior of "page A browsing", it indicates that the probability that the predicted evaluation matches the actual evaluation is "10%". That is, when the behavior of "page A browsing" is included in the behavior information, it indicates that the predicted evaluation varies widely.

For example, in the example shown in FIG. 7, the search behavior using the target "query A" identified by the target ID "TG11" is associated with a certainty degree of "80%". In this case, when the evaluation is predicted only by the search behavior using "query A", it indicates that the probability that the predicted evaluation matches the actual evaluation is "80%". That is, when the search behavior using "query A" is included in the behavior information, the predicted evaluation shows that the variation is small.

The confidence level information storage unit 124 is not limited to the above, and various information may be stored depending on the purpose. The certainty information storage unit 124 may store the certainty in association with each feature, not limited to the action.

(Control unit 130)
Returning to the description of FIG. 3, the control unit 130 is a controller, and is stored in a storage device inside the prediction device 100 by, for example, a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). It is realized by executing various programs (corresponding to an example of a prediction program) using the RAM as a work area. Further, the control unit 130 is a controller, and is realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 3, the control unit 130 includes an acquisition unit 131, a generation unit 132, a prediction unit 133, an estimation unit 134, and a provision unit 135, and functions and operations of information processing described below. To realize or execute. The internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 3, and may be another configuration as long as it is a configuration for performing information processing described later.

(Acquisition unit 131)
The acquisition unit 131 acquires various types of information. For example, the acquisition unit 131 acquires various information from the user information storage unit 121, the action information storage unit 122, the model information storage unit 123, the certainty information storage unit 124, and the like. Further, the acquisition unit 131 may acquire various information from an external information processing device.

For example, the acquisition unit 131 acquires behavioral information about one user on the Internet. Further, for example, the acquisition unit 131 acquires behavior information regarding one user other than a predetermined service. Further, for example, the acquisition unit 131 acquires behavior information about a non-user who is not using a predetermined service as behavior information about one user.

For example, the acquisition unit 131 acquires information on a search on the Internet by one user as behavioral information on one user. Further, for example, the acquisition unit 131 acquires information on a query used by one user for a search on the Internet as action information on one user. Further, for example, the acquisition unit 131 acquires information regarding viewing of content on the Internet by one user as behavioral information regarding one user. Further, for example, the acquisition unit 131 acquires information on a purchase on the Internet by one user as action information on one user.

In the example of FIG. 1, the acquisition unit 131 acquires the evaluation in the service X of the users U1 to U4 and the like. For example, the acquisition unit 131 acquires the evaluation in the service X of the users U1 to U4 and the like from the terminal devices 10-1 to 10-4 and the like. For example, the acquisition unit 131 acquires information indicating that the evaluation of the user U1 in the service X is "3" from the terminal device 10-1.

In the example of FIG. 1, the acquisition unit 131 acquires the behavior information of the users U1 to U4 and the like on the Internet. For example, the acquisition unit 131 acquires behavior information of users U1 to U4 and the like other than the service X from the terminal devices 10-1 to 10-4 and the like.

(Generator 132)
The generation unit 132 generates various information. For example, the generation unit 132 generates a model that predicts the evaluation in a predetermined service based on the behavior information about the other user and the evaluation of the other user in the predetermined service.

In the example of FIG. 1, the generation unit 132 generates a model based on the evaluation in the service X of each user U1 to U4 and the behavior information of each user U1 to U4 and the like. For example, the generation unit 132 can be used for the evaluation in the service X stored in the user information storage unit 121 in association with each user and the action information of each user stored in the action information storage unit 122 in association with each user. Based on this, a model that predicts the evaluation in service X is generated. For example, the generation unit 132 generates a model that predicts the evaluation of a user in the service X when the behavior information of the user is input.

In the example of FIG. 1, the generation unit 132 is based on the evaluation in the service X of each user stored in the user information storage unit 121 and the behavior information of each user stored in the behavior information storage unit 122 other than the service X. To generate a model. In the example of FIG. 1, the generation unit 132 generates the model M1 as shown in the model information storage unit 123. Further, for example, as shown in the model information storage unit 123, the generation unit 132 has the model information regarding the model M1 that the weight of the feature 1 is "0.5", the weight of the feature 2 is "-0.4", and the feature A model M1 having a weight of 3 such as "0.2" is generated.

(Prediction unit 133)
The prediction unit 133 predicts various types of information. For example, the prediction unit 133 is based on a model generated by behavior information about another user on the Internet and evaluation of another user in a predetermined service, and behavior information about one user acquired by acquisition unit 131. To predict the evaluation of the one user in a predetermined service. For example, the prediction unit 133 predicts the evaluation of one user in a predetermined service based on the model generated by the behavior information about another user other than the predetermined service and the evaluation of the other user in the predetermined service. ..

For example, the prediction unit 133 predicts the evaluation of the non-use user in a predetermined service based on the behavior information about the non-use user acquired by the acquisition unit 131 and the model. In the example of FIG. 2, the prediction unit 133 predicts the evaluation of the user U11 who is a non-user in the service X based on the model M1 stored in the model information storage unit 123.

In the example of FIG. 2, the prediction unit 133 predicts the evaluation based on the behavior information AC11 of the user U11 and the model M11. In the example of FIG. 2, the prediction unit 133 predicts the evaluation based on the behavior information AC11 of the user U11 and the model M11 stored in the model information storage unit 123. In the example of FIG. 2, the prediction device 100 predicts the evaluation of the user U11 in the service X as "3.2" based on the model M11 stored in the model information storage unit 123, as shown in the predicted value SC11. ..

(Estimating unit 134)
The estimation unit 134 estimates various information. For example, the estimation unit 134 estimates the certainty of the evaluation in a predetermined service predicted by the model based on the behavior information about another user. For example, the estimation unit 134 estimates the certainty about the evaluation in the predetermined service predicted by the model based on the certainty assigned to each of the behavioral information about the other user.

In the example of FIG. 2, the estimation unit 134 estimates the certainty of the evaluation predicted based on the behavior information AC11 of the user U11 and the certainty information. For example, the estimation unit 134 estimates the certainty of the evaluation predicted based on the behavior information AC11 of the user U11 and the certainty information stored in the certainty information storage unit 124. In the example of FIG. 2, the estimation unit 134 evaluates the user U11 in the service X based on the certainty associated with each action stored in the certainty information storage unit 124, as shown in the estimated value AD11. The certainty of "3.2" is estimated to be "75%". For example, since the behavior information AC11 of the user U11 includes a search behavior using "query A" having a high certainty of "80%", the estimation unit 134 evaluates the user U11 in the service X to "3". The certainty of ".2" is highly estimated to be "75%".

(Providing section 135)
The providing unit 135 provides various information. For example, the providing unit 135 provides various information to the terminal device 10 and the service providing device 200. For example, the provider 135 provides information on a combination of the evaluation of one user in a predetermined service and the certainty of the evaluation in a predetermined service predicted by the model. For example, the provider 135 provides the service providing device 200 with an evaluation of one user in a predetermined service and a certainty regarding the evaluation in a predetermined service predicted by the model.

[4. Flow of generation process]
Next, the procedure of the generation process by the prediction system 1 according to the embodiment will be described with reference to FIG. FIG. 8 is a flowchart showing an example of the generation process according to the embodiment.

As shown in FIG. 8, the prediction device 100 acquires the evaluations of a plurality of users in a predetermined service (step S101). In the example of FIG. 1, the prediction device 100 acquires evaluations in the service X of a plurality of users such as users U1 to U4.

Further, the prediction device 100 acquires the behavior information of a plurality of users (step S102). In the example of FIG. 1, the prediction device 100 acquires behavior information other than the service X such as users U1 to U4. Note that step S102 may be performed before step S101, and steps S101 and S102 may be performed a plurality of times.

After that, the prediction device 100 generates a model based on the evaluation and the behavior information (step S103). In the example of FIG. 1, the prediction device 100 is based on the evaluation in the service X of each user stored in the user information storage unit 121 and the behavior information of each user stored in the behavior information storage unit 122 other than the service X. To generate a model. For example, the prediction device 100 generates a model M1 based on the evaluation and the behavior information.

[5. Prediction processing flow]
Next, the procedure of the generation process by the prediction system 1 according to the embodiment will be described with reference to FIG. FIG. 9 is a flowchart showing an example of the prediction process according to the embodiment.

As shown in FIG. 9, the prediction device 100 acquires the behavior information of one user (step S201). In the example of FIG. 2, the prediction device 100 acquires the behavior information of the user U11 from the terminal devices 10-11 used by the user U11 who is a non-user of the service X.

Further, the prediction device 100 predicts the evaluation of one user in a predetermined service (step S202). In the example of FIG. 2, the prediction device 100 predicts the evaluation of the user U11 in the service X based on the model M11 stored in the model information storage unit 123.

Further, the prediction device 100 estimates the certainty of the predicted evaluation (step S203). In the example of FIG. 2, the prediction device 100 estimates the certainty of the evaluation of the estimated user U11 in the service X.

Then, the prediction device 100 provides information based on the evaluation and the certainty (step S204). In the example of FIG. 2, the prediction device 100 provides the service providing device 200 that provides the service X with information on a combination of the predicted evaluation and the estimated certainty.

[6. effect〕
As described above, the prediction device 100 according to the embodiment has an acquisition unit 131 and a prediction unit 133. The acquisition unit 131 acquires behavioral information about one user on the Internet. Further, the prediction unit 133 is based on a model generated by behavior information about another user on the Internet and evaluation of another user in a predetermined service, and behavior information about one user acquired by acquisition unit 131. To predict the evaluation of one user in a given service.

As a result, the prediction device 100 according to the embodiment relates to a model generated by behavior information about other users on the Internet and evaluations of other users in a predetermined service, and one user acquired by the acquisition unit 131. Based on the behavior information, it is possible to appropriately predict the user's evaluation in a predetermined service.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires behavior information regarding one user other than a predetermined service. Further, the prediction unit 133 predicts the evaluation of one user in the predetermined service based on the model generated by the behavior information about the other user other than the predetermined service and the evaluation of the other user in the predetermined service. ..

As a result, the prediction device 100 according to the embodiment is based on a model generated by behavioral information about other users other than the predetermined service and evaluation of the other user in the predetermined service, so that the user in the predetermined service The evaluation can be predicted appropriately.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires the behavior information about the non-user who is not using the predetermined service as the behavior information about one user. In addition, the prediction unit 133 predicts the evaluation of the non-use user in the predetermined service based on the behavior information about the non-use user acquired by the acquisition unit 131 and the model.

As a result, the prediction device 100 according to the embodiment uses the predetermined service by predicting the evaluation of the non-user in the predetermined service based on the acquired behavior information regarding the non-user and the model. It is possible to appropriately predict the evaluation of a user who does not have a given service.

Further, the prediction device 100 according to the embodiment has an estimation unit 134. The estimation unit 134 estimates the certainty of the evaluation in the predetermined service predicted by the model based on the behavior information about other users.

As a result, the prediction device 100 according to the embodiment estimates the certainty of the evaluation in the predetermined service predicted by the model based on the behavioral information about the other user, thereby providing the information regarding the certainty of the predicted evaluation. Can be estimated appropriately.

Further, in the prediction device 100 according to the embodiment, the estimation unit 134 estimates the certainty regarding the evaluation in the predetermined service predicted by the model based on the certainty assigned to each of the behavioral information regarding the other users. ..

As a result, the prediction device 100 according to the embodiment predicts by estimating the certainty regarding the evaluation in the predetermined service predicted by the model based on the certainty assigned to each of the behavioral information regarding the other users. Information on the certainty of the evaluation made can be estimated appropriately.

Further, the prediction device 100 according to the embodiment has a providing unit 135. The provider 135 provides information on the combination of the evaluation of one user in a predetermined service and the certainty of the evaluation in the predetermined service predicted by the model.

As a result, the prediction device 100 according to the embodiment provides information on the combination of the evaluation of one user in the predetermined service and the certainty regarding the evaluation in the predetermined service predicted by the model, so that the predicted evaluation can be performed. It is also possible to grasp the degree of certainty.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires information on a search on the Internet as action information on one user.

As a result, the prediction device 100 according to the embodiment can appropriately predict the user's evaluation in a predetermined service by acquiring the information about the search on the Internet as the behavior information about one user. it can.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires information on a query used by one user for a search on the Internet as action information on one user.

As a result, the prediction device 100 according to the embodiment appropriately evaluates the user in the predetermined service by acquiring the information regarding the query used by the one user for the search on the Internet as the action information regarding the one user. Can be predicted.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires information regarding viewing of content on the Internet as action information regarding one user.

As a result, the prediction device 100 according to the embodiment appropriately predicts the user's evaluation in a predetermined service by acquiring the information regarding the viewing of the content on the Internet as the behavior information regarding the one user. be able to.

Further, in the prediction device 100 according to the embodiment, the acquisition unit 131 acquires information regarding a purchase on the Internet by one user as action information regarding one user.

As a result, the prediction device 100 according to the embodiment can appropriately predict the user's evaluation in a predetermined service by acquiring the information about the purchase on the Internet as the behavior information about one user. it can.

Further, the prediction device 100 according to the embodiment has a generation unit 132. The generation unit 132 generates a model that predicts the evaluation in the predetermined service based on the behavior information about the other user and the evaluation of the other user in the predetermined service.

As a result, the prediction device 100 according to the embodiment generates a model that predicts the evaluation in the predetermined service based on the behavior information about the other user and the evaluation of the other user in the predetermined service. The model can be used to appropriately predict the user's evaluation in a given service.

[7. Hardware configuration]
The prediction device 100 according to the above-described embodiment is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 10 is a hardware configuration diagram showing an example of a computer that realizes the function of the prediction device. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM (Read Only Memory) 1300, an HDD (Hard Disk Drive) 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F). ) Has 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400, and controls each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, a program that depends on the hardware of the computer 1000, and the like.

The HDD 1400 stores a program executed by the CPU 1100, data used by such a program, and the like. The communication interface 1500 receives data from another device via the network N and sends it to the CPU 1100, and transmits the data generated by the CPU 1100 to the other device via the network N.

The CPU 1100 controls an output device such as a display or a printer, and an input device such as a keyboard or a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the generated data to the output device via the input / output interface 1600.

The media interface 1700 reads a program or data stored in the recording medium 1800 and provides the program or data to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable Disk), a magneto-optical recording medium such as MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. And so on.

For example, when the computer 1000 functions as the prediction device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the function of the control unit 130 by executing the program loaded on the RAM 1200. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be acquired from another device via the network N.

Although some of the embodiments and modifications of the present application have been described in detail with reference to the drawings, these are examples and vary based on the knowledge of those skilled in the art, including the embodiments described in the disclosure line of the invention. It is possible to carry out the present invention in another form obtained by modifying or improving the above.

[8. Others]
Further, among the processes described in the above-described embodiments and modifications, all or part of the processes described as being automatically performed can be manually performed, or the processes described as being manually performed. It is also possible to automatically perform all or part of the processed processing by a known method. In addition, the processing procedure, specific name, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

Further, each component of each of the illustrated devices is a functional concept, and does not necessarily have to be physically configured as shown in the figure. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or part of the device is functionally or physically distributed / physically in any unit according to various loads and usage conditions. It can be integrated and configured.

In addition, the above-described embodiments and modifications can be appropriately combined as long as the processing contents do not contradict each other.

Further, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the acquisition unit can be read as an acquisition means or an acquisition circuit.

1 Prediction system 100 Prediction device 121 User information storage unit 122 Behavior information storage unit 123 Model information storage unit 124 Confidence information storage unit 130 Control unit 131 Acquisition unit 132 Generation unit 133 Prediction unit 134 Estimation unit 135 Providing unit 10 Terminal device 200 Service Provided device N network

Claims (14)

An acquisition unit that acquires behavioral information about one user on the Internet,
Behavioral information about other users on the Internet, evaluation of the other user in a predetermined service which is an electronic commerce service, and evaluation based on a score of the other user by a user who uses the predetermined service. Based on the model generated by the acquisition unit and the behavioral information about one user acquired by the acquisition unit, a prediction unit that predicts the evaluation of the one user in the predetermined service, and a prediction unit.
A prediction device characterized by being equipped with. An acquisition unit that acquires behavioral information about one user on the Internet,
The predetermined behavior information about another user on the Internet, the model generated by the evaluation of the other user in the predetermined service, and the behavioral information about one user acquired by the acquisition unit. A prediction unit that predicts the evaluation of the one user in the service,
An estimation unit that estimates the certainty of evaluation in the predetermined service predicted by the model based on the behavior information about the other user.
Equipped with a,
The estimation unit
Based on the confidence assigned to each of the behavioral information about the other user, the confidence regarding the evaluation in the predetermined service predicted by the model is estimated.
A prediction device characterized by that. A provider that provides combination information of the evaluation of the one user in the predetermined service and the certainty of the evaluation in the predetermined service predicted by the model.
The prediction device according to claim 2 , further comprising. The acquisition unit
Acquire behavioral information about the one user other than the predetermined service,
The prediction unit
Based on the model generated by the behavioral information about the other user other than the predetermined service and the evaluation of the other user in the predetermined service, the evaluation of the one user in the predetermined service is predicted.
The prediction device according to any one of claims 1 to 3 , wherein the prediction device is characterized. The acquisition unit
As the behavior information about the one user, the behavior information about the non-user who does not use the predetermined service is acquired, and the behavior information is acquired.
The prediction unit
Based on the behavioral information about the non-use user acquired by the acquisition unit and the model, the evaluation of the non-use user in the predetermined service is predicted.
The prediction device according to claim 4 . The acquisition unit
As the action information about the one user, the one user acquires the information about the search on the Internet.
The prediction device according to any one of claims 1 to 5 , wherein the prediction device is characterized. The acquisition unit
As the action information about the one user, the information about the query used by the one user for the search on the Internet is acquired.
The prediction device according to claim 6 . The acquisition unit
As the behavioral information regarding the one user, the one user acquires information regarding browsing of contents on the Internet.
The prediction device according to any one of claims 1 to 7 , wherein the prediction device is characterized. The acquisition unit
As the action information about the one user, the one user acquires the information about the purchase on the Internet.
The prediction device according to any one of claims 1 to 8 , wherein the prediction device is characterized. A generation unit that generates a model that predicts an evaluation in a predetermined service based on behavior information about the other user and the evaluation of the other user in the predetermined service.
The prediction device according to any one of claims 1 to 9 , further comprising. It ’s a computer-based prediction method.
The acquisition process to acquire behavioral information about one user on the Internet,
Behavioral information about other users on the Internet, evaluation of the other user in a predetermined service which is an electronic commerce service, and evaluation based on a score of the other user by a user who uses the predetermined service. A prediction process that predicts the evaluation of the one user in the predetermined service based on the model generated by the above and the behavior information about the one user acquired by the acquisition process.
A prediction method characterized by including. Acquisition procedure to acquire behavior information about one user on the Internet,
Behavioral information about other users on the Internet, evaluation of the other user in a predetermined service which is an electronic commerce service, and evaluation based on a score of the other user by a user who uses the predetermined service. Based on the model generated by the above and the behavioral information about one user acquired by the acquisition procedure, a prediction procedure for predicting the evaluation of the one user in the predetermined service, and a prediction procedure.
A prediction program characterized by having a computer execute. It ’s a computer-based prediction method.
The acquisition process to acquire behavioral information about one user on the Internet,
Based on the model generated by the behavior information about another user on the Internet, the evaluation of the other user in the predetermined service, and the behavior information about one user acquired by the acquisition process, the predetermined A prediction process that predicts the evaluation of the one user in the service,
An estimation step of estimating the certainty of evaluation in the predetermined service predicted by the model based on the behavior information about the other user.
Only including,
The estimation process is
Based on the confidence assigned to each of the behavioral information about the other user, the confidence regarding the evaluation in the predetermined service predicted by the model is estimated.
A prediction method characterized by that. Acquisition procedure to acquire behavior information about one user on the Internet,
The predetermined behavior information about another user on the Internet, the model generated by the evaluation of the other user in the predetermined service, and the behavioral information about one user acquired by the acquisition procedure. A prediction procedure that predicts the evaluation of the one user in the service,
An estimation procedure for estimating the certainty of evaluation in the predetermined service predicted by the model based on the behavior information regarding the other user, and an estimation procedure.
Let the computer run
The estimation procedure is
Based on the confidence assigned to each of the behavioral information about the other user, the confidence regarding the evaluation in the predetermined service predicted by the model is estimated.
A prediction program characterized by that.

Images