JP6401428B1 - Information processing apparatus, information processing method, program, and storage medium - Google Patents

Abstract

The purpose is to appropriately suggest new keywords that the user is interested in. For this purpose, the information processing apparatus acquires information for acquiring text information input by the user for search and information related to a genre based on a web page browsed by the user during a predetermined period before the search. A genre information estimation unit that estimates a genre that the user is interested in from the information related to the genre as a short-term attention genre, an extraction unit that extracts a short-term suggestion word based on the text information and the short-term attention genre, and an extraction And a presentation unit for presenting the short term suggested word to the user.

Description

  The present invention relates to an information processing apparatus, an information processing method, a program, and a storage medium, and more specifically, to a suggestion word technique presented to a user at the time of search.

JP 2013-250677 A

  When information desired by a user is obtained by searching, it is common to specify a keyword for searching (hereinafter referred to as a search query). At that time, instead of letting the user input all the characters of the search query, the user's burden is reduced by complementing the remaining character strings according to a part of the input search query. Yes. More specifically, a completed character string candidate estimated from a part of character strings input by the user is presented to the user as a suggestion word. The user can omit part of the input operation of the search query by selecting the suggested suggestion word.

  When suggesting a suggestion word, a keyword that is most frequently used during a search may be presented, or a keyword corresponding to user attribute information may be presented. In addition, for example, Patent Document 1 discloses a technique for presenting a suggestion word in consideration of a periodic trend.

Since the periodic trend described in Patent Document 1 is estimated from the tendency of many users, an appropriate keyword is not always presented as a suggestion word for each user.
In addition, even if the keywords that are periodically searched for are estimated for each user, those keywords can be estimated due to previous searches by the user, and the user is newly interested. It is difficult to suggest keywords.
Accordingly, the present invention has been made in consideration of such a situation, and an object of the present invention is to appropriately suggest a keyword that a user is newly interested in.

The information processing apparatus according to the present invention acquires text information input by a user for a search and information related to a genre based on a web page browsed by the user during a predetermined period before the search. An acquisition unit; a genre information estimation unit that estimates a genre that the user is interested in from information related to the genre as a short-term attention genre; and an extraction unit that extracts a short-term suggestion word based on the text information and the short-term attention genre. And a presentation unit for presenting the extracted short-term suggest word to the user.
The web page browsed by the user during a predetermined period before the search is highly likely to be the web page that the user was most interested in, and genre information is acquired based on the web page that was browsed immediately before. When a suggestion word extracted based on information is presented to a user as a short-term suggestion word, there is a high possibility that a user's desired search query can be presented as a suggestion word.

In the information processing apparatus described above, the information related to the genre based on the web page may include a search query input by the user to browse the web page.
Thereby, it can be estimated what intention the user has viewed the web page.

The genre information estimation unit of the information processing apparatus described above, when a web page in which at least part of the information related to the genre is the same genre is browsed continuously for the predetermined period, It may be estimated as follows.
This increases the possibility of appropriately estimating the genre that the user is interested in immediately before.

In the information processing apparatus described above, the genre information estimation unit includes the short-term attention genre when information on the genre based on each of the web pages browsed at a predetermined number of times before the search is inconsistent. The extraction unit may not extract the short-term suggest word.
This prevents the presentation of an erroneous suggestion word that the user is not interested in.

In the above information processing apparatus, when a plurality of genres are extracted as genres in which the user is interested from the information related to the genre, the genre information estimation unit is the lowest genre among the plurality of genres. May be estimated as the short-term attention genre.
Accordingly, it is possible to increase the possibility that a suggestion word having a long number of characters or a suggestion word including a plurality of keywords can be presented at a stage where the user's input character string is small.

In the information processing apparatus described above, the extraction unit may extract a normal suggest word without using the short-term attention genre, and the presentation unit may present both the short-term suggest word and the normal suggest word to the user. Good.
As a result, the user can select an appropriate search query from both the short-term suggest word corresponding to the short-term attention genre, which is a relatively narrow range, and the normal suggest word.

In the information processing apparatus described above, when the web page browsing is not performed a predetermined number of times during the predetermined period, the extraction unit extracts a normal suggest word without using the short-term attention genre, and the presentation unit Ordinarily, a suggestion word may be presented to the user.
Thereby, the type of suggestion word presented according to the situation is selected.

The information processing apparatus described above may include a suggestion word management unit in which phonetic information and genre information are linked.
As a result, processing such as extraction of a suggestion word is made efficient.

In the information processing apparatus described above, the predetermined period may be shorter than a predetermined time.
Thereby, possibility that an optimal short-term attention genre will be estimated can be raised.

In the information processing apparatus described above, the predetermined period may be a period between browsing a web page a predetermined number of times before the search.
This corresponds to the case where the web page browsed before the search has a large amount of text.

According to the information processing method of the present invention, the user can input the web page as information related to a genre based on text information input by the user for a search and a web page browsed by the user during a predetermined period before the search. genre information estimating step of estimating a genre which the user having the taste Xing as casting eyes genre and information acquisition step of acquiring, and information including the search query entered in order to view, from the information relating to the genre, the an extraction step of extracting sub ingest word based on the text information and the previous Kichu eyes genre, a presentation step of presenting a sub ingest word and the extracted user, the one in which the information processing device executes.
In addition, the second information processing method according to the present invention includes text information input by a user for search, information on a genre based on a web page browsed by the user during a predetermined period before the search, , And an attention genre in which the user is interested in the same genre when a web page in which at least a part of the information related to the genre is the same genre is browsed continuously in the predetermined period. The information processing apparatus executes a genre information estimation step estimated as: an extraction step for extracting a suggestion word based on the text information and the attention genre; and a presentation step for presenting the extracted suggestion word to a user It is.
The information processing method of these, the process for the user to properly Suggest keywords newly interested is executed.

The program according to the present invention is a program that causes a computer device to execute processing executed as the information processing method.
A storage medium according to the present invention is a storage medium that stores the above-described program in a computer-readable manner.

  According to the present invention, it is possible to appropriately suggest a keyword that a user is newly interested in.

It is a figure which shows the whole structure of embodiment of this invention. It is a block diagram of the computer of this embodiment. It is a block diagram of the suggestion word server of this Embodiment. It is a figure which shows an example of the information memorize | stored in suggest word DB. It is a figure for demonstrating the flow of the whole process. It is a flowchart which shows the example of a suggestion word presentation process. It is a figure for demonstrating the presentation aspect of a suggestion word. It is a flowchart about the 1st example of a genre relation information acquisition process. It is a flowchart about the 2nd example of a genre relation information acquisition process. It is a flowchart about the 1st example of a short-term attention genre estimation process. It is a flowchart about the example of the process performed by a short-term attention genre estimation process. It is a flowchart about the example of the process performed by a short-term attention genre estimation process. It is a flowchart about the example of the process performed by a short-term attention genre estimation process. It is a flowchart about the 2nd example of a short-term attention genre estimation process. It is a flowchart about the 3rd example of a short-term attention genre estimation process. It is a flowchart about the 4th example of short-term attention genre estimation processing.

  Hereinafter, embodiments will be described in the following order.

<1. Overall configuration>
<2. Hardware configuration of computer device>
<3. Suggested server functional configuration and DB>
<4. Overall flow>
<5. Suggest word presentation process>
<6. Example of genre related information acquisition processing>
[6-1. First example]
[6-2. Second example]
<7. Example of short-term attention genre estimation processing>
[7-1. First example]
[7-2. Second example]
[7-3. Third example]
[7-4. Fourth example]
<8. Summary>
<9. Program>

<1. Overall configuration>

A configuration of the entire network system including the search server 7 according to the present embodiment will be described with reference to FIG.
The network system 1 functions as an EC (EC: Electronic Commerce (EC)) system.
The network system 1 is configured such that a shopping site management system 3 and a plurality of user terminals 4 can communicate with a plurality of store terminals 5 via a communication network 2 such as the Internet.

The configuration of the communication network 2 shown in FIG. 1 and the like is not particularly limited. In addition to the Internet described above, for example, an intranet, an extranet, a LAN, a CATV (Community Antenna TeleVision) communication network, a virtual private network ( Virtual Private Network), telephone line network, mobile communication network, satellite communication network, etc. are assumed.
Various examples of the transmission medium constituting all or part of the communication network 2 are also assumed. For example, IEEE (Institute of Electrical and Electronics Engineers) 1394, USB (Universal Serial Bus), power line carrier, telephone line, etc., infrared, IrDA (Infrared Data Association), Bluetooth (registered trademark), 802.11 wireless It can also be used wirelessly, such as mobile phone networks, satellite lines, and digital terrestrial networks.

The shopping site management system 3 includes a shopping server 6, a search server 7, a suggest server 8, a member DB (Database) 50, a product DB 51, a store DB 52, and a suggest word DB 53 each configured by a computer device. These devices can communicate with each other via a network such as a LAN (Local Area Network).
The network for enabling communication between the devices constituting the shopping site management system 3 is not particularly limited, as with the communication network 2.

  The shopping site management system 3 provides a user with a website (shopping site) of a virtual shopping mall managed by the shopping server 6 or the like. There are a plurality of stores (members of a virtual shopping street) in the shopping site. The staff of each store registers the product of the store in the shopping site management system 3 (for example, the shopping server 6) via the computer device as the store terminal 5, so that information on various products of various stores is stored in the shopping site. Uploaded above. The user can purchase a desired product by accessing the shopping site through the user terminal 4.

  Here, when using the shopping site, the user can perform membership registration for receiving various services provided by the shopping site management system 3. At the time of member registration, the user registers necessary information such as a user ID (Identification), product destination information (address information), and a credit card number. The user logs in to the shopping site with the registered user ID, thereby saving the trouble of inputting necessary information again when purchasing a product on the shopping site. The shopping server 6 performs various processes for managing these member information.

The shopping server 6 performs various processes based on the HTTP request sent from the user terminal 4. For example, generation and transmission of various web page data (for example, a product web page, a shopping cart web page, an order web page, etc.), a purchase process according to an order confirmation operation by the user, and the like are executed.
The web page data is, for example, a structured document file such as HTML (Hyper Text Markup Language) or XHTML (Extensible HyperText Markup Language). The structured document file describes text data such as product names and descriptions, image data such as images related to products, and their arrangement and display mode (character color, font, size, decoration, etc.).
Examples of the web page include a login page for allowing the user to input login information, a search page for searching for products, and a product detail page for displaying detailed information on products.

  On the web page provided by the shopping server 6, an input field for inputting a search query for searching for a product is arranged as a search form, and the user can input a keyword for search on the search form. . A search button is arranged in the vicinity of the search form, and the user can issue a search execution instruction based on an input keyword to the search form by operating the search button. In response to this execution instruction, the user terminal 4 makes a search request with the input keyword as a search query to the search server 7.

The search server 7 functions as a so-called search engine, receives a search query specified by a search request from the user terminal 4, and searches from various product web pages managed by the shopping server 6 based on the search query. Search for product web pages related to the query.
The search server 7 is a product web page related to the received search query, for example, a search query, based on the product information managed by the shopping server 6 or the character information (text information) included in the product web page file. Search for product web pages that contain text information that matches. For efficient search processing, a search query and a search result may be linked in advance and stored in a DB or the like. In this case, the search server 7 may acquire a search result corresponding to the search query from the DB or the like and transmit it to the user terminal 4.

As will be described later, in the network system 1 according to the present embodiment, when a character is input to the search form on the web page provided by the shopping server 6, a suggestion word corresponding to the input character to the search form is displayed. Presented on the web page. For example, suggested words such as “golf wear” and “golf ball free shipping” are presented according to the input character “golf”.
When one of the presented suggested words is selected by the user, the user terminal 4 makes a search request using the selected suggested word as a search query to the search server 7, and the search server 7 obtains the search result. A search result page for presenting is transmitted to the user terminal 4.
As a result, the user can execute a search using the search query without inputting all characters in the search query to be input, thereby reducing the user's character input burden when searching for the target product. It has been.

The suggest server 8 is an information processing apparatus that executes various processes related to the presentation of a suggest word. Specific processing will be described later.
As can be understood from the above description, a suggestion word is an option that is inferred from a character string input in a search form and presented as a completed form of a search character string that a user will input. is there. Specifically, it is “golf wear” or “golf ball” presented when “go” is entered, and “golf wear” or “golf watch” presented when “golfou” is entered. It is. In addition, a character string in which a plurality of words such as “new golf ball” and “used golf ball” presented when “golf ball” is input is one of the suggested words.

  The user terminal 4 is a computer device provided with a web browser. Examples of the user terminal 4 include a high-performance mobile phone (smart phone), a mobile phone, a personal digital assistant (PDA), a portable or stationary personal computer (PC), and the like. The kind of terminal 4 is not limited to these.

  The user terminal 4 requests the transmission of desired web page data or predetermined processing by transmitting an HTTP (Hypertext Transfer Protocol) request to the shopping site management system 3 (particularly, the shopping server 6 or the search server 7). Further, the user terminal 4 receives the web page data sent in response to the HTTP request and displays it by the web browser. Thereby, the user using the user terminal 4 can browse and operate a desired web page.

  The store terminal 5 is a terminal used by a seller who sells products using an electronic commerce service provided by the shopping site management system 3.

In the user terminal 4 and the store terminal 5, various transmission / reception processes and display processes are executed as necessary. The user terminal 4 and the store terminal 5 are, for example, a PC (Personal Computer), a feature phone, a PDA (Personal Digital Assistants) having a communication function, or a smart device such as a smartphone or a tablet terminal.
Note that the user terminal 4 and the store terminal 5 only use names for convenience depending on whether the person to be used is a person who is in a position to purchase a product or a person who is in a position to sell a product. The terminal configuration is the same.

In the above description, the user terminal 4 directly transmits a search request to the search server 7, so that the search using the input keyword or the selected suggest word as a search query is performed. A search query such as an input keyword or a selected suggest word may be transmitted to the search server 7 via the shopping server 6. In this case, the shopping server 6 receives the search result information from the search server 7, generates web page data representing the search result based on the received search result information, and transmits the generated web page to the user terminal 4. You may make it do.

<2. Hardware configuration of computer device>

2, the devices (member DB 50, product DB 51, store DB 52, suggest word DB 53, user terminal 4, and store terminal 5) including the shopping server 6, the search server 7, and the suggest server 8 shown in FIG. A CPU (Central Processing Unit) 101 of a computer device that constitutes a computer performs various processes according to a program stored in a ROM (Read Only Memory) 102 or a program loaded from a storage unit 108 to a RAM (Random Access Memory) 103. Run. The RAM 103 also appropriately stores data necessary for the CPU 101 to execute various processes.
The CPU 101, ROM 102, and RAM 103 are connected to each other via a bus 104. An input / output interface 105 is also connected to the bus 104.
An input unit 106, an output unit 107, a storage unit 108, and a communication unit 109 are connected to the input / output interface 105.
The input unit 106 includes a keyboard, a mouse, a touch panel, and the like.
The output unit 107 includes a display made up of an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), an organic EL (Electroluminescence) panel, etc., and a speaker.
The storage unit 108 includes an HDD (Hard Disk Drive), a flash memory device, or the like.
The communication unit 109 performs communication processing and inter-device communication via the network 2.
A media drive 110 is also connected to the input / output interface 105 as necessary, and a removable medium 111 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is appropriately mounted, and information can be written to the removable medium 111. Reading is performed.

In such a computer device, data and programs are uploaded and downloaded by communication by the communication unit 109. Data and programs can be exchanged via the removable medium 111.
As the CPU 101 performs processing operations based on various programs, each device (member DB 50, product DB 51, store DB 52, suggest word DB 53, user terminal 4, user terminal 4, etc.) including the shopping server 6, the search server 7, and the suggest server 8. And necessary information processing and communication as the store terminal 5) are executed.
In addition, the information processing apparatus which comprises each apparatus (Member DB50, goods DB51, store DB52, suggest word DB53, user terminal 4, and store terminal 5) including the shopping server 6, the search server 7, and the suggestion server 8 is as follows. The computer device as shown in FIG. 2 is not limited to a single computer device, and a plurality of computer devices may be configured as a system. The plurality of computer devices may be systemized by a LAN or the like, or may be arranged at a remote place by a VPN using the Internet or the like. The plurality of information processing apparatuses may include an information processing apparatus as a server group (cloud) that can be used by the cloud computing service.

<3. Suggested server functional configuration and DB>

FIG. 3 shows a functional configuration and various DBs as a suggest server 8 configured by one or a plurality of information processing apparatuses.
Each function as the suggest server 8 is a function realized by processing executed by the CPU 101 in accordance with the program in the information processing apparatus. However, all or some of the processes described below may be realized by hardware.
Further, when each function is realized by software, each function need not be realized by an independent program. Processing of a plurality of functions may be executed by one program, or one function may be realized by cooperation of a plurality of program modules.

The suggest server 8 includes a management unit 8a, an information acquisition unit 8b, a genre information estimation unit 8c, an extraction unit 8d, and a presentation unit 8e.
The management unit 8a executes various processes for managing the suggestion word to be presented to the user. Specifically, a process for registering a new suggestion word in the suggestion word DB 53, a process for changing the registration contents, and the like are performed.

The information acquisition unit 8b executes a process of acquiring information about input characters input by the user. In particular, in this embodiment, not only for each keyword for which input has been completed, but also for each time a character is input by the user (for example, every time an input key on the keyboard is pressed) get.
Moreover, the information acquisition part 8b acquires the information of the web page browsed before the input of the search query. The information on the web page acquired by the information acquisition unit 8b in the present embodiment is information on the web page at which genre information can be estimated. For example, URL information may be used as long as genre information can be estimated from URL (Uniform Resource Locator) information, or keyword information linked to a web page. Further, it may be genre information indicating what genre the web page is based on. Further, it may be a search query input for browsing a web page.

The genre information estimation unit 8c estimates a genre based on a web page that the user is browsing immediately before inputting a search query. Some examples are described below. In the following description, such a web page is referred to as a “previous browsing web page”.
The extraction unit 8d is suitable for presenting to the user from the suggestion word DB 53 based on the information related to the search query acquired by the information acquisition unit 8b (for example, a character string in the middle of input) and the genre information estimated by the genre information estimation unit 8c. A simple suggestion word.
The presentation unit 8e presents the suggestion word extracted by the extraction unit 8d to the user terminal 4. At this time, the order of presentation of the extracted suggest words may be determined, and priority information may be attached and transmitted to the user terminal 4 for display based on the order of presentation.

In order to realize the various functions described so far, the shopping site management system 3 includes each DB shown in FIG.
Each DB shown in FIG. 3 may be realized in any form as long as each information processing apparatus constituting the shopping site management system 3 can be accessed as necessary. For example, all DBs may be formed in the storage unit in the same system as the shopping site management system 3, or a part or all of each DB may be provided in a computer system such as a separate or remote place. Good. Of course, each DB does not need to be formed in one device (for example, one HDD). Each DB does not need to be configured as one DB. For example, information stored in the member DB 50 may be stored and managed by a plurality of member DBs (for example, a login member DB and a transaction member DB). Each DB described below is merely an example of a storage unit for information related to the processing of the embodiment in the form of one DB.

  In the member DB 50, information on users who use various services provided by the shopping site management system 3 is stored for each user. For example, personal information such as a login password, name, age, gender, annual income, address, e-mail address, hobby, etc. is linked and stored for one user ID that can identify one user. Further, payment information, purchase history of products, favorite registration information of products, and the like may be stored. Of course, it is not always necessary to link all these pieces of information to one user ID, and there is no information that the user does not intend to disclose.

The product DB 51 stores information about each product that can be bought and sold via the shopping site management system 3. For example, for a product ID that can uniquely identify a product, product genre, product image, manufacturer (maker) information, model number information given by the manufacturer, sales start date, handling product provider information, inventory information, A merchandise introduction sentence is linked and stored.
The product image information may be image data itself or link information (URL information or the like) of stored image data.
In addition to the above, the product DB 51 may store production locations, product specifications (color, size, performance information), and the like.
The product DB 51 also stores web page data such as product pages.

The store DB 52 stores information on member stores that are open on the shopping site.
For example, with respect to a store ID that can uniquely identify each store, a login password, a store name, a store logo, a product ID of a handled product (product-specific code information provided for each product by the shopping site management system 3), Information such as the product page URL, the price for each product, the URL of the store page, and the contact information (telephone number, e-mail address, etc.) is linked and stored.
The product page URL is a URL attached to each product page, and even if the product is the same, if the store is different, a different product page URL is attached.
The store logo information may be image data itself or link information (URL information) of stored image data.
These pieces of information are registered by an operator in the shopping site management system 3 or a store employee.

The suggestion word DB 53 stores information related to the suggestion word presented when the user enters a search query. This will be specifically described with reference to FIG. FIG. 4 is an excerpt of the suggestion word used in the description of the present embodiment from the suggestion words stored in the suggestion word DB 53.
One suggestion word includes an S_ID for uniquely identifying the suggestion word, character string information to be presented to the user, phonetic information used for searching the suggestion word, and first to nth genre information. Has been.
The phonetic information is, for example, a Hebon-style Roman alphabet.
The genre information is information representing a hierarchical structure of genres, and the first genre represents the largest category. That is, the second genre is a sub-genre of the first genre. Similarly, the third genre is a sub-genre of the second genre.
Specifically, the suggest word with ID = S — 2152 is “head cover”, and the phonetic information is “hedokaba-”. In the genre information, the first genre is “sports”, the second genre is “golf”, the third genre is “accessories”, and the fourth genre is “head cover”.

<4. Overall flow>

A description will be given of the overall processing flow in the case where characters are entered in the search field for product search after browsing several product pages using the shopping site provided by the shopping site management system 3 (see FIG. 5). ).

First, login processing is executed between the user terminal 4 and the shopping server 6.
Specifically, in step S101, the user terminal 4 executes a login screen request process in response to the user performing an operation for displaying a login page. The login screen request process is to request web page data related to a web page for inputting login information. When a login screen request is transmitted from the user terminal 4 to the shopping server 6 by the login screen request process, the shopping server 6 executes a login screen transmission process in step S201.
Thereby, for example, a login screen (web page) to the shopping site received from the shopping server 6 is displayed on the user terminal 4.

Next, in step S <b> 102, the user terminal 4 executes login information transmission processing for transmitting login information (user ID and password) input by the user to the shopping server 6. When the login information is transmitted from the user terminal 4 to the shopping server 6, the shopping server 6 executes an authentication process in step S202, and executes an authentication result notification process in the subsequent step S203.
Specifically, the shopping server 6 compares the user ID and password input on the user terminal 4 with information stored in the member DB 50 to determine whether or not the user can log in, and sends the authentication result to the user terminal 4. Notice. In addition, while returning an authentication result to the user terminal 4, you may transmit the web page data of the top page of a shopping site. Thereby, user authentication is performed and the top page of the shopping site is displayed on the user terminal 4.
Note that the series of flows shown in FIG. 5 shows a case where it is determined that login is possible in the authentication processing in step S202. If it is determined in step S202 that login is not possible, the user terminal 4 executes the process of step S102 again, and the shopping server 6 executes the process of step S202 accordingly.

Subsequently, based on the product search operation performed by the user, the user terminal 4, the suggest word server 8, and the search server 7 execute various processes. FIG. 5 shows an example in which the user performs a single search operation and selects and browses one product information from the presented search results.
When a user inputs characters into a search box (input form for inputting a keyword for search) for product search, the user terminal 4 accepts a search character string input operation in step S103, and the suggest server 8 In S301, a suggestion word presenting process for presenting a search query suitable for the user as a suggestion word is executed.
The search character string input operation accepting process in step S103 performs a search query in the middle of input each time the user performs character input as a search query once, for example, every time the user presses a keyboard key for character input. It transmits to the suggestion word server 8 as a search character string during input.
The search character string during input is transmitted not only when a key operation is performed to add a character but also when a key operation is performed to delete one input character. Specifically, when “g” is input, “g” is transmitted as an input search character string, and when “o” is pressed next, “G” is transmitted as an input search character string. When “k” is further pressed, “G” is transmitted as an input search character string. Next, when the user presses the backspace key, “go” from which “k” has been deleted is transmitted to the suggestion word server 8 as an input search character string.
The suggestion word server 8 executes a process for presenting a suggestion word to the user every time a search character string during input is received. Details will be described later.

Next, when the user finishes inputting a search query for product search and presses the search button, the user terminal 4 executes a search operation reception process in step S104.
In this process, the search query finally designated by the user is transmitted to the search server 7 and the search server 7 is notified that a search instruction has been made.
Note that the user may instruct a search by pressing the Enter key instead of pressing the search button.

Receiving the search query, the search server 7 executes search result presentation processing in step S401. In the search result presentation process, for example, information on a web page on which a plurality of products extracted as search results are displayed as a list is transmitted to the user terminal 4.
The user terminal 4 that has received the web page data as the search result executes a process for displaying the search result on a display device such as a monitor in step S105. Thereby, information on various products is presented as a search result on the user terminal 4.

Next, when the user performs an operation of selecting one product from the products listed as a search result, the user terminal 4 executes a product selection operation reception process in step S106. In this process, the shopping server 6 is requested for product page information of the selected product.
The shopping server 6 requested for the merchandise page information executes a process of transmitting the web page data of the merchandise page to the user terminal 4 in step S204.
The user terminal 4 that has received the web page data of the product page executes a process for displaying the product page in step S107.
Thereby, the user can browse the detailed information of the product selected by the user.

  Among the processes shown in FIG. 5, the process from step S103 to the process of step S107 is an example of a process performed for browsing one product page. In addition to this, the product page may be browsed directly without going through the search result page. In addition, a plurality of product pages may be browsed from one search result page.

Note that the processes after step S103 may be performed without the user logging in. For example, when the user is not logged in, the shopping server 6 issues a session ID to the user terminal 4 used by the user or software (such as a web browser) operating on the user terminal 4, and the session Within the validity period of the ID, each subsequent process may be executed assuming that the same user is using it.

<5. Suggest word presentation process>

A specific processing example of the suggestion word presentation processing shown in step S301 of FIG. 5 will be described with reference to FIG.
In order to present a suggestion word to the user, the search character string being input from the user terminal 4 needs to be sent. The suggestion word server 8 executes a process for determining whether or not the input search character string is received in step S501. Note that the process of step S501 does not need to be repeated until an input search character string is received, and the suggestion word server 8 is placed in a standby state until an input search character string is sent from the user terminal 4, The processing after step S502 may be executed by using the input search character string as a trigger.

  In step S502, the suggest word server 8 that has received the input search character string executes a process of acquiring genre relation information. The genre-related information is information that can be used to guess the genre that the user is most likely interested in, for example, information on the product genre that can be acquired from the browsed web page. A specific example will be described later.

The suggestion word server 8 that has acquired genre-related information (genre-related information) from the web page browsed by the user performs a process of estimating a product genre (short-term attention genre) that the user is interested in in the short term in the subsequent step S503. Run.
In the estimation of the short-term attention genre, there are a case where a specific genre is estimated as the user's short-term attention genre and a case where it is estimated that there is no genre that the user is paying attention to in the short-term. A specific example will be described later.

Subsequently, in step S504, the suggestion word server 8 determines whether there is a genre estimated as a short-term attention genre. If there is, the short-term suggest word extraction process is executed in step S505.
In the short-term suggest word extraction process, for example, a suitable suggest word is extracted from the suggest word DB 53 as shown in FIG. For example, when the user is inputting a search query and the search character string being input is “head”, if it can be estimated that the user is paying attention to the “climbing” genre in the short term, Suggestion words related to the “mountain climbing” genre are presented from among various suggestion words starting with “”. Specifically, as shown in FIG. 7A, the search character string “head” being input is displayed in the search box 9, and a suggestion word starting from “head” is displayed in the suggestion word presentation column 10 provided below the search character string. Presented. The suggestion word located at the top is a suggestion word with a higher presentation priority. As shown in the figure, the “headlight” based on the “mountain climbing” genre is presented at the top, and then the “mountain climbing” genre is also based. “Headlight LED” is presented.

In addition, as shown in parentheses in step S505 and FIG. 7A, not only a suggestion word belonging to “climbing”, which is a short-term attention genre, but also other suggestion words (in FIG. 7A, “head cover” belonging to the “golf” genre). ) May be presented together. This is only an example, and only suggest words belonging to the short-term attention genre may be presented. When the number of suggest words belonging to the short-term attention genre is less than a predetermined number, normal suggest words extracted by other methods are combined. May be presented. Normal suggestion words extracted by other methods may be, for example, suggestion words based on the genre of interest estimated from the user's usual purchase history or browsing history, etc. The search query may be a suggestion word, or may be a suggestion word extracted from the tendency of the user whose attribute information matches the user who is inputting the search query.
By submitting only the suggestion word corresponding to the short-term attention genre, it becomes possible to select the search query desired by the user without hesitation. Moreover, even when the user searches for products other than the products belonging to the short-term attention genre, the possibility that the user can present an appropriate suggestion word is increased by presenting the normal suggestion word extracted by another method together. be able to.

When the subdivided genre can be estimated as the short-term attention genre, it is effective to present the short-term suggest word based on the genre as fine as possible.
Specifically, even if a user who is usually interested in the “golf” genre browses a web page (such as a product page) related to the “golf” genre immediately before, the suggested word presented to the user does not change (see FIG. 7B). However, if the web page that the user was browsing just before was only about the “golf shoes” genre among the “golf” genres, the short-term attention genres are “golf” and “golf shoes”. At this time, since the “golf shoes” genre is a sub-genre of the “golf” genre (see FIG. 4), a suggestion word is presented based on the more subdivided “golf shoes” genre.
Specifically, as shown in FIG. 7C, when the input search character string is “golf”, the short-term suggest word is “golf shoes”, “golf shoes spikeless”, “golf shoes 鋲”, etc. It is based on the “golf shoes” genre. That is, while the input search character string is the same “golf” as compared to FIG. 7B, the search query shown in FIG. 7C is narrowed down to a more genre is presented as a suggestion word. Therefore, even if the same number of characters are input, it is possible to increase the possibility that the search query input by the user may be reduced by the amount of the suggested narrowed words.

Returning to the description of FIG. If there is no short-term attention genre in step S504, the suggestion word server 8 executes normal suggestion word extraction processing in step S506.
Subsequently, the suggestion word server 8 performs a suggestion word presentation process in step S507. Thereby, the short-term suggest word extracted at step S505 and the normal suggest word extracted at step S506 are presented on the display device of the user terminal 4.
Note that the processing in step S506 is not essential, and if there is no short-term attention genre, the suggestion word need not be presented. In this case, when the suggestion word is presented to the user, the short-term suggestion word is presented, so that the user can be impressed that the suggestion word that surely captures his / her short-term interest is presented.

<6. Example of genre related information acquisition processing>

Several examples of the genre relation information acquisition processing shown in step S502 of FIG. 6 will be described with reference to the drawings.

[6-1. First example]
A first example of genre relation information acquisition processing will be described with reference to FIG.
In the first example, the suggest word server 8 acquires the web page data browsed in the predetermined period immediately before in step S601. Subsequently, the suggest word server 8 extracts keywords from the web page data in step S602.
Here, the immediately preceding predetermined period will be described. An example of the immediately preceding predetermined period is shown below.
・ Case 1: Period from the predetermined time before the current search to the current search ・ Case 2: Period from browsing the n previous web pages to the current search before the current search

For example, in case 1, the predetermined time is 5 minutes, and the web page data of a web page browsed 5 minutes before the current search operation is targeted. Although the predetermined time may be one hour or the like, in many cases when the user searches (searches) for a genre that the user is interested in in the short term, there is a high tendency to browse a plurality of web pages continuously. Therefore, the keyword extraction in step S602 can be performed for a web page that the user has focused on in the short term by setting the time such as a few minutes to be a predetermined time.
When there is no web page browsed during the immediately preceding predetermined time, genre related information is not acquired. Further, when the number of web pages browsed during the predetermined time immediately before is small (for example, only one page), the genre related information may not be acquired.

  In case 2, among the web pages browsed before the current search operation, for example, a period from the previous five browses to the current search operation is set as a predetermined period. Since the last viewed web page is more likely to be a web page based on the user's short-term interests, keyword extraction is performed accordingly to increase the possibility of extracting the optimum short-term suggest word. be able to.

It is also conceivable that the predetermined period is determined by mixing case 1 and case 2. For example, if you browse 30 web pages in the last 5 minutes, consider the case 2 and use the 5 most recent web pages for keyword extraction to find the optimal short-term suggest word The possibility of being able to present is increased. In addition, when only 2 items are viewed in the last 5 minutes, in case 2, keywords are extracted from 5 web pages including web pages viewed more than 5 minutes ago. In addition, keyword extraction can be performed for the two most recently viewed items.
In addition, as a result of considering case 1 and case 2, when there are few keyword extraction target web pages, it is not necessary to acquire genre relation information.

The keyword extraction in step S602 is processing for extracting a keyword suitable for acquiring genre information from various texts included in the web page data and file names given to images. The keywords that appear in the web page data may be simply extracted, or may be extracted in consideration of the number of appearances.
Moreover, you may extract what the other keyword which belongs to the same genre or a similar genre appears. For example, if the keyword “fishing rod” appears but no other keyword belonging to the “fishing” genre appears, the keyword “fishing ball” appears without extracting the keyword “fishing rod”. In addition, when keywords belonging to the “golf” genre such as “golf wear”, “marker”, and “approach” appear, these keywords may be extracted.

[6-2. Second example]
A second example of the genre related information acquisition process will be described with reference to FIG.
In the second example, in step S611, the suggestion word server 8 acquires a search query specified when searching for a web page browsed during a predetermined period. The predetermined period takes into account the previous case 1 and case 2, or both case 1 and case 2.

  For example, if the web page immediately before browsing is a product page related to the “golf” genre and the search query specified for browsing is a search query such as “Golf Shoes A Company”, these search queries are used as genre-related information. get.

In the first example, an example in which keywords are extracted from information included in web page data (for example, an HTML file) has been described. However, even if genre-related information is acquired in consideration of both the first and second examples. Good. Specifically, both the search query specified for searching the immediately preceding browsing web page and the keyword information included in the immediately preceding browsing web page may be acquired as genre related information.

<7. Example of short-term attention genre estimation processing>

Each example of the short-term attention genre estimation process shown in step S503 of FIG. 6 will be described with reference to the drawings.

[7-1. First example]
A first example of short-term attention genre estimation processing will be described with reference to FIG.
In the first example, the suggestion word server 8 determines in step S701 whether or not genre relation information has been acquired. If it can be obtained, the suggestion word server 8 extracts genre information from the genre relation information and estimates it as a short-term attention genre in step S702. On the other hand, if it cannot be acquired, the suggestion word server 8 estimates in step S703 that there is no short-term attention genre.
As described above, the genre-related information is a keyword or a search query extracted from a web page. When extracting genre information from such keywords and search queries, for example, the suggestion word DB 53 shown in FIG. 4 may be used. Since each suggestion word stored in the suggestion word DB 53 is associated with genre information, a keyword or search query is regarded as a suggestion word and a search is performed from the suggestion word DB 53 to which genre the keyword or search query is assigned. It is possible to grasp whether it belongs. Of course, you may provide DB for exclusive use for acquiring genre information from a keyword or a search query.

In the genre related information acquisition process, a large number of keywords and search queries may be acquired as genre related information. If short-term attention genre estimation processing is performed based on a large number of genre-related information, there will be a large number of short-term attention genres, and the possibility that the search query desired by the user will be presented as a short-term suggest word may be reduced. is there.
Therefore, even if keywords are narrowed down in the genre-related information acquisition process as described above, the short-term attention genre may be estimated after narrowing down keywords and search queries again in the process of step S702.

For example, there are five web pages A to E immediately before browsing, and the genre-related information acquired from the five web pages is “golf”, “golf shoes”, “spikeless”, “golf ball”, It is assumed that the keyword is “golf wear”, “golf shirt”, “polo shirt”, “fishing rod”, or the like. Specifically, it is assumed that the keywords acquired from each page are as follows.
・ Page A: “Golf”, “Golf Shoes”
・ Page B: "Golf", "Spikeless", "Fishing rod"
-Page C: "Golf", "Golf Ball", "Golf Wear", "Golf Shirt"
・ Page D: "Golf", "Golf shoes", "Polo shirt"
・ Page E: "Golf", "Golf shoes", "Golf wear", "Polo shirt"
The suggest word server 8 converts these keywords into genre information. For example, the genre information extracted from each page is as follows.
・ Page A: “Golf”, “Golf Shoes”
・ Page B: "Golf", "Golf shoes", "Spikeless", "Fishing", "Fishing rod"
-Page C: "Golf", "Golf Ball", "Golf Wear"
-Page D: "Golf", "Golf Shoes", "Golf Wear"
・ Page E: "Golf", "Golf shoes", "Golf wear"
The keyword “polo shirt” itself is not necessarily the “golf” genre, but is converted to the “golf wear” genre by determining that it is a polo shirt as golf wear from the context and the like.
Here, the product genre extracted from the most web pages is “golf”, and the next is “golf shoes”. Therefore, it is considered that the short-term attention genre is “golf” and “golf shoes”. The “golf” genre associated with the most web pages may be the short-term attention genre. Further, “golf”, “golf shoes”, and “golf wear” that are genres acquired from a plurality of web pages may be estimated as short-term attention genres.
Considering the hierarchical structure of the above keywords, the “golf” genre is the largest genre information, and “golf shoes” genre, “golf ball” genre, and “golf wear” genre are subgenres of the “golf” genre. positioned. Further, a “spikeless” genre is positioned as a further subgenre of “golf shoes”. Therefore, if the most popular genre “golf” genre is a short-term attention genre, the short-term suggest word presented to the user based on the browsing behavior of the most recent user is no different from the suggest word normally presented to the user. It may become a thing.
Thus, for example, among the genres satisfying a predetermined condition such as genres acquired from a plurality of web pages, the lowest genre may be set as the short-term attention genre. According to this example, “golf shoes” and “golf wear”, which are lower genres among the genres acquired from a plurality of web pages, may be used as short-term attention genres. As a result, a search query based on a narrow genre that is deeply related to the web page that was just viewed for the user is presented as a short-term suggest word, so the number of characters input by the user can be reduced. , Can increase convenience.

Specifically, this can be realized by the suggestion server 8 performing each process shown in FIG. 11 instead of step S702 of FIG.
First, the suggestion word server 8 extracts genre information from genre relation information in step S801.
Next, the suggestion word server 8 selects a short-term attention genre candidate from the extracted genre information in step S802. As a method for selecting candidates for the short-term attention genre, various methods can be considered in the same manner as the short-term attention genre estimation method performed in step S702.
Subsequently, in step S803, the suggestion word server 8 estimates the genre having the lowest genre hierarchy as the short-term attention genre among the candidates for the short-term attention genre. When there are a plurality of short-term attention genre candidates in the same hierarchy, all of them may be estimated as short-term attention genres, or one of the plurality of candidates may be selected. When selecting one, for example, the genre information extracted from the keyword or the search query that the user is most interested in when the genre information is extracted from the genre relation information in step S801 may be used as the short-term attention genre. . Specifically, it is genre information extracted from the keyword that appears most on the immediately preceding browsing web page.

Note that it may be desirable to have only one short-term attention genre. For example, if a suggestion word is extracted from a plurality of short-term attention genres, the directionality of the suggestion word presented to the user will be blurred. Specifically, when “golf shoes” and “golf wear” are used as short-term attention genres, a limited number of short-term suggest words on the display area of the user terminal 4 include “golf” as the search character string during input. "Golf shoes", "Golf shoes spikeless", "Golf wear", "Golf wear men" and so on.
However, by narrowing the short-term attention genre to “Golf Shoes”, the short-term suggest word presented to the user is “Golf Shoes”, “Golf Shoes Spikeless”, “Golf Shoes 鋲”, “Golf Shoes Spikeless A Company Or the like, it can be specialized in one genre. In other words, it is possible to increase the possibility that the search word desired by the user who is paying attention to “golf shoes” is included in the suggestion word. Moreover, it becomes possible to present a composite search query including more keywords such as “Golf Shoes Spikeless Company A”. Thereby, the user can select a long search query such as “golf shoes spikeless company A” by simply inputting “golf”, and can further contribute to shortening the input time.

  In order to make the lower genre a short-term attention genre, it is conceivable to change the selection criteria of the short-term attention genre according to the genre hierarchy. For example, if the genre is the highest genre, the genre based on the keyword extracted from 80% of the web pages viewed immediately before is set as the short-term attention genre. If the genre is one lower level, the genre is based on a keyword extracted from 60% of the last-viewed web pages. In the case of another lower genre, it is based on a keyword extracted from 40% of the immediately preceding browse web pages. By doing in this way, since a lower genre becomes easy to be estimated as a short-term attention genre, the possibility of presenting a suggestion word that is highly convenient for the user can be increased. When there are a plurality of short-term attention genres extracted in this way, as described above, the shortest genre that is the lowest genre may be used as the short-term multi-item genre.

Specifically, the suggestion server 8 can change the selection criteria of the short-term attention genre according to the genre hierarchy by performing each process shown in FIG. 12 instead of step S702 in FIG.
First, the suggestion word server 8 extracts genre information from genre relation information in step S811. This process is the same as the previous step S801.
Subsequently, in step S812, the suggestion word server 8 selects a short-term attention genre candidate based on a criterion corresponding to the genre hierarchy. If the criteria according to the hierarchy is the highest genre as described above, the genre based on the keyword extracted from 80% of the previous browsing web pages may be the short-term attention genre, or the short-term attention genre candidate It is good also considering the frequency | count that it appeared as a keyword and search query as a basis according to a hierarchy. Specifically, it can be considered that the higher-level genres do not become candidates for short-term attention genres unless they appear more frequently in keywords and search queries.

In addition, when the genre of interest of the usual user is the same, a genre that is a lower genre than the genre may be set as the short-term attention genre. For example, when a user who normally browses a web page related to “Golf” or purchases a product, and presents a suggested word specialized for the user, “Golf Wear” or “Golf Ball” Golf related keywords such as are suggested.
In such a situation, when a user browses a product or article of spikeless golf shoes, the genre of “golf” or “golf shoes” is not simply set as a short-term attention genre, but “golf shoes” It is desirable to set the “spikeless” genre, which is a subordinate genre, as a short-term attention genre. Thereby, it is possible to increase the possibility of presenting a search query desired by the user as a suggestion word.

  Conversely, when a user who does not normally browse products or web pages of the “fishing” genre browses a web page related to “spinning reels”, the genre is a lower genre of the “reel” genre, which is a lower genre of the “fishing” genre. If a certain “spinning reel” genre is set as a short-term attention genre, there is a possibility that a suggestion word in which a genre with which the user has little knowledge is dug down unnecessarily is presented. Therefore, in such a case, a short-term suggest word that is appropriate and highly convenient for the user can be presented by setting the “fishing” genre and the “reel” genre, which are higher genres, as the short-term attention genre.

Specifically, the suggestion server 8 can estimate the short-term attention genre in consideration of the usual user's interest genre by performing each process shown in FIG. 13 instead of step S702 in FIG.
First, the suggestion word server 8 extracts genre information from genre relation information in step S821. This process is the same as the previous step S801.
Next, in step S822, the suggest word server 8 determines whether or not the usual interest genre is included in the extracted genre information.
If it is included, the suggestion word server 8 estimates the genre having the lowest genre hierarchy as a short-term attention genre in step S823. In addition, when various genres are included in the extracted genre information, the lowest genre among the genres included in the lower level of the usual interest genre is estimated as the short-term attention genre. Specifically, when the usual genre of interest is “golf” and the extracted genre information is “golf”, “golf shoes”, or “spinning reel”, the lowest genre is the fourth genre. The “spinning reel” genre (see FIG. 4) is “golf” because the usual genre of interest is “golf”, so “golf shoes”, which is the lowest in the genre included in the “golf” genre, is the short-term attention genre. Presumed.

On the other hand, if the extracted genre information does not include the usual genre of interest, the suggestion word server 8 estimates the genre of the predetermined genre hierarchy as the short-term attention genre in step S824.
The predetermined genre hierarchy is, for example, a uniform second hierarchy (in the example of FIG. 4, “golf” genre or “climbing” genre), a third hierarchy (in the example of FIG. 4, “caddy bag” genre, “golf ball”). “Genre”). Moreover, you may respond | correspond to the number of subordinate genres. For example, if the number of subordinate genres of the “A” genre that is the first genre is small, the “A” genre that is the first genre may be estimated as the short-term attention genre, or the “B” genre that is the third genre. If the number of subordinate genres is large, the fourth genre may be estimated as the short-term attention genre.

[7-2. Second example]
A second example of short-term attention genre estimation processing will be described with reference to FIG.
In the second example, the suggestion word server 8 determines whether or not genre-related information is acquired in step S711. If not acquired, the suggestion word server 8 determines in step S714 that there is no short-term attention genre and ends the process.
If it has been acquired, the suggestion word server 8 determines in step S712 whether the keyword acquired from the immediately preceding browsing web page or the genre information obtained from the search query includes the same genre twice in succession.
Specifically, it is assumed that there are five web pages A to E just before browsing, and the genre information converted from the keywords acquired from each page is as follows. It is assumed that the pages are browsed in the order of pages A, B, C, D, and E.
・ Page A: “Golf”, “Golf Shoes”
・ Page B: "Fishing", "Fishing rod", "Fishing line"
-Page C: "Golf", "Golf Ball", "Golf Wear"
・ Page D: “Climbing”, “Trekking shoes”, “Water bottle”
・ Page E: "Golf", "Golf shoes", "Golf wear"
In this example, since web pages related to the same genre are not browsed continuously, the determination in step S712 is “No”. Therefore, the suggestion word server 8 determines that there is no short-term attention genre in step S714.
On the other hand, it is assumed that the genre information converted from the keywords acquired from each page is as follows.
・ Page A: “Golf”, “Golf Shoes”
・ Page B: "Fishing", "Fishing rod", "Fishing line"
-Page C: "Golf", "Golf Ball", "Golf Wear"
-Page D: "Golf", "Golf Shoes", "Golf Wear"
・ Page E: "Climbing", "Trekking shoes", "Water bottle"
In this example, since the page D including the information of the same genre is browsed after the page C is browsed, the determination in step S712 is “Yes”. Specifically, two genres of “golf” genre and “golf wear” genre are candidates for the short-term attention genre.

In step S713, the suggestion word server 8 that has determined “Yes” in step S712 executes a process of estimating the short-term attention genre from the short-term attention genre candidates.
As described in the first example, various examples are conceivable as to which genre is the short-term attention genre or both genres are the short-term attention genre.

[7-3. Third example]
A third example of short-term attention genre estimation processing will be described with reference to FIG.
In the third example, attention is paid to the immediately preceding two web pages. Specifically, the suggestion word server 8 determines whether or not genre-related information has been acquired in step S721. This process is the same process as step S711 in FIG.
Subsequently, in step S722, the suggestion word server 8 determines whether the genre information acquired from the web page browsed at the last predetermined number of times (two times in the example shown in FIG. 15) among the previous browse web pages includes the same genre. Determine whether or not.
Specifically, each page of the immediately preceding browsing web pages A to E is taken as an example. Assume that the genre information converted from the keywords acquired from each page is as follows. It is assumed that the pages are browsed in the order of pages A, B, C, D, and E.
・ Page A: “Golf”, “Golf Shoes”
・ Page B: "Fishing", "Fishing rod", "Fishing line"
・ Page C: "Climbing", "Trekking shoes", "Water bottle"
-Page D: "Golf", "Golf Ball", "Golf Wear"
・ Page E: "Golf", "Golf shoes", "Golf wear"
In this example, the last two last viewed web pages are page D and page E.
When the same genre can be acquired from page D and page E, the determination result of step S722 is “Yes”, and when it cannot be acquired, the determination result of step S722 is “No”. That is, in the above example, the determination result is “Yes”.

If the determination in step S722 is “Yes”, the suggest word server 8 executes a process of estimating the short-term attention genre from the short-term attention genre candidates in step S723. As described in the first example, various examples are conceivable as to which genre is the short-term attention genre or both genres are the short-term attention genre.
On the other hand, if the determination in step S722 is “No”, the suggestion word server 8 determines in step S724 that there is no short-term attention genre.

[7-4. Fourth example]
A fourth example of short-term attention genre estimation processing will be described with reference to FIG.
In the fourth example, attention is paid to the number of web pages viewed immediately before.
In step S731, the suggestion word server 8 determines whether or not the number of immediately preceding browsing web pages is equal to or greater than a predetermined number (for example, two).
If the number of web pages read immediately before is a predetermined number or more, the suggest word server 8 estimates the genre extracted from the genre relation information in step S732 as the short-term attention genre. This estimation may be any of the various examples described above.
If the number of web pages read immediately before is less than the predetermined number, the suggest word server 8 determines in step S733 that there is no short-term attention genre.

For example, when the predetermined time is 5 minutes and only one web page has been browsed in the 5 minutes before the current search, it is determined whether or not the user is going to perform a search related to the web page. Since it is difficult to determine appropriately, it is conceivable to perform a normal suggest word without setting a short-term attention genre. On the other hand, if you browse two web pages for 5 minutes before the search, and those web pages are based on the same product genre, this search also tries to search for products based on the same genre. Therefore, by setting a short-term attention genre and presenting a short-term suggest word, the user's input assistance is appropriately performed.

<8. Summary>

The suggestion word server 8 described in each of the above examples allows the user to view text information (part of the search query, search string being entered) input by the user for the search and a predetermined period before the search. Information acquisition unit 8b for acquiring information related to a genre (genre related information) based on the web page (web page viewed immediately before), and estimating a genre that the user is interested in from the information related to the genre as a short-term attention genre A genre information estimation unit 8c, an extraction unit 8d for extracting a short-term suggestion word based on text information and a short-term attention genre, and a presentation unit 8e for presenting the extracted short-term suggestion word to the user.
There is a high possibility that the web page browsed by the user during a predetermined period before the search is a web page that the user has recently been interested in. Acquiring genre information based on such a web page and presenting the suggested word extracted based on the genre information to the user as a short-term suggested word may be able to present the user's desired search query as the suggested word Therefore, it is possible to contribute to improvement of user convenience from the viewpoint of user input assistance.
For example, if the web page that the user performing the product search has viewed so far is a product that is biased to a specific genre, there is a high possibility that a product belonging to the same product genre will be searched for. Even in such a case, according to the above configuration, it is possible to present the user's desired search query as a suggestion word, and to shorten the user input time.
In addition, by limiting the genre information based on the web page that was browsed immediately before, the number of suggestion words presented to the user is selected rather than selecting the suggestion word from keywords belonging to all genres and presenting it to the user. Less. When the user selects from the suggested suggestion words, the number of suggested suggestion words is small, and it is highly possible that the keyword desired by the user is presented as a suggestion word. It is possible to select a suggestion word very efficiently.
Also, instead of simply acquiring genre information and extracting suggestion words from web pages that the user has browsed before, the period is limited to the web pages browsed during a predetermined period before the search. Suggestion words can be presented based on genres that are interested in the short term or genres in which the user's interest has increased rapidly. This makes it possible to flexibly suggest the search query that the user currently desires without being caught by the product genre that the user is usually interested in or the product genre that is commonly searched.
Of course, even if it is a genre that the user is always interested in, if you browse a web page of the same genre for a predetermined period, it will be presented as a suggestion word, so the normal suggestion word will not be presented at all Absent.
In addition, according to this structure, the presentation of the suitable suggestion word based on a short-term attention genre can be performed at the early stage of the input process of the search query by a user. As a result, it is possible to reduce the burden on the information processing apparatus that performs the input character acquisition processing and the suggestion word extraction / presentation processing performed for each character as well as the user input burden. For example, in comparison with the case where an appropriate suggestion word is presented after five characters are input, if an appropriate suggestion word is presented only by inputting one character, acquisition processing and suggestion word extraction performed for each subsequent input Since the information processing apparatus does not have to perform the processing / presentation process, the processing load on the information processing apparatus can be reduced.

Further, as described in the second example of the genre-related information acquisition process, FIG. 9 and the like, the information related to the genre based on the web page includes a search query input by the user for browsing the web page. May be.
In addition to extracting keywords from web pages browsed by the user and acquiring genre information on the web pages, genre information is extracted based on keywords (such as search queries) used for browsing web page searches. Thereby, the genre information in which the user was interested can be correctly estimated.
For example, in the case where information on “golf” and “fishing” is described on the web page browsed by the user, if information on the genre is acquired from the browse web page, both genres of “golf” and “fishing” Is estimated as a short-term attention genre that attracted the user's interest. However, if the user has reached the browsing of the web page as a result of using the search query “golf” (or a keyword such as “iron” related to “golf”), the user is “golf” rather than “fishing”. There is a high possibility of being interested in the genre. According to this configuration, since the genre information is estimated in consideration of the user's search query, the possibility that the search query desired by the user is suggested can be increased, and the convenience of the user can be improved. .

Furthermore, as described in the second example of the short-term attention genre estimation process and FIG. When browsed, the same genre may be estimated as a short-term attention genre.
The web pages browsed by the user during the predetermined period before the search are not all web pages that attract the user's interest. Therefore, estimating the genre information from the web page just by browsing for a predetermined period before the search may present a suggestion word based on the genre information that the user is not interested in even in the short term There is.
According to this configuration, when a plurality of web pages browsed in a predetermined period are of the same genre and the web pages are browsed continuously, the genre is selected as a short-term attention genre. Estimate as Thereby, possibility that the suggestion word based on the genre which the user was interested in can be raised.
It should be noted that the more web pages that have been browsed consecutively, the more immediately before the search (for example, browsing in the last minute or browsing in the last three times), the more short-term interest of the user It is possible to increase the possibility that an appropriate suggestion word can be presented.

Furthermore, as described in the third example of the short-term attention genre estimation process, the genre information estimation is performed when the information related to the genres based on the web pages browsed a plurality of times before the search is inconsistent. The unit 8c estimates that the short-term attention genre does not exist, and the extraction unit 8d may not extract the short-term suggest word.
Even if the user browses various web pages during a predetermined period before the search, if there is no genre common to these web pages, it is not preferable to estimate the genre that the user is interested in in the short term. There is a case. That is, there is a possibility that an appropriate short-term suggest word cannot be presented.
According to this configuration, when there is no genre common to a plurality of web pages viewed most recently (for example, the two web pages browsed in the last two times) among the web pages browsed before the search, the short-term suggest word By not extracting and presenting, it is possible to prevent an erroneous suggestion word that the user is not interested in from being presented.

In addition, as described in the first example of the short-term attention genre estimation process, when a plurality of genres are extracted from the genre-related information as the genre in which the user is interested, the genre information estimation unit 8c Of these, the lowest genre may be estimated as the short-term attention genre.
By estimating the lowest genre as a short-term attention genre, there is a high possibility that a search query or compound search query with a large number of characters will be presented as a suggestion word even if the search string entered by the user is small Therefore, it is possible to contribute to shortening the user input time. Further, the processing load on the user terminal 4 that performs the suggestion word presentation process and the like can be reduced so that the search query desired by the user can be presented as a suggestion word when the number of characters in the search character string being entered by the user is small. It becomes possible to do.

Further, as described in the suggestion word presentation process, FIG. 7A and the like, the extraction unit 8d extracts the normal suggestion word without using the short-term attention genre, and the presentation unit 8e displays both the short-term suggestion word and the normal suggestion word as the user. May be presented.
According to this configuration, when a short-term attention genre can be specified, a short-term suggest word is presented. Furthermore, a normal suggest word that does not depend on the web page viewed immediately before is also presented. Thereby, even if it is a case where a user performs a search unrelated to the web page browsed immediately before, even if it is a search related to the web page browsed immediately before, the suggestion word presentation process which can respond is performed be able to. Therefore, it is possible to save the user input and improve convenience.

Further, as described in the fourth example of the short-term attention genre estimation process and FIG. 16 and the like, if the web page has not been browsed a predetermined number of times during the predetermined period, the extraction unit 8d does not use the short-term attention genre. The normal suggest word may be extracted, and the presentation unit 8e may present the normal suggest word to the user.
If the web page has not been browsed a predetermined number of times such as twice in a predetermined period before the search, it is highly likely that it is difficult to appropriately estimate the short-term attention genre of the user. In such a case, the process switches to normal suggestion word presentation processing. This prevents a suggestion word that is not appropriate for the user from being presented.

Furthermore, as described with reference to the functional configuration of the suggest server and FIG. 3, a suggest word management unit 8a in which phonetic information and genre information are associated may be provided.
By managing the phonetic information and genre information as a single piece of information, it is possible to efficiently present a suggestion word according to the short-term attention genre and reduce the burden on the information processing apparatus. . In particular, when the normal suggest word is also presented to the user, the same DB can be referred to in the extraction of the short-term suggest word and the normal suggest word, so that the processing efficiency can be improved.

Then, as described in the first example of the genre related information acquisition process and FIG. 8 and the like, the predetermined period may be shorter than the predetermined time.
By setting the predetermined period before the search to be a fixed time, it is possible to appropriately estimate the genre that the user is paying attention to in the short term. For example, the short-term attention genre can be estimated more appropriately by setting the predetermined time to 5 minutes or 10 minutes.

Further, as described in the first example of the genre-related information acquisition process and FIG. 8 and the like, the predetermined period may be a period between browsing and searching for a web page a predetermined number of times ago.
There are cases where the web page browsed before the search has a large amount of text. A case where the user carefully reads the web page is also conceivable. In that case, if the predetermined period is set to a certain time, the browsing becomes outside the predetermined period even though the web page related to the genre that is focused on in the short term was browsed. There is a possibility that the short-term attention genre cannot be estimated. According to this configuration, since the short-term attention genre is estimated according to the web page that has been browsed a predetermined number of times before the search, it is possible to increase the possibility that the search query desired by the user is presented as a suggestion word. it can.
In addition, even if it was a genre that attracted attention in the short term, but it happened that other tasks happened before the search and time passed, forgetting what you were paying attention to When a keyword related to a short-term attention genre is presented as a suggestion word at the time of the search, it is possible to remember that the user has focused on the genre and to assist in performing an appropriate search. . In particular, since the short-term attention genre is unlikely to be universally noticed, the user is likely to forget, so that the use of this configuration can provide a high effect.

In each example described above, after browsing a web page presented by the shopping server 6 which is a part of the shopping site management system 3, a product search is performed on the web page presented by the same shopping server 6, and the shopping site management system is similarly used. 3, the search server 7 and the suggestion server 8 which are a part of 3 are used to search for products and present a suggestion word. Therefore, various types of information such as information on the web page read immediately before can be smoothly transmitted and received between the devices. . Specifically, it is possible to transmit and receive not only various types of web page data but also accompanying information. Therefore, processing such as keyword extraction is performed by preliminarily managing genre information linked to product pages and various special features pages, and estimating short-term attention genres based on the managed genre information. The above-mentioned various effects can be obtained.
However, even when a product search using the shopping site management system 3 is performed after browsing an external web page other than the web page under the management of the shopping site management system 3 without being caught by these configurations, If the external web page data (HTML data) can be acquired, or information such as a search query when browsing the web page can be acquired, the product genre can be specified by extracting a keyword from the web page data. Therefore, the various effects described above can be obtained.

In the above example, the product search is taken as an example. However, not only this but also when obtaining some information by the search, it is applicable if there is a search query input step.

<9. Program>

The program in each embodiment is a program to be executed by an arithmetic processing device (such as the CPU 101) provided in the suggest server 8.

This program computes an information acquisition procedure for acquiring text information input by a user for search and information related to a genre based on a web page browsed by the user during a predetermined period before the search. Let the device run.
Further, the arithmetic processing apparatus is caused to execute a genre information estimation procedure for estimating a genre in which the user has been interested immediately before from the information related to the genre as a short-term attention genre.
Further, the processing unit is caused to execute an extraction procedure for extracting a short-term suggest word based on the text information and the short-term attention genre.
Furthermore, the processing unit is caused to execute a presentation procedure for presenting the extracted short-term suggest word to the user.
In other words, this program is a program that causes the arithmetic processing unit to execute the process in step S301 in FIG. 5 and the processes described in FIG. 6 and FIGS.

The above-described suggest server 8 can be realized by such a program.
Such a program can be stored in advance in an HDD as a storage medium built in a device such as a computer device or a ROM in a microcomputer having a CPU. Alternatively, it can be stored (stored) temporarily or permanently in a removable storage medium such as a semiconductor memory, memory card, optical disk, magneto-optical disk, or magnetic disk. Such a removable storage medium can be provided as so-called package software.
Further, such a program can be installed from a removable storage medium to a personal computer or the like, or can be downloaded from a download site via a network such as a LAN or the Internet.

  DESCRIPTION OF SYMBOLS 1 Network system 2 Communication network 3 Shopping site management system 4 User terminal 5 Store terminal 6 Shopping server 7 Search server 8 Suggest server 8a Management part 8b Information acquisition part 8c Genre information estimation part 8d Extraction unit, 8e presentation unit, 50 member DB, 51 product DB, 52 store DB, 53 suggest word DB

Claims (11)

An information acquisition unit that acquires text information that is input by the user for search, and information related to a genre based on a web page that the user browses during a predetermined period before the search;
And genre information estimation unit that estimates a genre the user is interested as Note eyes genre from the information related to the genre,
An extraction unit that extracts the difference ingest word on the basis of the text information and before Kichu eyes genre,
And a presentation unit for presenting the service ingest word and the extracted user,
The information relating to the genre based on the web page is an information processing apparatus including a search query input by a user to browse the web page . An information acquisition unit that acquires text information that is input by the user for search, and information related to a genre based on a web page that the user browses during a predetermined period before the search;
A genre information estimation unit that estimates a genre in which the user is interested from information related to the genre,
An extraction unit that extracts a suggestion word based on the text information and the genre of interest;
A presentation unit for presenting the extracted suggestion word to a user,
The genre information estimation unit
When a web page in which at least a part of information related to the genre is the same genre is browsed continuously for the predetermined period, the same genre is estimated as the attention genre.
Information processing device. When the information related to the genre based on each of the web pages browsed at a predetermined number of times before the search is inconsistent,
The genre information estimation unit estimates that there is no pre-Kichu eyes genre,
The extraction unit, before the information processing apparatus according to any one of claims does not extract the hexa ingest word 1 or claim 2. When a plurality of genres are extracted as genres that the user is interested in from the information related to the genres,
The genre information estimating unit, the information processing apparatus according to claim 1 or claim 2 for estimating what the lowest genre out of the plurality of genres as a pre Kichu eyes genre. The extractor also extracts typically suggestion word without using the pre-Kichu eyes genre,
The presentation unit is an information processing apparatus according to claim 1 or claim 2 presents both the normal suggestion word before and hexa ingested word to the user. If a predetermined number of web page browsing has not been performed during the predetermined period,
The extraction section extracts a normal suggestion word without using the pre-Kichu eyes genre,
The presentation unit is an information processing apparatus according to claim 1 or claim 2, presents the user with the usual suggestions word. The information processing apparatus according to claim 1, further comprising a suggestion word management unit in which phonetic information and genre information are linked. The information processing apparatus according to claim 1, wherein the predetermined period is shorter than a predetermined time. The information processing apparatus according to claim 1 , wherein the predetermined period is a period between browsing of the web page a predetermined number of times before the search. Search query entered by the user for browsing the web page as text information entered by the user for the search and information relating to a genre based on the web page browsed by the user during the predetermined period before the search Including information , and an information acquisition step for acquiring
And genre information estimation step of estimating a genre which the user with a taste Xing as Note eyes genre from the information related to the genre,
An extraction step of extracting the difference ingest word on the basis of the text information and before Kichu eyes genre,
An information processing method by the information processing device executes the presenting step, the presenting the sub ingest word and the extracted user. An information acquisition step of acquiring text information that the user has entered for search, and information related to a genre based on a web page that the user browsed during a predetermined period before the search;
A genre information estimation step of estimating the same genre as an attention genre in which the user is interested when a web page in which at least a part of the information related to the genre is the same genre is browsed continuously in the predetermined period;
An extraction step of extracting a suggestion word based on the text information and the noted genre;
Presenting the extracted suggested words to the user; and
An information processing method executed by the information processing apparatus.

Images