JP2022180112A - Information processing system - Google Patents

Abstract

To recommend a commodity in accordance with the taste of a user.SOLUTION: An information processing system includes: an e-mail acquisition unit which acquires e-mail data of a user; a purchase data extraction unit which analyzes the e-mail data and extracts purchase data relating to a commodity or a service purchased by the user; a purchase history storage unit which stores the purchase data; a purchase intension input unit which receives an input of intention to purchase a commodity or a service again from the user; and a recommendation unit which recommends a commodity or a service to the user in accordance with the intention.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information processing system.

A product is recommended according to the purchase history (see, for example, Patent Literature 1).

JP-A-2002-024278

However, it is difficult to grasp how users feel about products.

SUMMARY OF THE INVENTION The present invention has been made in view of such a background, and an object of the present invention is to provide a technology capable of proposing products according to user's taste.

The main invention of the present invention for solving the above problems is an information processing system, comprising: an e-mail acquisition unit for acquiring e-mail data of a user; further comprising a purchase data extraction unit for extracting purchase data related to a service, a purchase history storage unit for storing the purchase data, and a purchase intention input unit for receiving an input of intention to purchase the product or service again from the user, and a recommendation unit that proposes the product or service to the user according to the intention.

Other problems disclosed by the present application and solutions thereof will be clarified by the section of the embodiment of the invention and the drawings.

According to the present invention, it is possible to propose products according to user's taste.

It is a figure which shows the whole configuration example of the recommendation system which concerns on this embodiment. It is a figure which shows the hardware structural example of the server apparatus 10 which concerns on this embodiment. 2 is a diagram showing a software configuration example of the server device 10 according to the present embodiment; FIG. It is a figure explaining an example of e-mail data. It is a figure explaining the decision of the goods to propose according to both favorable impression and purchase intention. It is a figure explaining operation|movement of the recommendation system of this embodiment.

<Overview of the invention>
The contents of the embodiments of the present invention are listed and explained. The present invention has, for example, the following configurations.
[Item 1]
an e-mail acquisition unit for acquiring e-mail data of a user;
a purchase data extraction unit that analyzes the e-mail data and extracts purchase data related to the product or service purchased by the user;
a purchase history storage unit that stores the purchase data;
further comprising a purchase intention input unit that receives an input of intention to purchase the product or service again from the user;
a recommendation unit that proposes the product or service to the user according to the intention;
An information processing system comprising:
[Item 2]
The information processing system according to item 1,
The recommendation unit recommends the first product or service to the user who has entered the intention to repurchase the first product or service, and repurchase the first product or service. recommending a second product or service different from the first product or service to the user who has input the intention of not wanting to use the product;
An information processing system characterized by
[Item 3]
The information processing system according to item 2,
the purchase data includes information indicating the provider of the product or service;
The recommendation unit is provided by a provider different from the provider included in the purchase data to the user who has entered the intention not to purchase the first product or service again. recommending a first product or service;
An information processing system characterized by
[Item 4]
The information processing system according to any one of items 1 to 3,
The recommendation unit estimates a purchase interval at which the user purchases the product or service based on the purchase data, and a period corresponding to the purchase interval has elapsed since the date of the latest purchase data corresponding to the user. Proposing the product or service to the user at the timing of
An information processing system characterized by
[Item 5]
The information processing system according to any one of items 1 to 4,
further comprising a point issuing unit that issues points to the user in response to the input of the intention;
An information processing system characterized by
[Item 6]
The information processing system according to any one of items 1 to 5,
a favorability input unit that receives an input of favorability regarding the product or service from the user;
The recommendation unit proposes the product or service to the user according to the favorability and the intention;
An information processing system characterized by

<System Overview>
A recommendation system according to an embodiment of the present invention will be described below. The recommendation system of this embodiment attempts to recommend products (including services; the same shall apply hereinafter) to users. The recommendation system of this embodiment receives user evaluations of products that the user has purchased in the past, and attempts to recommend the same or different products according to the user's evaluations of the products. In this embodiment, the evaluation received from the user includes the intention to purchase the product again, and can also include the favorable impression of the product.

In this embodiment, the repurchase intention may be in two categories of whether or not the purchase is desired. For example, it can be a category of "want to buy again" or "never buy again", or a category of "want to continue" or "want to withdraw" in the case of subscription type purchase (continuous purchase). The intention to purchase again may be a category other than that, or may be a score indicating the degree of desire to purchase, such as 5 levels or 10 levels.

Also, in this embodiment, the favorable impression may be two categories of whether or not the user has a favorable impression of the product. For example, it can be a "like" or "dislike" category. Note that the favorable impression may be a category other than that, or may be a score indicating the degree of favorable impression, such as 5 levels or 10 levels.

The recommendation system of the present embodiment recommends products according to the repurchase intention and favorability received from the user.

FIG. 1 is a diagram showing an example of the overall configuration of a recommendation system according to this embodiment. The recommendation system of this embodiment includes a server device 10 . The server device 10 is communicably connected to the user terminal 20 via a communication network. The communication network is, for example, the Internet, and can be constructed by a public telephone line network, a mobile telephone line network, a wireless communication channel, Ethernet (registered trademark), or the like.

<Hardware configuration example>
The server device 10 may be a general-purpose computer such as a workstation or personal computer, or may be logically implemented by cloud computing.

FIG. 2 is a diagram showing a hardware configuration example of the server device 10 according to this embodiment. The server device 10 may be a general-purpose computer such as a workstation or personal computer, or may be logically implemented by cloud computing. Note that the illustrated configuration is an example, and other configurations may be employed. The information processing apparatus includes a CPU 101 , a memory 102 , a storage device 103 , a communication interface 104 , an input device 105 and an output device 106 . The storage device 103 is, for example, a hard disk drive, solid state drive, flash memory, etc., which stores various data and programs. The communication interface 104 is an interface for connecting to a communication network. Examples include a USB (Universal Serial Bus) connector and an RS232C connector for serial communication. The input device 105 is, for example, a keyboard, a mouse, a touch panel, a button, a microphone, etc. for inputting data. The output device 106 is, for example, a display, printer, speaker, or the like that outputs data. Note that each functional unit provided in the server device 10, which will be described later, is implemented by the CPU 101 reading out a program stored in the storage device 103 into the memory 102 and executing it. It can be implemented as part of the storage area provided by the storage device 103 . Each functional unit of the server device 10, which will be described later, is implemented by the CPU 101 reading a program stored in the storage device 103 into the memory 102 and executing it. is implemented as part of the storage provided by

<Software configuration example>
FIG. 3 is a diagram showing a software configuration example of the server device 10 according to this embodiment. The server device 10 includes an e-mail acquisition unit 111, a purchase data extraction unit 112, a purchase intention input unit 113, a favorability input unit 114, a point issuing unit 115, a recommendation unit 116, a purchase history storage unit 131, a rule storage unit 132, a user An evaluation storage unit 133 and a product information storage unit 134 are provided.

The purchase history storage unit 131 stores information about purchased products (hereinafter referred to as purchase data). The purchase data may include a user ID indicating the user, the date of purchase, the product ID identifying the purchased product, the name of the store where the transaction was made, the amount paid for the product, the shipping cost of the product, the purchase price, and the like. The purchase amount is the final amount after adding the shipping fee to the payment amount and performing additions and subtractions according to points, fees, etc., and can be used as the settlement amount to be settled. In addition, you may employ|adopt a payment amount as a purchase amount.

The rule storage unit 132 stores rules for analyzing email data. The rule storage unit 132 can store rules for each store, for example. A rule may be, for example, a learning model learned by machine learning of e-mail data. The rule storage unit 132 further associates and stores the pattern of the address that is the transmission source of the e-mail (it may be the address itself, or it may be a pattern expressing the address using a regular expression or the like). can also be

The user evaluation storage unit 133 stores user evaluations of purchased products. The user evaluation storage unit 133 can store the purchase intention and favorable rating in association with the user ID and the product ID. The purchase intention and favorability are values input by the user for the purchased product. In this embodiment, the purchase intention includes whether or not to purchase again ("I want to buy again", "I want to continue my subscription", etc., and "I will not buy again" or " I want to cancel the subscription" and do not wish to purchase again). Also, in this embodiment, the favorability is either "like" or "dislike". In addition, as described above, both the purchase intention and favorable impression can be expressed by numerical values.

The product information storage unit 134 stores information on products (product information). The product information stored in the product information storage unit 134 includes the product name, the purpose of the user related to the product, the category of the product, and the attributes of the product in association with the product ID that identifies the product. can be done. The product name may have a pattern that allows identification of variations in the notation of the product. Multiple objectives can be set. For example, it is possible to set the purpose of "writing" and "decoration" for a ballpoint pen. Categories can be set for each purpose. A category may have a plurality of hierarchies, and for example, large classification, medium classification, small classification, etc. can be set. The attribute may be, for example, information (demographic information, etc.) specifying a user segment assumed by this product, or information (segment information) for specifying a user or segment who likes this product. The segment information can be a feature quantity common to users corresponding to purchase data including a product ID.

The e-mail acquisition unit 111 acquires e-mail data addressed to the user. The e-mail acquisition unit 111 can access a mail server and acquire e-mail data from a user's mailbox according to protocols such as POP (Post Office Protocol) and IMAP (Internet Message Access Protocol). The server apparatus 10 is provided with a mail server storage section for managing mail servers to be accessed in association with user IDs, and the e-mail acquisition section 111 accesses all mail servers registered in the mail server storage section. can also be configured to collect email data.

The purchase data extraction unit 112 analyzes the e-mail data and extracts purchase data related to the product purchased by the user. Among the e-mail data, those related to the purchased product contain information (purchase data) related to the product. The e-mail data related to the purchased product is, for example, an e-mail sent from the EC site to the user for confirmation after the user purchases the product on the EC site. A user can confirm what and when he purchased products purchased from various product or service providers by looking only at e-mail data without being conscious of the servers of each provider. The purchase data extraction unit 112 can extract purchase data from e-mail data according to rules registered in the rule storage unit. The purchase data extracting unit 112 reads from the rule storage unit 132 a rule corresponding to an address pattern matching the source address included in the e-mail data, and extracts the purchase data from the e-mail data based on the read rule. can be extracted. Further, the purchase data extraction unit 112 may apply all the rules stored in the rule storage unit 132 and use the rule that successfully extracts the purchase data. A store name corresponding to the rule can be specified. By extracting purchase data from e-mail data, the purchase data extraction unit 112 can acquire a wide variety of purchase data from e-mails without accessing servers of providers such as various EC sites. .

FIG. 4 is a diagram illustrating an example of e-mail data. FIG. 4 shows an example of purchase notification mail 21, which is an e-mail notified when a product is purchased on an EC (electronic commerce) site. The purchase notification mail 21 contains text data indicating the date and time 211 when the product was purchased, the product name 212 of the purchased product, and the transaction amount 213 (including the payment amount and shipping fee for the product). The purchase data extractor 112 can extract these data from the e-mail data according to the rules and create purchase data based on the extracted data. The purchase data extraction unit 112 can identify the product ID corresponding to the product name 212 from the product information storage unit 134 .

Incidentally, the purchase data extracting unit 112 may extract the favorability rating of the purchase data containing the data corresponding to the favorability rating. For example, if the e-mail data contains an evaluation value (stars, etc.) set by the user for the product on an EC site or the like, the purchase data extraction unit 112 can acquire the evaluation value as the favorability rating. .

In addition, when the e-mail data includes a review written by the user about the product, the purchase data extracting unit 112 can analyze the review and estimate the favorability or repurchase intention. The purchase data extracting unit 112 counts, for example, the number of reviews containing keywords indicating favorable impression or intention to purchase again and the number of keywords indicating disgust or unwillingness to purchase again, and Favorability or repurchase intention can be estimated according to these. Moreover, the purchase data extracting unit 112 may estimate the favorability of the product and the intention of repurchasing the product by analyzing the syntax in addition to the presence or absence of the keyword. Machine learning can also be used to estimate review favorability and repurchase intent.

The purchase data extraction unit 112 can register the extracted purchase data in the purchase history storage unit 131 .

The purchase intention input unit 113 receives an intention to purchase again the product related to the purchase data. For example, the purchase intention input unit 113 extracts product IDs included in purchase data corresponding to a specific user from the purchase history storage unit 131 without duplication, and transmits product information corresponding to the product ID to be extracted to the user terminal 20. Then, a list of product information (which may be limited to those for which the user ID and purchase intention corresponding to the product ID are not registered in the user evaluation storage unit 133) is transmitted to the user terminal 20, and the user terminal 20 , the product can be specified and the purchase intention can be input. The purchase intention input unit 113 receives the designated product and the input purchase intention from the user terminal 20, and registers (or updates) the purchase intention in the favorability storage unit 133 in association with the user ID and product ID. can be done.

The favorability input unit 114 receives input of favorability for the product related to the purchase data. For example, the favorability input unit 114 extracts product IDs included in purchase data corresponding to a specific user from the purchase history storage unit 131 without duplication, and transmits product information corresponding to the product ID to be extracted to the user terminal 20. Then, a list of product information (which may be limited to those for which the user ID and favorable rating corresponding to the product ID are not registered in the user evaluation storage unit 133) is transmitted to the user terminal 20, and the user terminal 20 , the product can be specified and the favorability can be input. The favorable rating input unit 114 receives the designated product and the inputted favorable rating from the user terminal 20, and registers (or updates) the favorable rating in the favorable rating storage unit 133 in association with the user ID and the product ID. can be done. In addition, it is possible to simultaneously receive the purchase intention and favorable impression.

The point issuing unit 115 can issue points in accordance with at least one of purchase intent and favorability. Points can be given to the user as a reward for inputting at least one of purchase intent and favorability. The point issuing unit 115 can issue points, for example, according to the number of products for which purchase intent and favorability are input.

The recommendation unit 116 proposes products according to favorable ratings. The recommendation unit 116 recommends the first product to a user who has a favorable impression of the first product, and recommends a product different from the first product to a user who does not have a favorable impression of the first product. A second product can be recommended. When recommending a second product, the recommendation unit 116 can select, as the second product, a product similar to or a substitute for the first product. If the products are categorized, similar or alternative products can be selected depending on the category. For example, when major classifications, middle classifications, and small classifications are set for commodities, it is possible to select commodities that share the same small classification as similar goods, and commodities that share the middle classification as substitute goods.

In addition, the recommendation unit 116 can suggest products to the user according to both the favorability rating and the purchase intention. FIG. 5 is a diagram for explaining the determination of products to be proposed according to both favorable ratings and purchase intentions.

As shown in the figure, the recommendation unit 116 proposes the same product as the product purchased by the user when the favorable impression is good (“like”) and the user intends to purchase again (“purchase again”). can be done. Also, even when the favorable impression is good (“like”), when there is no intention to purchase again (“never buy again”), the recommendation unit 116 proposes a product similar to or a substitute for the product purchased by the user. can do.

The recommendation unit 116 can also suggest the same product as the product purchased by the user when there is a repurchase intention (“purchase again”) even if the favorable impression is poor (“dislike”). You can also suggest similar or alternative products. In addition, when the favorable impression is poor (“dislike”) and there is no repurchase intention (“never buy again”), the recommendation unit 116 can suggest a product similar to or a substitute for the product purchased by the user. can.

In addition, the recommendation unit 116 can propose a similar product or a substitute product for each purpose when recommending products. For example, the recommendation unit 116 reads the category for each purpose included in the product information corresponding to the product ID of the purchase data, and recommends other You can suggest products.

Also, the recommendation unit 116 can estimate the purchase interval at which the user purchases products based on the purchase data. The recommendation unit 116 can propose a product to the user when a period corresponding to the estimated purchase interval has elapsed from the date of the latest purchase data corresponding to the user. For example, the recommendation unit 116 selects, from the purchase history storage unit 131, among the purchase data corresponding to the user ID and the product ID, the past predetermined period (the entire period may be used, or the past one year may be used. , may be a predetermined number in order of newest), arranging the extracted purchase data in order of purchase date, and summarizing the interval between the purchase dates of two consecutive purchase data (which may be an average value) However, it may be the median value or the expected value of the distribution estimated by Bayesian learning or the like.) can be estimated as the purchase interval. The recommendation unit 116 can recommend products on a date obtained by adding the estimated purchase interval to the date of the latest purchase data.

In addition, based on the purchase data, the recommendation unit 116 determines the number of times the product was purchased (purchase frequency) during a predetermined period (for example, one month, six months, or one year). If it is equal to or greater than a predetermined threshold, it is possible to propose the product even when the favorable impression of the product is low (“dislike”). In addition, when the favorable impression is poor, the recommendation unit 116 can propose products similar to the product in question (products in the same minor classification). In addition, when the favorable impression is poor, the recommendation unit 116 can also propose alternative products (products in the same middle classification) as the product in question. In addition, if the favorable impression is not good and there is an intention to purchase again ("want to buy again"), other products (similar products or alternative products) having the same purpose as the product concerned can be proposed.

In addition, the recommendation unit 116 detects information specifying a product (product name, etc.) and keywords related to the product (eg, “broken”, “unusable”, “lost”) from the e-mail data. As an opportunity, the product concerned, a similar product, or a substitute product can be proposed.

In addition, the recommendation unit 116 is triggered by the start of a campaign such as a discount by a product seller, and the user who has set a high favorable rating for the target product or the intention to purchase again for the target consumption (“want to buy again”). ) can be proposed to the target product.

<Action>
FIG. 6 is a diagram for explaining the operation of the recommendation system of this embodiment.

The server device 10 acquires the user's e-mail data (S301), analyzes the acquired e-mail data to extract purchase data (S302), and registers the extracted purchase data in the purchase history storage unit 131 (S303). ). The server device 10 accepts input of the user's favorability rating and repurchase intention for the product (S305), registers the user ID and product ID in the user evaluation storage unit 133, and registers the favorable impression and repurchase intention in the user evaluation storage unit 133. points can be issued according to the input of (S306).

Next, the server device 10 determines the suggested product according to the favorability rating and purchase intention (S306). For example, as shown in FIG. 5, the server device 10 can determine one or a plurality of suggested products from the same product, alternative product, or similar product according to the favorability rating and purchase intention.

Server device 10 estimates the purchase interval for each product based on the purchase data stored in purchase history storage unit 131 (S307), and determines whether the estimated purchase interval has passed since the date of the previous purchase. Judgment is made (S308), and if the purchase interval has passed (S308: YES), the determined suggested product is proposed (S309).

As described above, according to the recommendation system of the present embodiment, proposed products corresponding to the evaluation (favorability and repurchase intention) input by the user for the product are presented at the timing of the purchase interval according to the purchase history. can be proposed.

Although the present embodiment has been described above, the above-described embodiment is intended to facilitate understanding of the present invention, and is not intended to limit and interpret the present invention. The present invention can be modified and improved without departing from its spirit, and the present invention also includes equivalents thereof.

For example, in the present embodiment, the server device 10 is described as being one computer, but the functional units and storage units of the server device 10 can be distributed among a plurality of computers.

In addition, in the present embodiment, the product is recommended according to both the intention to purchase again and the favorable impression, but the present invention is not limited to this. can be recommended. For example, when the user has an intention to purchase a certain product again ("want to buy again"), the server device 10 can recommend the same product, alternative products, and similar products. The server device 10 can also recommend purchasing the same product from a different store when the user has no intention of purchasing the product again (“I will not buy it again”). In addition, when there is no intention to purchase again, the server device 10 can propose other products (alternative products or similar products) having the same purpose as the relevant product.

In addition, in the present embodiment, the purchase history of products or services is acquired from e-mail data, but this is not the only option. It is also possible to specify the purchased product.

Alternatively, data may be extracted (scraped) from a web page of an EC site or the like, or purchase data may be obtained from a server of another company such as an EC site or a credit card company. Complementary use of these purchase data to acquire purchase data that could not be extracted from e-mail data, or to acquire some items of purchase data that could not be extracted from e-mail data can be done.

Further, the server device 10 can detect related products from the contents of the calendar data and TODO data, and can receive favorable ratings and repurchase intentions for the detected products.

In addition, the recommendation unit 116 of the server device 10 detects the user's free time from the calendar data, and offers entertainment-related products (including movie viewing and visits to commercial facilities) at a predetermined time before the free time. can be proposed.

10 server device 20 user terminal

Claims (6)

an e-mail acquisition unit for acquiring e-mail data of a user;
a purchase data extraction unit that analyzes the e-mail data and extracts purchase data related to the product or service purchased by the user;
a purchase history storage unit that stores the purchase data;
further comprising a purchase intention input unit that receives an input of intention to purchase the product or service again from the user;
a recommendation unit that proposes the product or service to the user according to the intention;
An information processing system comprising: The information processing system according to claim 1,
The recommendation unit recommends the first product or service to the user who has entered the intention to repurchase the first product or service, and repurchase the first product or service. recommending a second product or service different from the first product or service to the user who has input the intention of not wanting to use the product;
An information processing system characterized by The information processing system according to claim 2,
the purchase data includes information indicating the provider of the product or service;
The recommendation unit is provided by a provider different from the provider included in the purchase data to the user who has entered the intention not to purchase the first product or service again. recommending a first product or service;
An information processing system characterized by The information processing system according to any one of claims 1 to 3,
The recommendation unit estimates a purchase interval at which the user purchases the product or service based on the purchase data, and a period corresponding to the purchase interval has elapsed since the date of the latest purchase data corresponding to the user. Proposing the product or service to the user at the timing of
An information processing system characterized by The information processing system according to any one of claims 1 to 4,
further comprising a point issuing unit that issues points to the user in response to the input of the intention;
An information processing system characterized by The information processing system according to any one of claims 1 to 5,
a favorability input unit that receives an input of favorability regarding the product or service from the user;
The recommendation unit proposes the product or service to the user according to the favorability and the intention;
An information processing system characterized by

Images