JP7034197B2 - Information processing equipment, information processing methods, and programs - Google Patents

Description

The present invention relates to an information processing apparatus, an information processing method, and a program.

A technique is known in which a computer is made to select an ideal product to be displayed on a web page of an electronic commerce website (see, for example, Patent Document 1).

Special Table 2019-504406 Gazette

However, in the conventional technology, in electronic commerce, there are cases where the user cannot predict with sufficient accuracy the category of goods or services that are likely to lead to conversion.

The present invention has been made in view of the above problems, and provides an information processing device, an information processing method, and a program capable of accurately predicting a category of a product or service that is likely to lead to conversion by a user. I am aiming.

One aspect of the present invention has passed since the user finally reached the first conversion for the products or services classified into the above categories for each of the plurality of categories to which the products or services handled in the electronic commerce are classified. For the first hour, the first cumulative number of times the user has reached the first conversion in the past period for the goods or services classified in the category, and for the goods or services classified in the category. The second time that has passed since the user last reached the second conversion, and the second cumulative number of times that the user reached the second conversion in the past for a product or service classified in the category. Based on the first calculation unit that calculates, and the first hour, the first cumulative number of times, the second hour, and the second cumulative number of times calculated for each category by the first calculation unit. Based on the second calculation unit that calculates the score representing the user's preference for the category for each category and the score calculated for each category by the second calculation unit, the user has a time point in the future. It is an information processing apparatus including a prediction unit for predicting a conversion category, which is a category of goods or services expected to reach the first conversion or the second conversion.

According to one aspect of the present invention, it is possible to accurately predict the category of goods or services that users are likely to convert.

It is a figure which shows an example of the information processing system 1 including the information processing apparatus 100 in 1st Embodiment. It is a figure which shows an example of the web page provided by the service providing apparatus 20. It is a figure which shows an example of the structure of the information processing apparatus 100 in 1st Embodiment. It is a flowchart which shows the flow of the series of processing of the runtime by the information processing apparatus 100 in 1st Embodiment. It is a flowchart which shows the flow of the series of processing of the training by the information processing apparatus 100 in 1st Embodiment. It is a figure which shows an example of a layered category. It is a figure which shows an example of the hardware composition of the information processing apparatus 100 of an embodiment.

Hereinafter, the information processing apparatus, the information processing method, and the program to which the present invention is applied will be described with reference to the drawings.

[Overview]
The information processing device is realized by one or more processors. The information processing device calculates recency and frequency for each of a plurality of categories to which goods or services handled in electronic commerce are classified.

Recency is the time elapsed since the user last converted to a product or service in a category. Conversion means, for example, that a product or service is viewed, purchased, added to a shopping cart, added to favorites, reviewed or reviewed.

Frequency is the cumulative number of conversions a user has made to a product or service in a certain category over a period of time in the past.

When the information processing apparatus calculates the recency and frequency for each category, the information processing apparatus calculates a score representing the user's preference for the category for each category based on the recency and frequency. Then, the information processing device predicts the conversion category based on the calculated score. A conversion category is a category of goods or services that a user is expected (or expected) to convert at some point in the future. By calculating the score based on the recency and frequency in this way, it is possible to accurately predict the category of the product or service that the user is likely to convert. Hereinafter, the goods or services will be collectively referred to as "goods" and described.

<First Embodiment>
[overall structure]
FIG. 1 is a diagram showing an example of an information processing system 1 including an information processing apparatus 100 according to the first embodiment. The information processing system 1 in the first embodiment includes, for example, one or more terminal devices 10, a service providing device 20, and an information processing device 100. These devices are connected via the network NW. Further, a part of these devices may be included in another device as a virtual device. For example, a part or all of the functions of the service providing device 20 are realized by the functions of the information processing device 100. It may be a virtual machine.

Each device shown in FIG. 1 transmits and receives various information via the network NW. The network NW includes, for example, the Internet, a WAN (Wide Area Network), a LAN (Local Area Network), a provider terminal, a wireless communication network, a wireless base station, a dedicated line, and the like. It should be noted that not all combinations of the devices shown in FIG. 1 need to be able to communicate with each other, and the network NW may include a local network in part.

The terminal device 10 is, for example, a terminal device including an input device, a display device, a communication device, a storage device, and an arithmetic unit. More specifically, the terminal device 10 is a mobile phone such as a smartphone, a tablet terminal, various personal computers, and the like. The communication device includes a network card such as a NIC (Network Interface Card), a wireless communication module, and the like. In the terminal device 10, a UA (User Agent) such as a web browser or an application program is activated, and a request according to the content input by the user is transmitted to the service providing device 20. Further, the terminal device 10 in which the UA is activated causes the display device to display various information based on the information acquired from the service providing device 20.

The service providing device 20 is, for example, a web server that provides a web page to the terminal device 10 in response to a request from a web browser activated as a UA. The web page may be, for example, a web page constituting various websites such as a shopping site where a plurality of stores are opened on the Internet, an auction site where products are bought and sold in an auction format, and a free market site. Further, the service providing device 20 provides contents to the terminal device 10 in response to a request from an application started as a UA, and thus provides the same service as various websites such as a shopping site, an auction site, and a free market site. It may be an application server that provides. Hereinafter, as an example, the service provided by the service providing device 20 will be described as a “shopping site”, but this shopping site may be read as another service such as an auction site or a flea market site. In this case, the sale on the site may be read as an exhibition, and the purchase may be read as a successful bid.

FIG. 2 is a diagram showing an example of a web page provided by the service providing device 20. In the illustrated example, a web page on which a product can be purchased is schematically shown on a shopping site. On such a web page, for example, products in the conversion category predicted by the information processing apparatus 100 are also displayed as in the area of R1 in the figure. For example, suppose that two categories, category A and category B, are predicted as conversion categories of a user who accesses a web page. In such a case, the product classified into the category A and the product classified into the category B are displayed as recommended products on the web page. This can encourage users to make desired conversions such as browsing and purchasing. The details of the conversion category prediction method will be described later.

For example, the information processing apparatus 100 acquires various conversion information of a user who has accessed the shopping site, calculates the above-mentioned recency and frequency based on the acquired conversion information, and based on the calculated recency and frequency, the category. The score representing the user's preference for the information is calculated for each category. Then, the information processing apparatus 100 predicts the conversion category of the user based on the calculated score.

[Information processing device configuration]
FIG. 3 is a diagram showing an example of the configuration of the information processing apparatus 100 according to the first embodiment. As shown in the figure, the information processing apparatus 100 includes, for example, a communication unit 102, a control unit 110, and a storage unit 130.

The communication unit 102 includes, for example, a communication interface such as a NIC. The communication unit 102 communicates with the terminal device 10, the service providing device 20, and the like via the network NW.

The control unit 110 includes, for example, an acquisition unit 112, a first calculation unit 114, a second calculation unit 116, a prediction unit 118, a learning unit 120, and a communication control unit 122.

The components of the control unit 110 are realized by, for example, a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) executing a program stored in the storage unit 130. Further, some or all of the components of the control unit 110 are hardware such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), or GPU (Graphics Processing Unit). It may be realized by (circuit unit; circuitry), or it may be realized by the cooperation of software and hardware.

The storage unit 130 is realized by, for example, a storage device such as an HDD (Hard Disk Drive), a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), a ROM (Read Only Memory), and a RAM (Random Access Memory). In addition to various programs such as firmware and application programs, the storage unit 130 stores the first conversion information 132, the second conversion information 134, and the model information 136. These various types of information will be described later.

[Runtime processing flow]
Hereinafter, the processing contents of each component of the control unit 110 will be described with reference to the flowchart. FIG. 4 is a flowchart showing a flow of a series of run-time processes by the information processing apparatus 100 according to the first embodiment. The runtime is to use (execute) the trained logistic regression model MDL described later. The processing of this flowchart may be repeated, for example, at a predetermined cycle.

First, the acquisition unit 112 acquires the first conversion information 132 and the second conversion information 134 of the user who accessed the shopping site (hereinafter referred to as an access user) from the service providing device 20 via the communication unit 102 (hereinafter referred to as an access user). Step S100).

The first conversion information 132 is information about the first conversion reached by the access user in a certain first predetermined period in the past. The first conversion is, for example, browsing a product on a shopping site. The first predetermined period may be a period of several days, several weeks, several months, several years, and typically several months. The first conversion information 132 includes, for example, the date and time when the first conversion was reached.

The second conversion information 134 is information about the second conversion reached by the access user in a certain second predetermined period in the past. The second conversion is, for example, purchasing a product on a shopping site. The second predetermined period may be a period of several days, weeks, months, years, or the like, and is typically several years, as in the first predetermined period. The second conversion information 134 includes, for example, the date and time when the second conversion was reached.

Next, the first calculation unit 114 calculates the recency and frequency of the access user for each of the plurality of categories based on the first conversion information 132 and the second conversion information 134 acquired by the acquisition unit 112 (. Step S102).

For example, the first calculation unit 114 refers to the first conversion information 132, and calculates the time elapsed since the access user finally reached the first conversion as the recency of the first conversion. In other words, the first calculation unit 114 sets the period (number of days) from the date and time when the access user accessed the shopping site to the latest date and time when the first conversion was reached within the first predetermined period in the past. Calculated as a recency. The recency of the first conversion is an example of the "first hour".

Further, for example, the first calculation unit 114 refers to the first conversion information 132, and sets the cumulative number of times (total number of times) that the access user has reached the first conversion within the first predetermined period in the past as the first conversion. Calculated as a frequency. The frequency of the first conversion is an example of the "first cumulative number of times".

Further, for example, the first calculation unit 114 refers to the second conversion information 134, and calculates the time elapsed since the access user finally reached the second conversion as the recency of the second conversion. In other words, the first calculation unit 114 calculates the period from the date and time when the access user accesses the shopping site to the latest date and time when the second conversion is reached within the second predetermined period in the past as the recency of the second conversion. do. The second conversion recency is an example of a "second time".

Further, for example, the first calculation unit 114 refers to the second conversion information 134 and calculates the cumulative number of times the access user has reached the second conversion within the second predetermined period in the past as the frequency of the second conversion. .. The frequency of the second conversion is an example of the "second cumulative number of times".

Next, the second calculation unit 116 is based on the recency of the first conversion, the frequency of the first conversion, the recency of the second conversion, and the frequency of the second conversion calculated for each category by the first calculation unit 114. A score representing the preference of the access user for each category (hereinafter referred to as category preference score S) is calculated for each category (step S104).

For example, the second calculation unit 116 calculates the category preference score S for each category using the logistic regression model MDL indicated by the model information 136.

Model information 136 is information (program or data structure) that defines a logistic regression model MDL. For example, the model information 136 includes a plurality of partial regression coefficients a, a bias component b, and the like. The logistic regression model MDL is represented by, for example, mathematical formulas (1) and (2).

x ₁ to x ₄ represent explanatory variables of the logistic regression model MDL. a ₁ represents the partial regression coefficient of the explanatory variable x ₁ , a ₂ represents the partial regression coefficient of the explanatory variable x ₂ , a ₃ represents the partial regression coefficient of the explanatory variable x ₃ , and a ₄ represents the partial regression coefficient of the explanatory variable x ₄ . It represents the partial regression coefficient. Further, b represents a bias component as described above. The plurality of partial regression coefficients a and the bias component b are predetermined by the learning unit 120 described later.

R ₁ represents the recency of the first conversion calculated for a certain target category c _i , F ₁ represents the frequency of the first conversion calculated for the target category c _i , and R ₂ represents the frequency of the first conversion calculated for the target category c _i . The calculated second conversion recency is represented, and F ₂ represents the calculated second conversion frequency for the target category _ci . As shown in equation (2), the explanatory variable x ₁ is the logarithmic value of the recency R ₁ of the first conversion, the explanatory variable x ₂ is the logarithmic value of the frequency F ₁ of the first conversion, and the explanatory variable x ₃ is. , The logarithmic value of the recency _R2 of the _second conversion, and the explanatory variable x4 is the logarithmic value of the frequency F2 of the _second conversion. Depending on the user, the recency R or frequency F may be much larger or smaller than others. Therefore, by taking the logarithm of each recency R or frequency F, the score S can be easily calculated even when the values of the recency R and frequency F are significantly different from those of others.

For example, the second calculation unit 116 inputs the logarithmic value of the recency R 1 of the first conversion as the explanatory variable x 1 to the mathematical formula (1) representing the logistic regression model MDL, and inputs the logarithmic value of the recency R ₁ of the _first conversion as the explanatory variable x ₁ . The logarithmic value is entered as the explanatory variable x ₂ , the logarithmic value of the recency _R2 of the second conversion is entered as the explanatory variable _x3 , and the logarithmic value of the frequency F ₂ of the _second conversion is entered as the explanatory variable x4. As a result, the logistic regression model MDL outputs the category preference score S of the target category _ci as the objective variable. The category preference score S output by the logistic regression model MDL may be treated as a probability.

For example, when there are m types of categories to which the products handled on the shopping site are classified, the second calculation unit 116 calculates the category preference score S while changing the target category _ci among the m categories. This is repeated, and the category preference score S is calculated for all m categories.

Next, the prediction unit 118 assumes or expects that the access user will reach the first conversion or the second conversion at a certain point in the future based on the category preference score S calculated for each category by the second calculation unit 116. The category of the product to be produced, that is, the conversion category is predicted (step S106).

For example, the prediction unit 118 may compare the category preference scores S of each of the m categories, and predict the category with the highest score S as the conversion category. Further, the prediction unit 118 may predict a high-ranking predetermined number of categories having a high score S as conversion categories.

Next, the communication control unit 122 transmits the information of the conversion category predicted by the prediction unit 118 to the service providing device 20 via the communication unit 102 (step S108). In response to this, the service providing device 20 may display a web page or the like on which products classified in the conversion category are posted on the terminal device 10 used by the access user. As a result, the interest of the user can be increased, and the conversion (for example, purchase) desired by the seller who sells the product on the shopping site or the like is likely to occur.

[Training process flow]
Hereinafter, the flow of a series of processing of the training of the information processing apparatus 100 in the first embodiment will be described according to a flowchart. Training is training the logistic regression model MDL used at runtime. FIG. 5 is a flowchart showing a flow of a series of training processes by the information processing apparatus 100 according to the first embodiment. The processing of this flowchart may be repeated, for example, at a predetermined cycle.

First, the acquisition unit 112 acquires the first conversion information 132 and the second conversion information 134 of each of the unspecified number of used users from the service providing device 20 via the communication unit 102 (step S200). A used user is a user who has accessed a shopping site in the past and has reached a second conversion at the shopping site. For example, when the second conversion is to purchase a product, the used user may be read as a user who has purchased the product on the shopping site in the past. The used user is an example of a "learning target user".

Next, the first calculation unit 114 uses an unspecified number of used users for each of the plurality of categories based on the first conversion information 132 and the second conversion information 134 acquired by the acquisition unit 112. The recency and frequency of the above are calculated (step S202).

For example, the first calculation unit 114 sets a certain past reference date in the first conversion information 132, and the user who has used the user has reached the first conversion within the first predetermined period further past the reference date. The period until the date and time is calculated as the recency of the first conversion.

Further, for example, in the first conversion information 132, the first calculation unit 114 determines the cumulative number of times that the used user has reached the first conversion within the first predetermined period further before the reference date, as the frequency of the first conversion. Calculated as.

Further, for example, in the second conversion information 134, the first calculation unit 114 sets the period up to the latest date and time when the used user reaches the second conversion within the second predetermined period further than the reference date. Calculated as a recency of 2 conversions.

Further, for example, in the second conversion information 134, the first calculation unit 114 determines the cumulative number of times that the used user has reached the second conversion within the second predetermined period further before the reference date, as the frequency of the second conversion. Calculated as.

Next, the learning unit 120 learns the logistic regression model MDL based on the recency and frequency calculated by the first calculation unit 114 (step S204).

For example, the learning unit 120 learns the logistic regression model MDL with the category of the product for which the used user has reached the second conversion in the third predetermined period after the reference date as the correct conversion category. The third predetermined period may be a period of several days, several weeks, several months, several years, and typically several weeks, as in the first predetermined period and the second predetermined period. For example, if you want to predict future conversion categories beyond weeks, the third predetermined period may be set to a period of months or years.

More specifically, in the learning unit 120, when the logarithmic values of the regression and frequency before the reference date are input as the explanatory variables x, the used user is second in the third predetermined period after the reference date. Category preference score of the category of the product that led to the conversion Higher (closer to 1), and the category preference score of the category of the product that the used user did not reach the second conversion in the third predetermined period. The partial regression coefficient a and the bias component b of the logistic regression model MDL are determined by the least squares method or the like so that S becomes low (close to 0).

Then, the learning unit 120 stores the model information 136 of the logistic regression model MDL in which the partial regression coefficient a and the bias component b are determined by learning in the storage unit 130 (step S206). This ends the processing of this flowchart. Through such training, the category of products that are likely to reach the second conversion at runtime is predicted as the conversion category.

In the above-mentioned training flowchart, the learning unit 120 sets the category of the product for which the used user has reached the "second conversion" in the third predetermined period after the reference date as the correct conversion category, and the logistic regression model MDL. I explained it as learning, but it is not limited to this. For example, the learning unit 120 may learn the logistic regression model MDL by using the category of the product for which the used user has reached the "first conversion" in the third predetermined period after the reference date as the correct conversion category. As a result, the category of products that easily leads to the first conversion at runtime is predicted as the conversion category.

According to the first embodiment described above, the information processing apparatus 100 has a recency of the first conversion, a frequency of the first conversion, and a frequency of the first conversion for each of a plurality of categories to which the products handled in the shopping site or the like are classified. Calculate the recency of the second conversion and the frequency of the second conversion. The information processing apparatus 100 calculates the category preference score S for each category based on the calculated recency and frequency. Then, the information processing apparatus 100 sets the category of the product that the user is expected to reach the first conversion or the second conversion at a certain point in the future as the conversion category based on the calculated category preference score S of each category. Predict. By such processing, it is possible to accurately predict the category of the product or service that the user is likely to convert.

<Second Embodiment>
Hereinafter, the second embodiment will be described. The second embodiment differs from the first embodiment described above in that the categories are layered. Hereinafter, the differences from the first embodiment will be mainly described, and the points common to the first embodiment will be omitted. In the description of the second embodiment, the same parts as those of the first embodiment will be described with the same reference numerals.

FIG. 6 is a diagram showing an example of hierarchical categories. As shown in the figure, the category "food" included in the first layer has a category of the second layer, and the category of the second layer includes "rice," vegetables "," noodles ", and" bread ". It includes multiple categories such as "" and "drinks". In the category of "rice" and the category of "drink" in the second layer, there is a category of the third layer. The third layer category, which is a lower layer of the second layer category called "rice", includes multiple categories such as "polished rice", "washless rice", "brown rice", and "grain rice". .. The third-level category, which is a lower level of the second-level category called "drink", includes a plurality of categories such as "wine", "shochu", "sake", and "beer".

When a plurality of categories are layered in this way, the second calculation unit 116 does not depend on the hierarchical structure, and as in the case of the first embodiment, the category preference score of each category of each layer. Calculate S. For example, it is assumed that the second calculation unit 116 selects the third layer as the target layer from the plurality of layers. In this case, the second calculation unit 116 sets the category preference score S of each of the plurality of categories included in the selected third layer to the categories included in the first layer and the second layer higher than the third layer. , It is calculated independently of the categories included in the fourth and subsequent layers below the third layer.

Then, the prediction unit 118 predicts the conversion category in the target hierarchy based on the category preference score S of the category included in the hierarchy higher than the target hierarchy and the category preference score S of the category included in the target hierarchy.

For example, when the third layer is set as the target layer, the prediction unit 118 not only has the category preference score S of the category such as "polished rice" or "washless rice", but also the category preference of the category "rice" of the upper layer. The conversion category may be predicted in the third layer in consideration of the score S. Specifically, the prediction unit 118 weighted averages, arithmetically averages, and weights the category preference score S of the second layer category with respect to the category preference score S of the third layer category. You may do it.

As a result, for example, if the category preference score S of the category "drink" is higher than that of the category "rice" in the second layer, "polished rice" or "washless rice" in the third layer Even if the category preference score S of the category related to "drinks" such as "wine" and "drink" is the same or lower than the category preference score S of the category related to "rice", "wine" or "wine" or Categories related to "drinks" such as "shochu" are more likely to be selected as conversion categories.

According to the second embodiment described above, the information processing apparatus 100 is based on the category preference score S of the category included in the hierarchy higher than the target hierarchy and the category preference score S of the category included in the target hierarchy. , Predict conversion categories in the target hierarchy. This makes it possible to accurately predict the categories of products or services that users are more likely to convert.

<Hardware configuration>
The information processing apparatus 100 of the above-described embodiment is realized by, for example, a hardware configuration as shown in FIG. 7. FIG. 7 is a diagram showing an example of the hardware configuration of the information processing apparatus 100 of the embodiment.

The information processing device 100 includes NIC100-1, CPU100-2, RAM100-3, ROM100-4, a secondary storage device 100-5 such as a flash memory or an HDD, and a drive device 100-6 on an internal bus or a dedicated communication line. It is configured to be interconnected by. A portable storage medium such as an optical disk is mounted on the drive device 100-6. A program stored in a portable storage medium mounted on the secondary storage device 100-5 or the drive device 100-6 is expanded in the RAM 100-3 by a DMA controller (not shown) or the like, and executed by the CPU 100-2. As a result, the control unit 110 is realized. The program referred to by the control unit 110 may be downloaded from another device via the network NW.

Although the embodiments for carrying out the present invention have been described above using the embodiments, the present invention is not limited to these embodiments, and various modifications and substitutions are made without departing from the gist of the present invention. Can be added.

1 ... Information processing system, 10 ... Terminal device, 20 ... Service providing device, 100 ... Information processing device, 102 ... Communication unit, 110 ... Control unit, 112 ... Acquisition unit, 114 ... First calculation unit, 116 ... Second calculation Unit, 118 ... Prediction unit, 120 ... Learning unit, 122 ... Communication control unit, 130 ... Storage unit

Claims (8)

For each of the plurality of categories to which the goods or services handled in the electronic commerce are classified, the first time elapsed since the user last reached the first conversion for the goods or services classified in the above categories, and the above. The first cumulative number of times the user has reached the first conversion for a product or service classified in the category in the past period, and the user last for the product or service classified in the category. The first to calculate the second time elapsed after reaching the second conversion and the second cumulative number of times the user has reached the second conversion in the past period for the goods or services classified in the category. Calculation unit and
Based on the first time, the first cumulative number of times, the second time, and the second cumulative number of times calculated for each category by the first calculation unit, a score representing the user's preference for the category is given. , The second calculation unit calculated for each category,
A category of goods or services in which the user is expected to reach the first conversion or the second conversion at some point in the future based on the score calculated for each category by the second calculation unit. A predictor that predicts conversion categories,
Information processing device equipped with. One conversion of the first conversion and the second conversion is to browse the goods or services classified in the category, and the other conversion is to purchase the goods or services classified in the category. Is,
The information processing apparatus according to claim 1. The second calculation unit inputs the first time, the first cumulative number of times, the second time, and the second cumulative number of times into the logistic regression model as explanatory variables to obtain the logistic regression model. Obtain the score as the objective variable,
The information processing apparatus according to claim 1 or 2. In the logistic regression model, when the first time, the first cumulative number, the second time, and the second cumulative number of the learning target user are input, the learning target user performs the first conversion. Or it is learned so that the score of the category of the product or service that has reached the second conversion is high.
The information processing apparatus according to claim 3. When the plurality of categories are hierarchized, the second calculation unit calculates the score for each of the one or more categories included in a certain target hierarchy.
The prediction unit predicts the conversion category in the target hierarchy based on the score of the category included in the hierarchy higher than the target hierarchy and the score of the category included in the target hierarchy.
The information processing apparatus according to any one of claims 1 to 4. The second calculation unit calculates the score of the category of the target hierarchy without depending on the category included in the hierarchy higher than the target hierarchy and the category included in the hierarchy lower than the target hierarchy. do,
The information processing apparatus according to claim 5. The computer
For each of the plurality of categories to which the goods or services handled in the electronic commerce are classified, the first time elapsed since the user last reached the first conversion for the goods or services classified in the above categories, and the above. The first cumulative number of times the user has reached the first conversion for a product or service classified in the category in the past period, and the user last for the product or service classified in the category. The second time elapsed after reaching the second conversion and the second cumulative number of times the user has reached the second conversion in the past for a certain period in the past for the goods or services classified in the category are calculated.
Based on the first hour, the first cumulative number of times, the second hour, and the second cumulative number of times calculated for each category, a score representing the user's preference for the category is calculated for each category. ,
Based on the score calculated for each of the categories, the conversion category, which is the category of goods or services that the user is expected to reach the first conversion or the second conversion at some point in the future, is predicted.
Information processing method. On the computer
For each of the plurality of categories to which the goods or services handled in the electronic commerce are classified, the first time elapsed since the user last reached the first conversion for the goods or services classified in the above categories, and the above. The first cumulative number of times the user has reached the first conversion for a product or service classified in the category in the past period, and the user last for the product or service classified in the category. 2 To calculate the second time elapsed since the conversion was reached and the second cumulative number of times the user has reached the second conversion in the past for a certain period of time for the goods or services classified in the category.
Based on the first hour, the first cumulative number of times, the second hour, and the second cumulative number of times calculated for each category, a score representing the user's preference for the category is calculated for each category. matter,
Predicting a conversion category, which is a category of goods or services that the user is expected to reach the first conversion or the second conversion at some point in the future, based on the score calculated for each category.
A program to execute.

Images