JP6661737B2 - Information processing apparatus and information processing method - Google Patents

Description

  Embodiments described herein relate generally to an information processing apparatus and an information processing method for estimating a funnel structure of a user on the Internet or the like.

  The user has a predetermined intent with respect to the conversion of a predetermined advertisement distributed on the Web, and belongs to various layers (cognition, interest / interest, examination, immediately before, etc.) based on the intent and the like. The distribution of users in each hierarchy has a funnel structure. As an advertiser's point of view, it is important to correctly evaluate the value of the delivered advertisement and to optimize the structure of the advertisement such as the content and volume of the advertisement, and it is necessary to know the funnel structure as a premise .

  However, conventionally, an analog method such as a questionnaire has been used to visualize the funnel structure. Analog methods such as questionnaires are costly and difficult to obtain a sufficient amount of data. Also, the questionnaire does not always provide objective information, such as including a response based on misidentification or a response to show oneself better.

  Therefore, a technique for estimating the user hierarchy in the funnel structure from an access log on the Internet or the like is required.

  In addition, the required information of the user differs depending on the layer to which the user belongs. Therefore, a technique for distributing different advertisements for each hierarchy is required.

JP, 2014-512613, A

  In order to solve the above-described problem, an information processing apparatus and an information processing method capable of distributing an advertisement according to a hierarchy of a user are provided.

According to the embodiment, the information processing apparatus includes a conversion the user log extracting unit, a path extractor, a setting unit, an estimation unit, a. The conversion user log extraction unit extracts a conversion user log of a user who has completed the conversion. The path extraction unit extracts, based on the conversion user log, an action performed by the user who has completed the conversion until the user reaches the conversion and a transition path of the action. The setting unit sets an action ratio of the action for each layer of the funnel structure based on the action extracted by the path extraction unit and the transition path of the action. The estimating unit estimates a user hierarchy of a predetermined user terminal based on the action ratio. The conversion user log is a log relating to at least one of the user's location, radio listening, healthcare information, purchase, application use, television viewing, or operation of a device owned by the user.

  According to the embodiment, the information processing apparatus can model the funnel hierarchy (generate the hierarchy table and the transfer probability table) based on the action log of the user who has completed the conversion. Further, the information processing apparatus can estimate the hierarchy of the user who has not completed the conversion by modeling the funnel structure.

  That is, the information processing apparatus can estimate the number of users for each tier and the transfer probability between tiers. Therefore, from the above estimation, it is possible to estimate the volume of each tier and which tier will be expanded or reduced in the future.

  Further, the information processing apparatus can estimate a funnel structure of a user who has viewed the advertisement and a funnel structure of a user who has not viewed the advertisement. Therefore, the information processing device can present the effect of the advertisement as the difference between the two funnel structures. For example, if the number of users in a specific hierarchy of the funnel structure of the user who viewed the advertisement increases and the hierarchy immediately above (for example, if the specific hierarchy is “interest,“ cognition ”) decreases, Is presumed to have the effect of transitioning the user from the layer immediately above to the specific layer.

  Further, the information processing device estimates both funnel structures after the advertisement distribution. Assume that both funnels are affected by factors other than advertising (for example, TV commercials), but that the effects of both funnels are not different (or the difference is sufficiently small) Can be. Therefore, the difference between the two funnels is not affected by factors other than the advertisement. Therefore, it can be determined that the difference between the two funnels results from browsing the advertisement. As a result, the effect of the advertisement can be grasped more clearly.

  Further, the information processing device can measure the effect of each layer by randomly distributing a plurality of advertisements. As a result, the information processing device can determine an effective advertisement for each layer.

  Further, upon receiving the advertisement request from the user terminal, the information processing apparatus estimates the user hierarchy of the user terminal. The information processing device distributes an advertisement corresponding to the estimated hierarchy to the user terminal. As a result, the information processing device can deliver an optimal advertisement to the user.

FIG. 1 is a diagram illustrating a configuration example of a funnel structure according to the embodiment. FIG. 2 is a block diagram illustrating a configuration example of the information processing apparatus according to the embodiment. FIG. 3 is a diagram illustrating an example of category transition according to the embodiment. FIG. 4 is a diagram illustrating an example of category transition according to the embodiment. FIG. 5 is a hierarchy table according to the embodiment. FIG. 6 is a transfer probability table according to the embodiment. FIG. 7 shows a transition example of the funnel structure according to the embodiment. FIG. 8 is a flowchart illustrating an operation example of the information processing apparatus according to the embodiment. FIG. 9 is a flowchart illustrating an operation example of the information processing apparatus according to the embodiment. FIG. 10 is a flowchart illustrating an operation example of the information processing apparatus according to the embodiment. FIG. 11 is a flowchart illustrating an operation example of the information processing apparatus according to the embodiment. FIG. 12 is a flowchart illustrating an operation example of the information processing apparatus according to the embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
First, the funnel structure will be described.
FIG. 1 shows a configuration example of the funnel structure 1.

The funnel structure 1 indicates the number of users for each tier associated with the conversion of a given advertisement.
Conversions are the end result of an advertiser performing an ad. For example, the conversion is a purchase of a predetermined article, a request for materials, a browsing of a predetermined site, or a visit to a predetermined store. The content of the conversion is not limited to a specific configuration.
The advertisement is presented to the user through the user terminal.

  As shown in FIG. 1, the funnel structure 1 includes layers 2 to 6 and the like. The layers 2 to 6 are, for example, “cognition”, “interest / interest”, “examination”, “immediately before”, and “conversion”, respectively. In the example shown in FIG. 1, it is shown that the larger the area of the hierarchy is, the more users are. Note that the funnel structure 1 does not need to include “conversion” as a hierarchy.

  “Recognition” is a hierarchy that recognizes matters related to results. For example, when the conversion is a purchase of a predetermined product, “recognition” is a hierarchy in which the product is recognized as a matter related to the result. Further, when the conversion is browsing of the event notification site, “recognition” is a hierarchy in which the event or the like is recognized as a matter related to the result.

  The “interest / interest” is a hierarchy having an interest or interest in matters related to the result. In other words, “interest / interest” is a hierarchy in which there is a certain level of interest in matters related to results, although the degree of conversion is not considered.

  “Consideration” is a hierarchy in which conversion is considered. In other words, “examination” is a hierarchy that considers performing conversion beyond mere interest.

  “Before” is the level immediately before performing the conversion. In other words, “immediately before” is the level at which the examination of the conversion is completed and the conversion is determined to be performed.

  “Conversion” is the level at which conversion has been completed.

  In the example shown in FIG. 1, the number of users decreases as the hierarchy deepens as “cognition”, “interest / interest”, “review”, “immediately”, and “conversion”.

  Note that the funnel structure may further include another hierarchy. Further, the funnel structure may not have the above-described hierarchy. Further, the number of layers of the funnel structure is not limited to a specific number. The configuration of the funnel structure is not limited to a specific configuration.

Next, an information processing apparatus according to the embodiment will be described.
FIG. 2 is a block diagram illustrating a configuration example of the information processing apparatus 10.
In the configuration example shown in FIG. 2, the information processing device 10 includes a CPU 11, a ROM 12, a RAM 13, a NVM 14, a communication unit 15, an operation unit 16, a display unit 17, and the like as a basic configuration. These units are connected to each other via a data bus. In addition, the information processing apparatus 10 may include a configuration as necessary in addition to the configuration illustrated in FIG. 2 or may exclude a specific configuration.

  The CPU 11 has a function of controlling the operation of the entire information processing device 10. The CPU 11 may include an internal cache and various interfaces. The CPU 11 implements various processes by executing programs stored in the internal memory, the ROM 12, or the NVM 14 in advance.

  A part of various functions realized by the CPU 11 executing the program may be realized by a hardware circuit. In this case, the CPU 11 controls a function executed by the hardware circuit.

  The ROM 12 is a nonvolatile memory in which a control program, control data, and the like are stored in advance. The control program and control data stored in the ROM 12 are incorporated in advance according to the specifications of the information processing device 10. The ROM 12 stores, for example, a program (for example, BIOS) for controlling the circuit board of the information processing device 10 and the like.

  The RAM 13 is a volatile memory. The RAM 13 temporarily stores data being processed by the CPU 11 and the like. The RAM 13 stores various application programs based on instructions from the CPU 11. Further, the RAM 13 may store data necessary for executing the application program, an execution result of the application program, and the like.

  The NVM 14 is a nonvolatile memory in which data can be written and rewritten. The NVM 14 includes, for example, a hard disk, an SSD, an EEPROM (registered trademark), a flash memory, or the like. The NVM 14 stores a control program, an application, and various data according to the operation purpose of the information processing device 10.

  Further, the NVM 14 includes a storage area 14a for storing an action log, which is a record of a user browsing the site, and a storage area 14b for storing an effect log indicating an effect of an advertisement. The action log and the effect log will be described later.

  The communication unit 15 is an interface for transmitting and receiving data to and from an external device via a communication network such as the Internet. The communication unit 15 is, for example, an interface that supports LAN connection.

  The communication unit 15 transmits and receives data to and from the user terminal 20 via a communication network. The user terminal 20 is a terminal used by the user. For example, the user terminal 20 is a desktop PC, a notebook PC, a tablet PC, a smartphone, or the like. The user terminal 20 requests an advertisement from the information processing device 10 and displays the advertisement.

  Various operation instructions are input to the operation unit 16 by an operator of the information processing apparatus 10. The operation unit 16 transmits an operation instruction signal input to the operator to the CPU 11. The operation unit 16 is, for example, a keyboard, a numeric keypad, and a touch panel.

  The display unit 17 is a display device that displays various information under the control of the CPU 11. The display unit 17 is, for example, a liquid crystal monitor. When the operation unit 16 is configured by a touch panel or the like, the display unit 17 may be formed integrally with the operation unit 16.

Next, the action log will be described.
The action log is a log related to the action of the user. For example, the action log stores information indicating the action in association with the date and time when the action was performed.

  For example, the actions targeted by the action log include Web access, user location, content listening, healthcare information, purchasing, posting to SNS, application use, content viewing, or operation of a device owned by the user. .

  The Web access is, for example, a search keyword, a browsing site, text in a content, an image or a moving image, a click, a mouse over, or a heat map.

  The location of the user is, for example, GPS information of the smartphone or the like possessed by the user, an IP address of the smartphone, a visit to the store by a beacon, or entry / exit of an automatic ticket gate.

The content listening is, for example, music listening or radio listening.
The health care information is information on the body of the user. The health care information is, for example, exercise, blood pressure, or heart rate.

Purchasing is the purchase of something that does not convert. For example, the purchase is payment with electronic money or purchase at an EC site.
Posting to the SNS is, for example, posting to a SNS site such as text, an image, or a moving image.

The use of the application is, for example, download of the application, activation of the application, use of the application, or deletion of the application.
The content viewing is, for example, watching a television or watching a moving image on a moving image site.

The operation of the device possessed by the user is, for example, the operation of a smart home appliance, the driving of a car, the automatic driving, or the communication between devices.
The action targeted by the action log is not limited to a specific configuration.

Next, functions realized by the CPU 11 will be described.
First, the CPU 11 has a function of extracting an action log (conversion user log) of a user who has completed conversion (a conversion user log extraction unit).
The CPU 11 extracts, from the storage area 14a, an action log of the user who has completed the conversion, the action log being retroactive for a predetermined period from the date and time when the conversion has been completed. For example, the CPU 11 extracts an action log 30 days, 60 days or 90 days before the completion of the conversion. The CPU 11 may determine the period to go back based on the conversion. For example, the CPU 11 may set a relatively long period for a conversion that is considered to take a long time to make a decision (for example, purchasing a high-priced product). Further, for example, the CPU 11 may set a relatively short period for a conversion that can be determined in a short period (for example, purchase of a low-priced product). The method by which the CPU 11 determines the period during which the action log is extracted is not limited to a specific method.

The CPU 11 has a function of specifying a user action that has completed conversion and a path along which the action transitions based on the extracted action log (path extracting unit).
For example, the CPU 11 sets a category for the extracted action log.
The category is set based on the characteristics of the action. The category may be manually set in advance for each action. The category may be automatically set by the CPU 11 according to a predetermined algorithm. The method of setting the category in the action log is not limited to a specific method.

  For example, the category is a set of Web sites browsed by the user. For example, a category is a set of websites related to specific content. For example, the category is “usually used media” or “related product information”.

  “Usually used media” is a site that users regularly use. For example, “usually used media” includes a search engine or a portal site.

  “Related product information” is a site that is strongly related to conversion. For example, when the conversion is the purchase of a product, the “related product information” is a criticism site of the product or a site for introducing a product similar to the product.

  A category is a set of locations of users. For example, the category indicates a specific area where the user is located, such as a predetermined store, building, or park. For example, the category is “convenience store”.

  “Convenience store” indicates that the user has entered a specific or arbitrary convenience store.

  The category is a set related to the health care of the user. For example, the category is a classification related to blood pressure. For example, the category is “high blood pressure”, “low blood pressure” or “average blood pressure”.

“High blood pressure” indicates that the user has a blood pressure value higher than a predetermined threshold.
“Low blood pressure” indicates that the user has a blood pressure value lower than a predetermined threshold.
“Average blood pressure” indicates that the user's blood pressure value does not belong to any of “high blood pressure” and “low blood pressure”.
Note that the CPU 11 need not set a category in the action log. In addition, the CPU 11 does not have to set a category for some action logs and set a category for other action logs.

  Next, the CPU 11 extracts what kind of action the user who has completed the conversion has changed and led to the conversion. That is, the CPU 11 extracts an action that is characteristically performed by the user who has completed the conversion, and extracts a transition path of the action.

  For example, the CPU 11 extracts an action (or a category) that is specifically performed by the user who has completed the conversion. Further, the CPU 11 calculates a transition probability between the actions. That is, the CPU 11 calculates the probability of performing another action after the user who has completed the conversion performs one action.

  The CPU 11 sets a path as a transition route between the actions based on the transition probability between the actions. For example, the CPU 11 may set a path between actions whose transition probability is higher than a predetermined threshold. Further, the CPU 11 may set a path having a different thickness between the actions (that is, a path having a different strength) according to the transition probability.

  For example, the CPU 11 may use a Bayesian network to extract an action performed by the user who has completed the conversion, and may extract a transition path of the action leading to the conversion.

3 and 4 show transition examples of the behavior of the user who has completed the conversion.
Here, it is assumed that the CPU 11 has extracted the actions 21 to 25. Actions 21 to 24 are categories. The actions 21 to 24 are “usually used media”, “high blood pressure”, “convenience store”, and “related product information”, respectively. The action 25 is “app DL”.

The “usually used media”, “high blood pressure”, “convenience store”, and “related product information” are as described above.
“App DL” indicates an action of downloading a specific application.

  3 and 4, arrows indicate paths. That is, the arrow indicates that the user makes a transition from the action at the start point of the arrow to the action at the end point of the arrow.

  As shown in FIG. 3 and FIG. 4, one of the transition paths leading to the conversion by the user is a path that follows action 21 (“usually used media”), action 24 (“related product information”), and “conversion” in order. is there. Another one of the transition paths leading to the conversion of the user follows action 21 (“usually used media”), action 23 (“convenience store”), action 24 (“related product information”), and “conversion” in this order. It is a route.

  Another one of the transition paths leading to the conversion of the user is a path that sequentially follows action 21 (“usually used media”), action 23 (“convenience store”), and “conversion”. Another one of the transition paths leading to the conversion of the user is action 21 (“usually used media”), action 22 (“high blood pressure”), action 23 (“convenience store”), action 24 (“related product information”). ) And “conversion”.

  Further, another one of the transition paths leading to the conversion by the user is a path that sequentially follows action 21 (“usually used media”), action 22 (“high blood pressure”), action 23 (“convenience store”), and “conversion”. is there.

  Further, another one of the transition paths leading to the conversion by the user is a path that sequentially follows action 21 (“usually used media”), action 22 (“high blood pressure”), action 25 (“app DL”), and “conversion”. It is.

  Note that the number of actions extracted by the CPU 11 is not limited to a specific number. The configuration of the action extracted by the CPU 11 is not limited to a specific configuration. The path extracted by the CPU 11 is not limited to a specific configuration.

  Further, the CPU 11 has a function of setting an action ratio of the extracted action (or category) for each hierarchy of the funnel structure (setting unit). The action ratio is a ratio of users who perform the extracted action in the hierarchy or a probability that each user takes the action in the hierarchy.

For example, the CPU 11 generates a table (hierarchy table) in which the hierarchy of the funnel structure is associated with the action ratio of each action in the hierarchy.
FIG. 5 shows a configuration example of a hierarchy table in which the hierarchy of the funnel structure is associated with the action ratio of the action in the hierarchy.

  As shown in FIG. 5, in the hierarchy table, in “recognition”, the action ratios of “usually used media”, “high blood pressure”, “convenience store”, “related product information”, and “app DL” are 77% and 10%, respectively. , 7%, 3% and 3%. For example, the hierarchy table indicates that 77% of the users belonging to “cognition” perform an action belonging to “usually used media”.

  Further, in the hierarchy table, in “interest / interest”, the action ratios of “usually used media”, “high blood pressure”, “convenience store”, “related product information”, and “application DL” are 45%, 15%, and 17%, respectively. , 15% and 8%.

  Also, in the hierarchy table, in the “examination”, the action ratios of “usually used media”, “high blood pressure”, “convenience store”, “related product information”, and “app DL” are 10%, 17%, 35%, and 28%, respectively. % And 10%.

  In the hierarchy table, the action ratios of “usually used media”, “high blood pressure”, “convenience store”, “related product information”, and “app DL” are “6%, 19%, 25%, and 35%”, respectively, in “immediately before”. % And 15%.

  The actions set by the CPU 11 for each layer and the ratio of the actions are not limited to specific values. That is, the configuration of the hierarchy table set by the CPU 11 is not limited to a specific configuration.

  In addition, the CPU 11 has a function of calculating a probability that a user belonging to each tier will move to each tier after a predetermined period based on the behavior and a transition path between the behaviors (transition probability) (transfer probability calculation unit). That is, the CPU 11 calculates a transfer probability between tiers. For example, the CPU 11 generates a table (relocation probability table) in which the hierarchy of the funnel structure is associated with the probability that a user in the hierarchy will move to each hierarchy after a predetermined period. Note that the transfer probability table also includes the probability that the user will stay in the hierarchy.

  FIG. 6 shows a configuration example of a transfer probability table in which the layers of the funnel structure are associated with the probabilities that users in the layers will transfer to each layer after a predetermined period.

  As shown in FIG. 6, the transfer probability table shows that the users belonging to “recognition” have 80%, “2”, “recognition”, “interest / interest”, “review”, “immediately before”, and “conversion” after a predetermined period, respectively. %, 3%, 15%, and 0%. In addition, the transfer probability table shows that users belonging to “interest / interest” are 81%, 5%, 14%, and 0% for “interest / interest”, “examination”, “immediately before”, and “conversion” after a predetermined period, respectively. Indicates that they will be relocated. Further, the transfer probability table indicates that the user belonging to “review” transfers to “review”, “immediately before”, and “conversion” after a predetermined period with a probability of 79%, 21%, and 0%, respectively. Further, the transfer probability table indicates that the user belonging to “immediately before” moves to “immediately before” and “conversion” after a predetermined period with a probability of 87% and 13%, respectively.

  Note that the probability that the CPU 11 calculates for each hierarchy is not limited to a specific value. That is, the configuration of the transfer probability table calculated by the CPU 11 is not limited to a specific configuration.

For example, the CPU 11 generates a hierarchy table and a transfer probability table as follows.
The CPU 11 assumes the number of layers until “conversion” and the probability of transfer from each layer to the next layer. According to the assumption, the CPU 11 continuously evaluates which layer a certain user belongs to, and the probability of transition to the next layer, and generates a layer table and a transfer probability table.

For example, the CPU 11 generates a hierarchy table and a transfer probability table according to a hidden Markov model or the like.
Note that the method of generating the hierarchy table and the transfer probability table is not limited to a specific method.

The CPU 11 has a function of extracting an action log of a user who has not completed conversion (unconverted user log).
The CPU 11 extracts an action log of a user who has not completed the conversion from the storage area 14a. Note that the CPU 11 may further extract an action log within a predetermined period from the action log of the user who has not completed the conversion.

  Further, the CPU 11 has a function of estimating the tier of each user who has not completed conversion from the action log and the tier table of the user who has not completed conversion.

The CPU 11 estimates a hierarchy for each user based on the extracted action log. For example, the CPU 11 extracts an action log including the action indicated by the hierarchy table from the action log of a certain user. The CPU 11 matches the action indicated by the extracted action log with each action ratio in the hierarchy table. The CPU 11 estimates the hierarchy that most closely matches the behavior of the user as the hierarchy to which the user belongs.
The CPU 11 executes the above-described operation for each user, and estimates the hierarchy of each user.

  Further, the CPU 11 has a function of estimating the funnel structure from the hierarchy of each user. For example, the CPU 11 counts the number of users for each layer, and estimates the number of users belonging to each layer as a funnel structure. Further, for example, the CPU 11 may estimate the funnel structure at a ratio for each hierarchy. Note that the CPU 11 may add “conversion”, which is the hierarchy of the user who has completed the conversion, to the funnel structure.

  Further, the CPU 11 has a function of estimating a funnel structure formed by a user who has viewed a predetermined advertisement and a funnel structure formed by a user who has not viewed the advertisement. Here, it is assumed that the advertiser has distributed a predetermined advertisement on a predetermined Web. FIG. 7 shows a configuration example of a funnel structure 1a of a user who has viewed the advertisement and a funnel structure 1b of a user who has not viewed the advertisement.

  The CPU 11 extracts, from the storage area 14a, an action log (viewing user log) of the user who has not completed the conversion and has viewed the advertisement. For example, the CPU 11 extracts a user having an action log indicating that the advertisement has been viewed, and extracts an action log of the user.

  The CPU 11 estimates the hierarchy of each user who has viewed the advertisement from the extracted action log and the hierarchy table. The CPU 11 estimates the funnel structure 1a of the user who has viewed the advertisement based on the estimation result. Note that the CPU 11 may add, to the funnel structure 1a, the hierarchy of the user who has completed the conversion after browsing the advertisement as the “conversion” hierarchy.

  Further, the CPU 11 extracts, from the storage area 14a, an action log (a non-viewing user log) of a user who has not completed the conversion and has not viewed the advertisement. For example, the CPU 11 extracts a user having an action log indicating that the user has viewed the advertisement, and extracts an action log other than the user.

  The CPU 11 estimates the hierarchy of each user who has not viewed the advertisement from the extracted user action log and the hierarchy table. The CPU 11 estimates the funnel structure 1b of the user who has not viewed the advertisement based on the estimation result. Note that the CPU 11 may add, to the funnel structure 1b, the hierarchy of the user who has completed the conversion without viewing the advertisement as the hierarchy of “conversion”.

  By the above operation, as shown in FIG. 7, after the advertisement distribution, the CPU 11 can estimate the funnel structure 1a of the user who has viewed the advertisement and the funnel structure 1b of the user who has not viewed the advertisement. Note that the CPU 11 may classify the user according to the number of times of viewing the advertisement and estimate the funnel structure for each number of times of viewing.

Further, the CPU 11 has a function of acquiring an effect log for each advertisement (effect log acquisition unit).
The effect log is a log indicating the effect of the advertisement. For example, the effect log includes a hierarchy of the user who presented the advertisement, an advertisement ID for specifying the advertisement, and effect information indicating the effect. The effect information indicates the effect of the advertisement. For example, the effect information is information indicating whether the purpose of the advertisement has been achieved. For example, the effect information is information indicating whether a conversion corresponding to the advertisement has been performed. Further, the effect information may be information indicating whether or not the advertisement is clicked.

  The CPU 11 randomly distributes a plurality of advertisements to the user. For example, the CPU 11 receives an advertisement request from the user terminal 20 via the communication unit 15. The advertisement request may be transmitted, for example, when the user terminal 20 executes a predetermined html or the like.

When receiving the request from the user terminal 20, the CPU 11 estimates the hierarchy of the user of the user terminal 20. When estimating the user hierarchy, the CPU 11 randomly selects one advertisement from a plurality of advertisements (selection unit). For example, the plurality of advertisements may be advertisements having different elements such as a display surface, a format, a size, or a creative. For example, the CPU 11 may select one advertisement from a plurality of advertisements in which elements such as a publication surface, a format, a size, and a creative are individually controlled.
The CPU 11 transmits the advertisement selected as a response to the request to the user terminal 20 (selected advertisement transmission unit).

  The CPU 11 monitors the operation of the user for a predetermined period after transmitting the advertisement to the user terminal 20 (monitoring unit). For example, based on a signal from the user terminal 20, the CPU 11 monitors behavior such as whether the user has clicked an advertisement or converted from the advertisement. For example, the CPU 11 acquires an action log indicating the action of the user after transmitting the advertisement to the user terminal 20. The CPU 11 generates effect information based on the monitoring result.

  The CPU 11 generates an effect log (effect log generator). For example, the CPU 11 generates an effect log that stores the hierarchy of the user, an advertisement ID that specifies the advertisement delivered to the user, and the generated effect information. Note that the effect log may further store a user ID or the like for specifying the user. The CPU 11 stores the generated effect log in the storage area 14b.

Next, the CPU 11 has a function of determining an effective advertisement (advertisement corresponding to each tier) for each tier based on the effect log (optimization unit).
For example, the CPU 11 acquires a plurality of effect logs from the storage area 14b. When acquiring a plurality of effect logs, the CPU 11 classifies the effect logs for each hierarchy. The CPU 11 calculates a performance rate for each advertisement based on the effect information of the effect log classified into a certain hierarchy. The performance rate is a rate at which the purpose of the advertisement is achieved with respect to the number of displayed advertisements. For example, the performance rate is a rate at which an advertisement is clicked (click rate) or a rate at which conversion is achieved (conversion rate). Further, the achievement rate may be a probability (transition rate) that the user's hierarchy changes after the advertisement is clicked.

For example, the CPU 11 counts the number of effect logs classified into the hierarchy and the number of effect logs including effect information indicating that the purpose has been achieved. The CPU 11 calculates the advertisement performance rate from the two numbers.
The CPU 11 determines an advertisement with the highest performance rate as an effective advertisement for the hierarchy.
The CPU 11 performs the same operation for each layer, and determines an effective advertisement for each layer.

Further, the CPU 11 has a function of distributing an advertisement according to a hierarchy of users (for example, users who have not completed conversion).
For example, the CPU 11 receives a request from the user terminal 20. Upon receiving the request, the CPU 11 estimates the hierarchy of the user who transmitted the request (estimating unit). For example, the CPU 11 acquires an action log of the user. The CPU 11 determines the hierarchy of the user by matching the behavior indicated by the behavior log with each behavior ratio in the hierarchy table. The estimation of the hierarchy is as described above.

  Note that the CPU 11 may determine the hierarchy of each user in advance and store it in the NVM 14. The CPU 11 may determine the hierarchy of the user who transmitted the request with reference to the NVM 14.

  The CPU 11 determines an advertisement according to the determined user hierarchy (advertisement determination unit). For example, the NVM 14 stores a hierarchy and an advertisement in advance in association with each other. The CPU 11 refers to the NVM 14 to acquire an advertisement according to the determined user hierarchy. The CPU 11 transmits the acquired advertisement to the user terminal 20 via the communication unit 15 (determined advertisement transmission unit).

  The advertisement associated with the hierarchy may be an effective advertisement determined based on the effect log. Further, the advertisement associated with the hierarchy may be an advertisement determined by the operator.

Next, an operation example of the information processing apparatus 10 will be described.
First, an operation example in which the information processing apparatus 10 estimates a funnel structure will be described.
FIG. 8 is a flowchart illustrating an operation example in which the CPU 11 of the information processing apparatus 10 estimates a funnel structure.

  First, the CPU 11 extracts, from the storage area 14a, an action log of a user who has completed the conversion retroactively for a predetermined period (S11). When the action log of the user who has completed the conversion is extracted, the CPU 11 extracts the action (or category) and the transition path between the actions based on the extracted action log (S12).

  When the action and the transition path between the actions are extracted, the CPU 11 generates a hierarchy table based on the action and the transition path between the actions (S13). After generating the hierarchy table, the CPU 11 generates a transfer probability table based on the behavior and the transition route between the behaviors (S14).

  After generating the transfer probability table, the CPU 11 extracts the action log of the user who has not completed the conversion (S15). When the action log of the user who has not completed the conversion is extracted, the CPU 11 estimates the tier of each user who has not completed the conversion based on the extracted action log and the tier table (S16).

  After estimating the hierarchy of each user who has not completed the conversion, the CPU 11 estimates the funnel structure based on the estimation result (S17). After estimating the funnel structure, the CPU 11 ends the operation.

  Next, an operation example in which the information processing apparatus 10 estimates a funnel structure of a user who has viewed the advertisement and a funnel structure of a user who has not viewed the advertisement will be described.

FIG. 9 is a flowchart illustrating an operation example in which the CPU 11 of the information processing apparatus 10 estimates a funnel structure of a user who has viewed the advertisement and a funnel structure of a user who has not viewed the advertisement.
Here, it is assumed that the CPU 11 has generated the hierarchy table.

  First, the CPU 11 extracts, from the storage area 14a, an action log of a user who has not completed the conversion and has viewed the advertisement (S21). After extracting the action log, the CPU 11 estimates the hierarchy of each user who has viewed the advertisement based on the hierarchy table (S22).

  After estimating the hierarchy of each user who has viewed the advertisement, the CPU 11 estimates the funnel structure of the user who has viewed the advertisement based on the estimation result (S23). When estimating the funnel structure of the user who has viewed the advertisement, the CPU 11 extracts, from the storage area 14a, the action log of the user who has not completed the conversion and has not viewed the advertisement (S24).

  After extracting the action log, the CPU 11 estimates the hierarchy of each user who has not viewed the advertisement based on the hierarchy table (S25). After estimating the hierarchy of each user who has not viewed the advertisement, the CPU 11 estimates the funnel structure of the user who has not viewed the advertisement (S26).

  After estimating the funnel structure of the user who has not viewed the advertisement, the CPU 11 ends the operation.

Next, an operation example in which the information processing apparatus 10 acquires an effect log will be described.
FIG. 10 is a flowchart illustrating an operation example in which the CPU 11 of the information processing apparatus 10 acquires an effect log.

  First, the CPU 11 determines whether an advertisement request has been received from the user terminal 20 via the communication unit 15 (S31). When determining that the request has not been received (S31, NO), the CPU 11 returns to S31.

  When determining that the request has been received (S31, YES), the CPU 11 estimates the hierarchy of the user of the user terminal 20 that has transmitted the request (S32). When estimating the user hierarchy, the CPU 11 selects one advertisement at random from a plurality of advertisements (S33).

  When one advertisement is selected, the CPU 11 transmits the selected advertisement to the user terminal 20 via the communication unit 15 (S34). When transmitting the selected advertisement to the user terminal 20, the CPU 11 monitors the operation of the user (S35).

  When monitoring the operation of the user, the CPU 11 generates an effect log based on the hierarchy of the user determined in S32, the advertisement ID of the advertisement selected in S33, and the monitoring result in S35 (S36).

  After generating the effect log, the CPU 11 stores the generated effect log in the storage area 14b (S37). After storing the effect log, the CPU 11 ends the operation.

Next, an operation example in which the information processing device 10 determines an effective advertisement will be described.
FIG. 11 is a flowchart illustrating an operation example in which the CPU 11 of the information processing device 10 determines an effective advertisement.

  First, the CPU 11 acquires a plurality of effect logs from the storage area 14b (S41). Upon acquiring a plurality of effect logs, the CPU 11 sets a hierarchy for determining an effective advertisement (S42).

  When a hierarchy is set, the CPU 11 calculates a performance rate for each advertisement in the hierarchy (S43). After calculating the performance rate for each advertisement, the CPU 11 determines an effective advertisement for the hierarchy based on the performance rate for each advertisement (S44).

  When determining an effective advertisement, the CPU 11 determines whether there is another hierarchy for determining an effective advertisement (S45). If it is determined that there is another hierarchy that determines an effective advertisement (S45, YES), the CPU 11 returns to S43. When determining that there is no other hierarchy for determining an effective advertisement, the CPU 11 ends the operation.

Next, an operation example in which the information processing apparatus 10 distributes an advertisement according to the user's hierarchy will be described.
FIG. 12 is a flowchart for explaining an operation example in which the CPU 11 of the information processing apparatus 10 distributes an advertisement according to a user hierarchy.

  First, the CPU 11 determines whether an advertisement request has been received from the user terminal 20 via the communication unit 15 (S51). If it is determined that the request has not been received (S51, NO), the CPU 11 returns to S51.

  When determining that the request has been received (S51, YES), the CPU 11 estimates the hierarchy of the user of the user terminal 20 that has transmitted the request (S52). After estimating the user's hierarchy, the CPU 11 determines an advertisement corresponding to the hierarchy (S53).

  When the advertisement corresponding to the hierarchy is determined, the CPU 11 transmits the determined advertisement to the user terminal 20 via the communication unit 15 (S54). When the determined advertisement is transmitted to the user terminal 20, the CPU 11 ends the operation.

  Note that the CPU 11 may acquire an action log or an effect log from an external device. Further, the CPU 11 may display data indicating the number of users in each layer on the display unit 17 as a funnel structure. Further, the CPU 11 may transmit data indicating the funnel structure to the external device through the communication unit 15 or the like.

The present invention is not limited to the above-described embodiment, and can be variously modified in an implementation stage without departing from the gist of the invention. In addition, the embodiments may be implemented in appropriate combinations as much as possible, in which case the combined effects can be obtained. Furthermore, the above embodiments include inventions at various stages, and various inventions can be extracted by appropriate combinations of a plurality of disclosed constituent elements. For example, even if some components are deleted from all the components shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved and the effects described in the column of the effect of the invention can be solved. Is obtained, a configuration from which this configuration requirement is deleted can be extracted as an invention.
Hereinafter, the inventions described in the claims at the beginning of the present application are additionally described.
[C1]
A conversion user log extraction unit that extracts a conversion user log of a user who has completed conversion,
Based on the conversion user log, a path extraction unit that extracts an action performed by the user who has completed the conversion until the conversion and a transition path of the action,
A setting unit that sets an action ratio of the action for each hierarchy of a funnel structure based on the action extracted by the path extraction unit and the transition path of the action;
A communication unit for transmitting and receiving data;
Upon receiving an advertisement request from a user terminal through the communication unit, an estimation unit that estimates a hierarchy of users of the user terminal based on the activity ratio,
An advertisement determining unit that determines an advertisement corresponding to the hierarchy,
A determined advertisement transmitting unit that transmits the advertisement through the communication unit,
An information processing apparatus comprising:
[C2]
further,
Based on the behavior extracted by the route extraction unit and the transition path of the behavior, a transfer probability calculation unit that calculates a transfer probability that a user belonging to each hierarchy of the funnel structure moves to each hierarchy,
The information processing device according to C1.
[C3]
The conversion user log extraction unit extracts, as the conversion user log, an action log dating for a predetermined period from a date and time when the conversion is completed,
The information processing device according to C1 or 2.
[C4]
The estimating unit estimates the hierarchy of the user by matching the action indicated by the action log of the user with the action ratio,
The information processing device according to any one of C1 to C3.
[C5]
An effect log acquisition unit that acquires an effect log indicating the effect of the advertisement,
An optimization unit that determines an advertisement corresponding to each tier based on the effect log;
With
The advertisement determining unit determines the advertisement determined by the optimization unit as an advertisement corresponding to the hierarchy,
The information processing apparatus according to any one of C1 to C4.
[C6]
The effect log includes effect information indicating whether the purpose of the advertisement has been achieved,
The optimization unit calculates, for each advertisement, a performance rate, which is a rate at which the purpose of the advertisement is achieved with respect to the number of displayed advertisements, based on the effect information of the effect log classified into a certain hierarchy, and Determining an advertisement corresponding to the hierarchy based on the
The information processing device according to C5.
[C7]
The performance rate is at least one of a click rate, a conversion rate, and a transition rate of the advertisement.
The information processing device according to C6.
[C8]
A selection unit that selects one advertisement from a plurality of advertisements when receiving the request through the communication unit;
A selected advertisement transmitting unit that transmits the advertisement selected by the selecting unit through the communication unit;
After the selected advertisement transmitting unit transmits the advertisement, a monitoring unit that acquires an action log of a user of the user terminal,
Based on the action log acquired by the monitoring unit, an effect log generation unit that generates the effect log,
Comprising,
The information processing apparatus according to any one of C5 to C7.
[C9]
The conversion is at least one of purchase of a predetermined product, request of materials, browsing of a predetermined site, or visit to a predetermined store.
The information processing device according to any one of C1 to C8.
[C10]
Extract the conversion user log of the user who completed the conversion,
Based on the conversion user log, extract the action performed by the user who has completed the conversion until the conversion and the transition path of the action,
Based on the action and the transition path of the action, set an action ratio of the action for each hierarchy of the funnel structure,
Upon receiving an advertisement request from a user terminal, estimating a user hierarchy of the user terminal based on the activity ratio,
Determining an advertisement corresponding to the hierarchy;
Sending the advertisement,
Information processing method.

  1 (1a and 1b) ... funnel structure, 2 to 6 ... hierarchy, 10 ... information processing device, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... NVM, 14a and 14b ... storage area, 15 ... communication unit, 16 operation unit, 17 display unit, 20 user terminal.

Claims (10)

A conversion user log extraction unit that extracts a conversion user log of a user who has completed conversion,
Based on the conversion user log, a path extraction unit that extracts an action performed by the user who has completed the conversion until the conversion and a transition path of the action,
A setting unit that sets an action ratio of the action for each hierarchy of a funnel structure based on the action extracted by the path extraction unit and the transition path of the action ;
An estimation unit that estimates a user hierarchy a given user terminal based on the previous SL behavior ratio,
With
The conversion user log the user's whereabouts, radio listening, health care information, purchase, application use, television viewing or, Ru at least one related to log der of the operation of the device carried by the user,
Information processing device. further,
Based on the behavior extracted by the route extraction unit and the transition path of the behavior, a transfer probability calculation unit that calculates a transfer probability that a user belonging to each hierarchy of the funnel structure moves to each hierarchy,
The information processing apparatus according to claim 1. The conversion user log extraction unit extracts, as the conversion user log, an action log dating for a predetermined period from a date and time when the conversion is completed,
The information processing apparatus according to claim 1. The estimating unit estimates the hierarchy of the user by matching the action indicated by the action log of the user with the action ratio,
The information processing apparatus according to claim 1. A communication unit for transmitting and receiving data;
An advertisement determining unit that determines an advertisement corresponding to the hierarchy,
A determined advertisement transmitting unit that transmits the advertisement through the communication unit,
An effect log acquisition unit that acquires an effect log indicating the effect of the advertisement,
An optimization unit that determines an advertisement corresponding to each tier based on the effect log,
The advertisement determining unit determines the advertisement determined by the optimization unit as an advertisement corresponding to the hierarchy,
The information processing apparatus according to claim 1. The effect log includes effect information indicating whether the purpose of the advertisement has been achieved,
The optimization unit calculates, for each advertisement, a performance rate, which is a rate at which the purpose of the advertisement is achieved with respect to the number of displayed advertisements, based on the effect information of the effect log classified into a certain hierarchy, and Determining an advertisement corresponding to the hierarchy based on the
The information processing apparatus according to claim 5. The performance rate is at least one of a click rate, a conversion rate, and a transition rate of the advertisement.
The information processing apparatus according to claim 6. A selection unit that selects one advertisement from a plurality of advertisements when receiving an advertisement request from the user terminal through the communication unit;
A selected advertisement transmitting unit that transmits the advertisement selected by the selecting unit through the communication unit;
After the selected advertisement transmitting unit transmits the advertisement, a monitoring unit that acquires an action log of a user of the user terminal,
Based on the action log acquired by the monitoring unit, an effect log generation unit that generates the effect log,
Comprising,
The information processing apparatus according to claim 5. The conversion is at least one of purchase of a predetermined product, request of materials, browsing of a predetermined site, or visit to a predetermined store.
The information processing device according to claim 1. Extract the conversion user log of the user who completed the conversion,
Based on the conversion user log, extract the action performed by the user who has completed the conversion until the conversion and the transition path of the action,
Based on the action and the transition path of the action, set an action ratio of the action for each hierarchy of the funnel structure,
Estimating the hierarchy of the user of the predetermined user terminal based on the action ratio ,
The conversion user log the user's whereabouts, radio listening, health care information, purchase, application use, television viewing or, Ru at least one related to log der of the operation of the device carried by the user,
Information processing method.