JP6443431B2 - Information processing apparatus, information processing method, and information processing program - Google Patents

Description

  The present invention relates to a technique for selecting an item related to an item.

In recent years, with the advancement of digital technology and network technology, items such as digital contents and articles are distributed and sold using a network. Opportunities to collect information through search sites are also increasing. Along with this, there is an increasing need for a technology that selects and provides information on items that are likely to be of interest to the user from among a large number of items and information that matches the purpose.
For example, there is a recommendation process using collaborative filtering as one of the techniques for recommending items. In the recommendation process using collaborative filtering, a lot of usage information such as browsing history and usage history is accumulated, and an item set related to a certain item is recommended using the usage information.
In addition, there is a growing need for a technique for recommending items even when there is little usage information, such as when services are started.
For example, Patent Document 1 discloses a technique for recommending even an item with little usage information.

JP2012-059183A

Patent Document 1 preferentially presents other items having many of the same keywords as highly relevant keywords assigned to the viewed item when the related item of the item viewed by the user is presented. Thus, it is possible to present related items even when the usage information is small.
However, in Patent Document 1 described above, since related items have many of the same keywords, it is difficult to present items that may not be matched but may be related.
In other words, since it is difficult to recommend various items, the user may feel that “only similar items are recommended” and lose interest in the recommendation.

  An object of this invention is to provide the technique which can recommend various items.

In order to solve the above-described problems of the related art, the present invention identifies a user entity identifier for identifying a user entity and an item used by the user entity in an information processing apparatus that selects an item related to a reference item. A usage information storage unit that stores usage information associated with item identifiers; an item information storage unit that stores attribute correspondence information in which item identifiers and attribute identifiers for identifying item attributes are associated; and Based on the information and the attribute correspondence information, an attribute relevance level, which is a relevance level between the attribute identifier and the item identifier, is calculated, and each item identifier is related to the item identifier based on the calculated attribute relevance level. Related attribute information that associates each item identifier with the selected related attribute. A related attribute selection unit, a related attribute information storage unit that stores related attribute information generated by the related attribute selection unit, and related attribute information that includes a reference item identifier that is an item identifier of the reference item. It is a set of item identifiers acquired from the item information storage unit and acquired from the item information storage unit, attribute correspondence information including any of the attribute identifiers acquired from the storage unit and included in the acquired related attribute information A related candidate item set is formed, a degree of association between each item identifier included in the relevant candidate item set and the reference item is calculated, and an item related to the reference item is selected based on the calculated degree of association An information processing apparatus including an item selection unit is provided.
In addition, in order to solve the above-described problems of the related art, the present invention provides an information processing method for selecting an item related to a reference item, a user entity identifier for identifying a user entity, and an item used by the user entity. Item information storing attribute correspondence information in which a usage information storage step for storing usage information in association with an item identifier to be identified is stored in a usage information storage unit, an item identifier, and an attribute identifier for identifying an item attribute Based on the item information storage step stored in the section, the usage information and the attribute correspondence information, the attribute association degree is calculated as an association degree between the attribute identifier and the item identifier, and based on the calculated attribute association degree For each item identifier, a related attribute that is an attribute identifier related to the item identifier is selected and each item identifier is selected. A related attribute selection step of creating related attribute information in which a bespoke and the selected related attribute are associated with each other, and a related attribute information storage step of storing the related attribute information generated in the related attribute selection step in the related attribute information storage unit And associated attribute information including a reference item identifier that is an item identifier of the reference item is acquired from the related attribute information storage unit, and attribute correspondence information including any of the attribute identifiers included in the acquired related attribute information A related candidate item set that is a set of item identifiers acquired from the item information storage unit and included in the acquired attribute correspondence information is formed, and the degree of association between each item identifier included in the related candidate item set and the reference item A function for selecting an item related to the reference item based on the calculated relevance level. An information processing method characterized by having a item selection step.
In addition, in order to solve the above-described problems of the related art, the present invention is used by an information processing program for causing a computer to select an item related to a reference item, a user entity identifier for identifying a user entity, and the user entity. Attribute correspondence information in which a usage information storage step for storing usage information in association with an item identifier for identifying an item in the usage information storage unit, an item identifier, and an attribute identifier for identifying an item attribute is associated with the item. Based on the item information storage step stored in the information storage unit, the usage information and the attribute correspondence information, an attribute relevance level, which is a relevance level between the attribute identifier and the item identifier, is calculated, and the calculated attribute relevance level For each item identifier, select the related attribute that is the attribute identifier related to that item identifier A related attribute selecting step for creating related attribute information in which each item identifier is associated with the selected related attribute, and a related attribute for storing the related attribute information generated in the related attribute selecting step in the related attribute information storage unit An attribute including one of the attribute identifiers included in the related attribute information storage unit, the related attribute information including the reference item identifier that is the item identifier of the reference item is acquired from the related attribute information storage unit Corresponding information is acquired from the item information storage unit, a related candidate item set that is a set of item identifiers included in the acquired attribute corresponding information is formed, each item identifier included in the related candidate item set, the reference item, The relevance level of the reference item is calculated based on the calculated relevance level. An information processing program characterized by executing the related item selection step of selecting an item to the computer.

  According to the present invention, various items can be recommended.

It is a figure which shows the structure of the whole system in the 1st Embodiment of this invention. It is a block diagram which shows the structure of the information processing server apparatus 1 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the user information storage part 131 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the item information storage part 132 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the utilization information storage part 133 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the related attribute information storage part 134 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the attribute priority value information storage part 135 in the 1st Embodiment of this invention. It is a figure which shows the storage state of the related item information storage part 136 in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the related attribute selection process in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the attribute priority value calculation process in the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the related item selection process in the 1st Embodiment of this invention. It is a table | surface which shows the related attribute with respect to the item i3 in the 1st Embodiment of this invention. It is a figure which shows the image of the item set of the related candidate item with respect to item i3 in the 1st Embodiment of this invention.

Hereinafter, an information processing apparatus, an information processing method, and a program according to the present invention will be described with reference to the accompanying drawings. The item in the present invention may be digital content such as text, voice, music, video, web page, and various articles, financial products, real estate,
It may be information about a person. That is, the item in the present invention may be tangible or intangible, whether it is paid or free.

<First Embodiment>
Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a configuration diagram of the entire system according to the first embodiment of the present invention. In the system according to the present embodiment, the information processing server device 1 and one or more terminal devices 3 (3a to 3n) are connected via a network 2. In each embodiment, only the information processing server device 1 may function as an information processing device, or the information processing server device 1 may function as an information processing device in cooperation with the terminal device 3.

The network 2 is a network such as the Internet, for example, and mediates exchange of information between the information processing server device 1 and the terminal device 3.
The terminal device 3 is a general computer having a CPU, RAM, ROM, hard disk drive, network interface, or the like, or a portable information terminal such as a mobile phone, a smartphone, or a tablet PC. The terminal control unit 31 and the terminal communication unit 32 And terminal input unit 33
And a terminal display unit 34.
The terminal communication unit 32 is a unit for communicating with the information processing server device 1 via the network 2.
The terminal input unit 33 is configured so that, for example, the terminal device 3 is a PC (Personal Computer).
) Is an interface for the user to operate the terminal device 3 such as a mouse, a keyboard, or a mobile phone, such as a button.
The terminal display unit 34 is an interface for displaying various information and visually showing it to the user, such as a display.

The terminal control unit 31 performs usage information transmission processing, related item acquisition request transmission processing, and related item detailed information display processing via the network 2.
The usage information transmission process is a process of transmitting usage information to the information processing server device 1 when a user (user) who uses the terminal device 3 performs an operation such as browsing or using some item. . The usage information is information related to browsing and usage of the user's items, and at least a user identifier for uniquely identifying the usage user, or a terminal identifier for uniquely identifying the terminal device 3, and a target for browsing and usage. It includes an item identifier that uniquely identifies the item that has become, and usage date information that indicates when it was viewed or used. In the present embodiment, a case where a user identifier is used will be described as an example, but the same applies to a case where a terminal identifier is used. The term user entity identifier is used as a general term for the meaning of the user identifier and terminal identifier combined. Also,
The user or the terminal device used by the user is appropriately referred to as a user. Further, the use time information may be information indicating only the date, or may be information indicated in units of milliseconds, and may be information that can be understood up to a necessary unit according to the form of service.

With related item acquisition request transmission processing, when operation which acquires the related item detailed information relevant to the item which the user browsed is performed, or operation which terminal device 3 acquires related item detailed information automatically is performed. In this case, the related item acquisition request is transmitted to the information processing server device 1. The related item detailed information is information related to one or more items related to the viewed item. The related item acquisition request includes at least the item identifier of the viewed item. The related item acquisition request may be associated with an acquisition limit number that limits the number of related items to be acquired.
The related item detailed information display process is a process of receiving the related item detailed information corresponding to the transmitted related item acquisition request from the information processing server device 1 and displaying the received item detailed information on the terminal display unit 34. .

The information processing server device 1 is a device that exchanges various information with the terminal device 3 via the network 2. The information processing server device 1 can be implemented as software (program) processing using a computer including a CPU, RAM, ROM, hard disk drive (HDD), network interface, and the like. In addition, you may comprise the information processing server apparatus 1 using a some computer. For example, to load balance
The distributed processing may be performed using a plurality of computers that perform processing corresponding to each unit of the information processing server device 1. Alternatively, the distributed processing may be performed in such a manner that the processing of a part of the information processing server device 1 is performed by a computer and the processing of another part is performed by another computer.
FIG. 2 is a configuration diagram of the information processing server apparatus 1 in the present embodiment. The information processing server device 1 in this embodiment includes an information processing server control unit 11, an information processing server communication unit 12, and an information processing server storage unit 13.
The information processing server communication unit 12 is a unit for communicating with the terminal device 3 via the network 2.

The information processing server storage unit 13 stores various data using a storage device such as a memory or an HDD. The information processing server storage unit 13 includes a user information storage unit 131, an item information storage unit 132, a usage information storage unit 133, a related attribute information storage unit 134, an attribute priority value information storage unit 135, and a related item information storage. Part 136.
The user information storage unit 131 stores a plurality of user information. FIG. 3 shows a user information storage unit 1
FIG. The user information associates a user identifier (user_id) that uniquely identifies a user of the terminal device 3 and user attribute information (user_info), and is stored in a table format as shown in FIG. User attribute information includes the user's name, age, gender, address (region), hobby, and membership date (year, month, date, date, etc.)
Information such as e-mail address and telephone number. If information processing server device 1 can purchase a product, it may include credit card information for paying the product.
The item information storage unit 132 includes item identification information, attribute type information, attribute value information,
A plurality of pieces of attribute correspondence information are stored. FIG. 4 (FIGS. 4A to 4D) is a diagram illustrating a storage state of the item information storage unit 132.

Item identification information is an item identifier (item_) for uniquely identifying an item.
id) and the item name (item_name) are associated with each other in FIG.
). Note that item_id is unique and cannot be registered repeatedly.
The attribute type information is an attribute type identifier (ty) for uniquely identifying the item attribute type.
pe_id) and an item attribute type name (type_name) are associated with each other and stored in a table format as shown in FIG. The item attribute type corresponds to the item attribute item. For example, the item “creator”, “genre”, and “production year”
"Price", "User conditions suitable for use", etc. Note that type_id is unique and cannot be registered repeatedly.

The attribute value information includes an attribute type identifier (type_id), an attribute value identifier (attr_id) for uniquely identifying an attribute value corresponding to the attribute type identifier, and an attribute value (attr_n).
ame) and is stored in a table format as shown in FIG. For example, when the item is related to “book” and type_id is an identifier corresponding to “genre”, the attribute values are “fiction”, “love”, “cooking”, and the like. (T
The combination of (type_id, attr_id) is unique and duplicate registration is not possible.
The attribute correspondence information is information in which attributes corresponding to items are associated with each other, and includes an item identifier (item_id), an attribute type identifier (type_id), and an attribute value identifier (attr).
_Id) and is stored in a table format as shown in FIG. Note that the combination of (item_id, type_id, attr_id) is unique, and duplicate registration is not possible.

By using the tables shown in FIGS. 4A to 4D, a plurality of attributes of an item can be obtained.
Each attribute type can be managed. The item identification information corresponding to an item, the attribute type information specified from the attribute correspondence information corresponding to the item, and the attribute value information are combined as item attribute information. Also, type_name and attr_name
The combination of e is an item attribute. A combination of type_id and attr_id is used as an attribute identifier. In this embodiment, the attribute identifier is treated as a combination of the attribute type identifier and the attribute value identifier. However, in the case of a service in which the item attribute type does not exist (the attribute type identifier does not exist), the attribute value Only identifiers may be treated as attribute identifiers. In this case, it is not necessary to store attribute type information.
The usage information storage unit 133 stores a plurality of usage information. FIG. 5 is a diagram illustrating a storage state of the usage information storage unit 133. As shown in FIG. 5, the user identifier (use) that constitutes the usage information
r_id), an item identifier (item_id), and usage time information (dl_date) are associated with each other and stored in a table format. When other information is included in the usage information, the storage format may be changed according to the form of the usage information so that all the information included in the usage information can be stored.

The related attribute information storage unit 134 stores a plurality of related attribute information. FIG. 6 is a diagram illustrating a storage state of the related attribute information storage unit 134. As shown in FIG. 6, the item identifier (item_id), the attribute type identifier (type_id), the attribute value identifier (attr_id), and the attribute association degree (a_value) obtained by quantifying the degree of association are included. Are stored in a table format. (Item_id, type_id, att
The combination of r_id) is unique, and duplicate registration is not possible.
The attribute priority value information storage unit 135 stores a plurality of attribute priority value information. FIG. 7 is a diagram illustrating a storage state of the attribute priority value information storage unit 135. As shown in FIG. 7, an item identifier (item_id), an attribute type identifier (type_id), an attribute value identifier (attr_id), and an attribute priority value obtained by quantifying the attribute priority ( p_va
lue) and store them in a table format. In addition, (item_id, type_
The combination of id, attr_id) is unique and duplicate registration is not possible.

The related item information storage unit 136 stores a plurality of related item information. FIG. 8 is a diagram illustrating a storage state of the related item information storage unit 136. As shown in FIG. 8, the item identifier (reference item identifier (base_item_id)) of the item (reference item) serving as a reference constituting the related item information, and the item identifier of the related item with respect to the reference item (related item identifier (sim_item_id)) Are stored in a table format in association with the degree of association (s_value) obtained by quantifying the degree of association between the reference item and the associated item. In addition, the combination of (base_item_id, sim_item_id) is unique, and duplicate registration is not possible. By specifying the reference item identifier, all related item identifiers corresponding to the reference item can be acquired. Here, in order to store two types of item identifiers in one table, base_item_id, sim_i
Although the names tem_id are used for distinction, these are the same as item_id stored in the item information storage unit 132 or the like. The reference item is, for example, an item to be browsed, but may be any other item specified by various methods.

The information processing server control unit 11 performs overall control on each unit configuring the information processing server device 1. The information processing server control unit 11 includes a usage information registration unit 111, a related attribute selection unit 112, an attribute priority value calculation unit 113, a related item selection unit 114, and a related item detailed information creation unit 115.
The usage information registration unit 111 receives usage information from the terminal device 3 via the network 2 and stores the received usage information in the usage information storage unit 133.
The related attribute selection unit 112 performs related attribute selection processing at each predetermined timing. As the predetermined timing, a predetermined time interval (for example, every 24 hours) may be used, or the usage information may be received a certain number of times. Also, the time interval may vary from Monday to Friday every 3 hours, Saturday every 6 hours, and Sunday every 12 hours. Also, the time interval may be changed according to the season, such as shortening the time interval in summer and increasing the time interval in winter.

The related attribute selection process will be described with reference to the flowchart of FIG.
First, the related attribute selection unit 112 extracts all item identifiers included in the usage information from the usage information storage unit 133 without duplication (step 101). At this time, the usage information for extracting the item identifier includes usage time information corresponding to a range that goes back for a predetermined period (eg, predetermined by the service provider such as 3 months) from the time of the related attribute selection process. By limiting, related attribute information may be selected only for the item used most recently.
Next, the related attribute selection unit 112 selects, for example, one of the item identifiers extracted in step S101 in the order of extraction (step S102).
Next, the related attribute selection unit 112 acquires all the usage information having the item identifier selected in Step S102 from the usage information storage unit 133 (Step S103). At this time, the usage information to be acquired is limited to information including usage time information corresponding to a range that goes back for a predetermined period (for example, set in advance by the service provider such as 3 months) from the time of the related attribute selection process. Thus, the related attribute information may be selected using only the latest usage information. In addition, by acquiring up to a predetermined number (for example, 10 is set in advance by the service provider) in the order close to the time of related attribute selection processing, the amount of calculation when selecting related attribute information is reduced. It may be.

Next, the related attribute selection unit 112 extracts all user identifiers without duplication from the usage information acquired in step S103 (step S104).
Next, the related attribute selection unit 112 acquires all the usage information including any of the user identifiers extracted in step S104 from the usage information storage unit 134 (step S105). At this time, the usage information to be acquired is limited to information including usage time information corresponding to a range that goes back for a predetermined period (for example, set in advance by the service provider such as 3 months) from the time of the related attribute selection process. Thus, the related attribute information may be selected using only the latest usage information. In addition, for each user identifier, by obtaining the usage information including the user identifier in the order close to the time of the related attribute selection process, up to a predetermined number (for example, 10 is set in advance by the service provider), You may make it reduce the computational complexity at the time of selecting related attribute information.
Next, the related attribute selection unit 112 acquires, from the item information storage unit 132, a target attribute selection target among the attribute type identifiers included in the attribute type information (step S106).
Note that a service in which the item attribute type is managed by an attribute type identifier or the like needs to set two or more attribute type identifiers as related attribute selection targets in order to realize the effect of the present invention. A service provider may set an attribute type identifier for selecting a related attribute in advance.

Next, the related attribute selection unit 112 selects, for example, one of the attribute type identifiers acquired in step S106 in the order of acquisition (step S107).
Next, among the attribute correspondence information having the attribute type identifier selected in step S107 from the item information storage unit 132 by the related attribute selection unit 112, any of the item identifiers included in the usage information acquired in step S105. Get everything including. Further, the attribute identifier included in the acquired attribute correspondence information is extracted without duplication as a related candidate attribute identifier (step S108).
Next, the related attribute selection unit 112 calculates attribute relevance for each related candidate attribute identifier extracted in step S108 (step S109).
As a method for calculating the attribute relevance, for example, a Jaccard coefficient can be used. When the Jaccard coefficient is used, Ui is a user set that has used item i, Ua is a user set that has used any item that belongs to item attribute a, and any one that belongs to item i and item attribute a The number of users who have used both items is | Ui ∩ Ua |, and the number of users who have used at least one of item i and any item belonging to item attribute a is | Ui ∪ Ua | At this time, the attribute relevance sim (i, a) can be calculated by the equation (1).

Further, by calculating the item preference level and the attribute preference level, the cosine distance and the Pearson product moment correlation coefficient can also be used for attribute relevance calculation.
The item preference level is a value indicating the degree of preference of a user for an item, and can be calculated using usage information (first usage information) including an item identifier to be calculated.
Specifically, for example, the attribute relevance calculation processing is performed from the number of first usage information, the number of users counted without duplication from the first usage information, and the usage time information included in the first usage information. It can be calculated using the elapsed time up to the point in time.

The attribute preference degree is a value indicating the degree of preference for a user attribute, and usage information (second usage information) including any item identifier corresponding to a related candidate attribute identifier to be calculated.
Can be used to calculate. Specifically, for example, attribute relevance calculation processing is performed from the number of second usage information, the number of users counted without duplication from the second usage information, and the usage time information included in the second usage information. It can be calculated using the elapsed time up to the point in time.
When the cosine distance is used, for example, a user set that has used the item i is Ui, and the item preference for the user i ui (uiεUi) of the item i is Vi (i,
ui), a user set that has used any item belonging to the item attribute a is Ua, and the attribute preference degree of the item attribute a with respect to the user ua (ua∈Ua) is Va (a,
ua), a user who has used both the item i and any item belonging to the item attribute a is Uc, and the item preference level of the item i with respect to the user uc (ucεUc) is Vi (i, uc), the item The attribute preference degree for the user uc of the attribute a is expressed as Va (a, uc
), The attribute relevance sim (i, a) can be calculated by equation (2).

When using the Pearson product moment correlation coefficient, for example, a user who has used both items i and any item belonging to the item attribute a is Uc, the number of users belonging to Uc is n, The item preference degree for the user uc (ucεUc) of i is expressed as V
The attribute relevance degree sim (i, a) can be calculated by Expression (3) where Va (a, uc) is the attribute preference degree of i (i, uc) and item attribute a with respect to the user uc.

Other than this, any index may be used as long as it is an index representing the relationship between items and attributes.

Next, according to the attribute relevance calculated by the related attribute selection unit 112 in step S109,
Related attributes are selected (step S110). To select related attributes, attribute relevance (
A related candidate attribute identifier that is set in advance by the service providing side and has a predetermined value or more may be used as the related attribute. Further, the related candidate attribute identifiers selected up to a predetermined number (set in advance by the service provider) in descending order of attribute relevance may be used as related attributes.
Next, for each related attribute selected in step S110, the related attribute selection unit 112 selects the item identifier selected in step S102, the combination of the attribute type identifier and attribute value identifier that are related attributes, and in step S109. The related attribute information associated with the attribute relevance calculated corresponding to the related attribute is stored in the related attribute information storage unit 134 (step S111).

Next, the related attribute selection unit 112 determines whether or not all attribute type identifiers have been selected in step S107 (step S112). If all are selected, the process proceeds to step S113, and if there is an unselected one, the process proceeds to step S107.
In step S113, the related attribute selection unit 112 determines whether all item identifiers have been selected in step S102. If all are selected, the series of processing from step S101 to step S113 is terminated, and if there are unselected ones, step S102 is performed.
Proceed to
With this procedure, in a service that manages item attribute types, for each item identifier, related attribute information corresponding to two or more attribute type identifiers is stored in the related attribute information storage unit 134. It will be.
In addition, the candidate of the item at the time of selecting a related item can be increased, so that the attribute type identifier used as the object which produces | generates related attribute information increases.
In addition, in a service that does not manage the item attribute type (no attribute type identifier), the processes of step S106, step S107, and step S112 can be omitted.
The above is the description of the procedure of the related attribute selection process.

The attribute priority value calculation unit 113 performs an attribute priority value calculation process at every predetermined timing. The attribute priority value calculation process will be described with reference to the flowchart of FIG.
First, the attribute priority value calculation unit 113 extracts all item identifiers included in the usage information from the usage information storage unit 133 without duplication (step 201).
Next, the attribute priority value calculation unit 113 selects, for example, one of the item identifiers extracted in step S201 in the order of extraction (step S202).
Next, the attribute priority value calculation unit 113 acquires all the usage information having the item identifier selected in step S202 from the usage information storage unit 133 (step S203). At this time, the usage information to be acquired is limited to information including usage time information corresponding to a range that goes back a predetermined period (for example, preset by the service provider such as 3 months) from the time of attribute priority value calculation processing. Thus, the attribute priority value may be calculated using only the latest usage information.
Next, the attribute priority value calculation unit 113 receives, from the item information storage unit 132, step S202.
All the attribute correspondence information having the item identifier selected in (1) is acquired (step S204).
. Note that when acquiring the attribute correspondence information, the attribute correspondence information may be limited to only one including one or more attribute type identifiers set in advance by the service provider.

Next, the attribute priority value calculation unit 113 selects, for example, one of the attribute correspondence information acquired in step S204 in the order of acquisition (step S205).
Next, the attribute priority value calculation unit 113 calculates an attribute priority value for the attribute correspondence information selected in step S205 (step S206).

Seven types of attribute priority value calculation methods will be described below.
The first method for calculating the attribute priority value is a method for uniformly calculating the same value (for example, “1”) for any item attribute corresponding to any attribute correspondence information. The first method of attribute priority value calculation treats any item attribute equally.

The second method of calculating the attribute priority value is a method of calculating the value so that the smaller the number of items belonging to the item attribute corresponding to the attribute correspondence information selected in step S205 is, the larger the value is. Specifically, the number of item sets I (tx, txy) belonging to the attribute identifier (tx, txy) of the item attribute (a pair of the attribute type identifier tx and the attribute value identifier) | I (tx,
txy) |, the constant α1 (> 0) is used and the attribute priority value p corresponding to the item i is used.
v (i, tx, txy) may be calculated by equation (4).

The second method for calculating the attribute priority value is to increase the weighting of finer item attributes such that the number of items belonging to the item attribute is small, and conversely the granularity such that the number of items belonging to the item attribute is large. The coarser the item attribute, the smaller the weight. Of course, if it is an expression that monotonously decreases with respect to the number of items belonging to the item attribute,
Any formula may be used.

The third method for calculating the attribute priority value is a method of calculating the value so that the larger the number of attribute value identifiers corresponding to the attribute type identifier corresponding to the attribute correspondence information selected in step S205 is, the larger the value is. . Specifically, when the number of attribute value identifiers A (tx) corresponding to the attribute type identifier tx of the item attribute is | A (tx) |, the constant β1 (≧ 1) is used to correspond to the item i. The attribute priority value pv (i, tx, txy) to be calculated may be calculated by equation (5).

The third method for calculating the attribute priority value is to increase the weight of an attribute type for which a fine-grained attribute value is set such that the number of attribute values corresponding to the attribute type is large. The weighting is reduced for attribute types for which coarse-grained attribute values are set such that the number of corresponding attribute values is small. Of course, any expression may be used as long as the expression monotonically increases with respect to the number of attribute values corresponding to the item type.

The fourth method for calculating the attribute priority value is a method for calculating the attribute priority value so that the larger the use of the item belonging to the item attribute corresponding to the attribute correspondence information selected in step S205 is, the larger the value is. Specifically, a set L (t of usage information related to the use of any item in the item set I (tx, txy) belonging to the attribute identifier (tx, txy) of the item attribute
When the number of x, txy) is | L (tx, txy) |, | A (tx) |
It can be calculated by replacing with | L (tx, txy) |. The fourth method of calculating attribute priority values can increase the weight of item attributes to which many popular items that are frequently used by many users belong. Of course, any expression may be used as long as the expression monotonically increases with respect to the number of items belonging to the item attribute.

The fifth method for calculating the attribute priority value is a method of calculating the value so as to increase the number of users who use items belonging to the item attribute corresponding to the attribute correspondence information selected in step S205. Specifically, there is no duplication from the set L (tx, txy) of usage information related to the use of any item included in the item set I (tx, txy) belonging to the attribute identifier (tx, txy) of the item attribute. Extracted user set U (tx, txy)
When | U (tx, txy) | is the number of | A (tx) | in Equation (5), | U (tx,
txy) | The fifth method of attribute priority value calculation can increase the weighting of popular item attributes that are used by many users. Of course, any expression may be used as long as the expression monotonically increases with respect to the number of users using items belonging to the item attribute.

The sixth method for calculating the attribute priority value is a method for calculating the attribute priority value so that the larger the number of items used corresponding to the item identifier selected in step S202, the larger the value. Specifically, when the number of sets L (i) of usage information related to the use of item i is | L (i) |, | A (tx) | in Expression (5) is | L (i) It can be calculated by replacing with |. The sixth method of attribute priority value calculation can increase the weighting of popular items that are used many times by many users. Of course, any formula may be used as long as the formula increases monotonously with the number of items used.

The seventh method for calculating the attribute priority value is a method of calculating the attribute priority value so that the value increases as the number of users using the item corresponding to the item identifier selected in step S202 increases. Specifically, if | U (i) | is the number of user sets U (i) extracted without duplication from the set L (i) of usage information related to the use of item i, the | It can be calculated by replacing A (tx) | with | U (i) |. The seventh method of attribute priority value calculation can increase the weight of popular items that are used by many users. Of course, any expression may be used as long as the expression monotonically increases with respect to the number of users used.
Further, the calculation may be performed by combining the second to seventh methods of attribute priority value calculation. For example,
You may calculate like Formula (6) and Formula (7) combining the 2nd and 4th method. Other than that, they may be freely combined so as not to change the properties of the respective calculation methods. This completes the description of the attribute priority value calculation method.

Next, the attribute priority value calculation unit 113 associates the item identifier, the attribute type identifier, the attribute value identifier, and the attribute priority value calculated in step S206, which are included in the attribute correspondence information selected in step S205. The information is stored in the attribute priority value information storage unit 135 (step S207).
Next, the attribute priority value calculation unit 113 determines whether or not all attribute correspondence information has been selected in step S205 (step S208). If all are selected, the process proceeds to step S209, and if there is any unselected, the process proceeds to step S205.
In step S209, the attribute priority value calculation unit 113 determines whether all item identifiers have been selected in step S202. If all are selected, the series of processing from step S201 to step S209 is terminated, and if there are unselected ones, step S20
Go to step 2.
The above is the description of the procedure of the attribute priority value calculation process.

The related item selection unit 114 performs related item selection processing at each predetermined timing. The procedure of related item selection processing will be described with reference to the flowchart of FIG.
First, the related item selection unit 114 extracts all item identifiers included in the related attribute information from the related attribute information storage unit 134 without duplication as reference item identifiers (step S301).
Next, the related item selection unit 114 selects, for example, one of the reference item identifiers extracted in step S301 in the order of extraction (step S302).
Next, the related item selection unit 114 receives a step S30 from the related attribute information storage unit 134.
All related attribute information having an item identifier that matches the reference item identifier selected in 2 is acquired (step S303).

Next, the related item selection unit 114 receives a step S3 from the attribute priority value information storage unit 135.
All attribute priority value information having an item identifier that matches the reference item identifier selected in 02 is acquired (step S304).
Next, the related item selection unit 114 receives a step S30 from the item information storage unit 132.
The attribute correspondence information including the attribute identifier included in the related attribute information acquired in 3 is specified, and all the item identifiers included in the specified attribute correspondence information are extracted without duplication as related candidate item identifiers (step S305). The set of related candidate item identifiers extracted here becomes the related candidate item set.
Next, the related item selection unit 114 selects, for example, one of the related candidate item identifiers extracted in step S305 in the order of extraction (step S306).

Next, the related item selection unit 114 receives a step S3 from the attribute priority value information storage unit 135.
All attribute priority value information having an item identifier that matches the related candidate item identifier selected in 06 is acquired (step S307).
Next, the related item selection unit 114 calculates the degree of association between the reference item identifier selected in step S302 and the related candidate item identifier selected in step S306 (step S308).
Five types of relevance calculation methods will be described below. In the description of the calculation method, a set of related attribute information corresponding to the reference item ib acquired in step S303 is ST (ib).
And

The first method of calculating the degree of association is a method of calculating the number of pieces of related attribute information including the attribute identifier corresponding to the related candidate item identifier selected in step S306 as the degree of association. The first method of calculating relevance has a small amount of calculation.

The second method of calculating the relevance is a method of calculating using the attribute relevance of the related attribute information including any of the attribute identifiers corresponding to the related candidate item identifier selected in step S306. Specifically, a set TA (ib, ir) of attribute identifiers extracted from the attribute identifiers corresponding to the related candidate item ir from ST (ib) including one of the attribute identifiers.
Is defined as | TA (ib, ir) |, and any attribute identifier included in TA (ib, ir) (
Attribute relevance sim (ib, tx, txy) with reference item ib corresponding to tx, txy)
), The relevance rate (ib, ir) is expressed by the equation (2) using α2 (0 ≦ α2 ≦ 1).
8).

The second method of calculating relevance is that any item attribute corresponding to a related candidate item is
The degree of association tends to increase as the degree of attribute relevance included in many pieces of related attribute information and included in the related attribute information increases. Equation (8) is the sum of attribute relevance when α2 is “0”, and the average value of attribute relevance when α2 is “1”. That is, in order to increase the influence of the number of related attribute information, α2 may be brought close to “0”. Of course, other expressions may be used as long as the number of related attribute information and the degree of attribute relevance increase monotonously.

The third method of calculating the relevance level is a calculation method that uses the attribute relevance level in the same manner as the second method of relevance level calculation and changes the weight of the corresponding attribute relevance level for each attribute type identifier. .
Specifically, for each attribute type identifier included in the set T of attribute type identifiers to be calculated for attribute relevance calculation, a weighting coefficient m1 (tx) corresponding to the attribute type identifier tx (∈T).
) Is set in advance, a set ST (ib, ir) of related attribute information including any attribute identifier among the attribute identifiers corresponding to the related candidate item ir is formed from ST (ib).
Next, a subset T (ib, ir) of T extracted without duplication from ST (ib, ir) is formed. Of the related attribute information included in ST (ib, ir), the attribute type identifier t
A set TA (ib, ir) of attribute identifiers extracted from those containing 'x (εT (ib, ir))
, T′x) is | TA (ib, ir, t′x) |, and the attribute corresponding to an arbitrary attribute identifier (t′x, t′xy) of TA (ib, ir, t′x) Relevance is av (ib, t'x, t
In the case of “xy”, the relevance rate (ib, ir) can be calculated by Expression (9) using α3 (0 ≦ α3 ≦ 1).

Note that the weighting coefficient m1 (tx) corresponding to the attribute type identifier tx (∈T) is calculated according to the granularity of the attribute type in the same manner as the third method for calculating the attribute priority value, so that It is desirable to absorb the granularity, but of course, the service providing side may freely set it regardless of the granularity between attribute types. In the third method of calculating the relevance level, in addition to the nature of the second method of calculating the relevance level, the attribute relevance level corresponding to which attribute type is prioritized by changing the weight for each attribute type. The provider can set it freely.

The fourth method of calculating the relevance level is a calculation method that uses the attribute relevance level in the same manner as the second method of relevance level calculation and changes the weight of the corresponding attribute relevance level for each attribute identifier. Specifically, for each set of attribute identifiers TA including the attribute type identifier that is the target of attribute relevance calculation, the weighting coefficient m2 (tx, txy) corresponding to the attribute identifier (tx, txy)
And a subset TA of TA extracted from ST (ib) without duplication from the related attribute information including any attribute identifier among the attribute identifiers corresponding to the related candidate item ir.
The attribute relevance level corresponding to an arbitrary attribute identifier (t′x, t′xy) of (ib, ir) is av (
In the case of ib, t′x, t′xy), the relevance rate (ib, ir) is α4 (0 ≦ α
4 ≦ 1), and can be calculated by the equation (10).

The weighting coefficient m2 (tx, txy) corresponding to the attribute identifier (tx, txy) is
Similar to the second method of attribute priority value calculation, by calculating according to the granularity of the item attribute,
It is desirable to absorb the granularity between item attributes, but, of course, the service provider may freely set it regardless of the granularity between item attributes. In the fourth method of calculating the relevance level, in addition to the nature of the second method of calculating the relevance level, by changing the weight for each item attribute, the attribute relevance level corresponding to which item attribute is given priority. The provider can set it freely.

The fifth method of calculating relevance is a method of calculating using attribute priority information for the reference item acquired in step S304 and attribute priority value information for the related candidate item acquired in step S307. Specifically, a set TA ′ (ib) of attribute identifiers extracted without duplication from the set TP (ib) of attribute priority value information acquired for the reference item ib,
A common part TA ′ (ib) ∩TA ′ (ir) with a set of attribute identifiers TA ′ (ir) extracted without duplication from the set of attribute priority value information TP (ir) acquired for the related candidate item ir
) Are attribute priority values corresponding to attribute identifiers (t′x, t′xy) included in pv (
When ib, t′x, t′xy) (reference item) and pv (ir, t′x, t′xy) (related candidate item) are set, the relevance rate rate (ib, ir) is expressed by equation (11). As such, it may be calculated as an inner product value.

Further, it may be calculated as a cosine distance or a Pearson product moment correlation coefficient. Relevance calculation fifth
In this method, by calculating the degree of association between the reference item and the related candidate item using the attribute priority value, it becomes easier to preferentially select items that are close to the reference item as item items. In addition, the nature of the method used for attribute priority value calculation is also reflected.

Alternatively, the attribute priority value calculation may be calculated by combining the second to fifth methods. For example,
You may calculate like Formula (12) combining the 3rd and 4th method.

In addition, a new association degree new_rate (ib, ir) is calculated using the association rate rate2 (ib, ir) calculated by the second method and the association rate rate5 (ib, ir) calculated by the fifth method.
ir) may be calculated as the degree of relevance as in Expression (13) or Expression (14). Other than that, they may be freely combined so as not to change the properties of the respective calculation methods.

Next, the related item selection unit 114 determines whether or not all related candidate item identifiers have been selected in step S306 (step S309). If all are selected, the process proceeds to step S310, and if there is any unselected, the process proceeds to step S306.
In step S310, the related item selection unit 114 selects related items according to the degree of relevance calculated in step S308. In order to select a related item, a related candidate item identifier having a relevance level equal to or higher than a predetermined value (set in advance by the service provider) may be used as the related item. Moreover, what is necessary is just to let the related candidate item identifier selected to the predetermined number (previously set by the service provision side) the descending order of the related degree as the related item.
Next, for each related item selected in step S310, the related item selection unit 114 sets the reference item identifier selected in step S302 and the related candidate item identifier that is a related item as the related item identifier, The related item information associated with the relevance calculated in S308 is stored in the related item information storage unit 136 (step S311).

Next, it is determined whether the related item selection part 114 has selected all the reference | standard item identifiers in step S302 (step S312). If all are selected, step S
A series of processing from 301 to step S312 ends, and if there is an unselected item, the process proceeds to step S302.
A specific example of the actual effect of this procedure will be described with reference to FIGS. FIG. 12 shows related attributes for the item i3. As shown in FIG. 12, the related attributes of the item i3 are (ta, ta7), (ta, ta13), (tb, t
b4) and (tb, tb8). FIG. 13 is an image of an item set that is a target of a related candidate item for the item i3. As shown in FIG.
An item set corresponding to each related attribute of the item i3, I (ta, ta7), I (t
a, ta13), I (tb, tb4), I (tb, tb8), and the shaded portion is the item set I (ta, ta7) ∪I (ta, ta13) ∪I (tb, tb4) of the related candidate items
) ∪I (tb, tb8). In other words, related items are represented by I (ta, ta7), I (t
a, ta13), I (tb, tb4), and I (tb, tb8) can be selected from the items belonging to any of the items, so that it is possible to recommend items even if the usage information is small, and By using various item attributes, recommendation of various items can be realized.

In the above procedure, if the attribute priority value is not used for calculating the relevance, step S
The processing of 304 and step S307 can be omitted. Furthermore, since it is not necessary to create attribute priority value information, the attribute priority value calculation unit 113 and the attribute priority value information storage unit 135 are unnecessary.
The above is the description of the related item selection process.

When the related item detailed information creation unit 115 receives a related item acquisition request from the terminal device 3 through the information processing server communication unit 12, the related item detailed information creation unit 115 creates related item detailed information by performing related item detailed information creation processing, Through the server communication unit 12, the terminal device 3 is
Send the created related item detailed information.
The related item detailed information creation processing is processing for creating related item detailed information. Specifically, first, the related item information storage unit 136 compares the item identifier included in the received related item acquisition request with the reference item identifier, and acquires all the related item information that matches. Then, item information corresponding to the related item identifier included in the acquired related item information is acquired from the item information storage unit 132, and related item detailed information is created using the acquired item information. If the related item acquisition request includes the acquisition limit number, when acquiring the related item information from the related item information storage unit 136, the acquisition limit number is acquired in descending order of the degree of association. Good.

As described above, by creating the related item information based on the attribute information, it is possible to recommend various items even with a small amount of usage information. For this reason, the user's interest in recommendation can be maintained and the use of the service can be promoted.

The above-described embodiments of the present invention are illustrative examples, and are not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention. For example, you may combine each embodiment, a modification, etc. In addition, the information processing apparatus (information processing server apparatus 1
The functions of the information processing apparatus may be realized by making a part of the configuration of () separate and communicating with the separate configuration via a network or the like.

The present invention also includes a program for causing a computer to realize the functions of the respective units. These programs may be read from a recording medium and loaded into a computer, or may be transmitted via a communication network and loaded into a computer.

DESCRIPTION OF SYMBOLS 1 Information processing server apparatus 2 Network 3 Terminal apparatus 31 Terminal control part 32 Terminal communication part 33 Terminal input part 34 Terminal display part 11 Information processing server control part 111 Usage information registration part 112 Related attribute selection part 113 Attribute priority value calculation part 114 Related Item selection unit 115 Related item detailed information creation unit 12 Information processing server communication unit 13 Information processing server storage unit 131 User information storage unit 132 Item information storage unit 133 Usage information storage unit 134 Related attribute information storage unit 135 Attribute priority value information storage unit 136 related item information storage

Claims (9)

An acquisition unit for acquiring usage information in which a user is associated with an item used by the user, and attribute correspondence information in which an item is associated with an item attribute;
Based on the obtained usage information and attribute corresponding information items that are utilized by utilizing mainly using one item, it identifies the corresponding item attributes in the attribute mapping information, and the one item Calculating an attribute relevance level indicating the strength of the relevance with the identified item attribute, and preferentially selecting an item attribute with a high attribute relevance level for each item as a related attribute of the item; ,
Using the selected related attribute, a relevance calculation unit that calculates relevance indicating the strength of relevance between the reference item and other items;
A related item selection unit that selects an item related to the reference item based on the calculated relevance;
Equipped with a,
The relevance calculation unit creates a related candidate item set whose elements are items corresponding to any of the related attributes of the reference item in the attribute correspondence information, and the related candidates that are elements of the reference item and the related candidate item set Calculate the relevance of the item,
An information processing apparatus characterized by that. The relevance calculation unit calculates the relevance between the reference item and the related candidate item using the number of item attributes corresponding to the related candidate item in the attribute correspondence information included in the set of related attributes of the reference item. calculate,
The information processing apparatus according to claim 1. The relevance calculation unit includes a reference item and a related candidate item such that the higher the attribute relevance between the item attribute corresponding to the related candidate item in the attribute correspondence information and the reference item, the higher the value. to calculate the related degree,
The information processing apparatus according to claim 1 . The relevance calculation unit specifies one or more item attributes corresponding to related candidate items in the attribute correspondence information, adds an attribute relevance between a reference item and the specified item attribute, and a total value thereof the value obtained by dividing a value based on the number of items attributes the identifying, a reference item and related degree between related candidate item,
The information processing apparatus according to claim 3 . The relevance calculation unit calculates an attribute priority value for each item attribute that is higher as the number of items belonging to one item attribute in the attribute correspondence information is smaller, and corresponds to the related candidate item in the attribute correspondence information. One or more item attributes are specified, and using the attribute relevance between the reference item and the specified item attribute and the attribute priority value of the item attribute corresponding to the related candidate item, the reference item and the related candidate item Calculate the relevance of
The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. The relevance calculation unit determines, for each item attribute, an attribute priority value that becomes higher as the number of times an item belonging to one item attribute is used in the attribute correspondence information or the number of using entities using the item increases. Calculate and specify one or more item attributes corresponding to the related candidate item in the attribute correspondence information, the attribute relevance between the reference item and the specified item attribute, and the item attribute corresponding to the related candidate item Using the attribute priority value of, calculate the relevance between the reference item and the related candidate item,
The information processing apparatus according to claim 1, wherein the information processing apparatus is an information processing apparatus. The relevance level calculation unit has a higher value as the attribute relevance level between the reference item and the identified item attribute is higher, and a higher value as the attribute priority value of the item attribute corresponding to the related candidate item is higher. To calculate the relevance between the reference item and the related candidate item,
The information processing apparatus according to claim 5, wherein the information processing apparatus is an information processing apparatus. An information processing method executed by one or more computers,
An acquisition step of acquiring usage information in which a user is associated with an item used by the user, and attribute correspondence information in which an item is associated with an item attribute;
Based on the obtained usage information and attribute corresponding information items that are utilized by utilizing mainly using one item, it identifies the corresponding item attributes in the attribute mapping information, and the one item, A related attribute selection step of calculating an attribute relevance indicating the strength of the relevance with the identified item attribute, preferentially selecting an item attribute having a high attribute relevance for each item as a related attribute of the item,
A relevance calculating step of calculating a relevance indicating the strength of relevance between the reference item and another item using the selected related attribute ;
Based on the calculated degree of association observed including a related item selection step of selecting an item associated with the reference item,
The relevance calculation step creates a related candidate item set whose elements are items corresponding to any of the related attributes of the reference item in the attribute correspondence information, and the related candidate that is an element of the reference item and the related candidate item set Calculate the relevance of the item ,
An information processing method characterized by the above. One or more computers,
An acquisition unit for acquiring usage information in which a user is associated with an item used by the user, and attribute correspondence information in which an item is associated with an item attribute;
Based on the obtained usage information and attribute corresponding information items that are utilized by utilizing mainly using one item, it identifies the corresponding item attributes in the attribute mapping information, and the one item , A related attribute selection unit that calculates an attribute relevance indicating the strength of the relevance with the identified item attribute, and preferentially selects an item attribute with a high attribute relevance for each item as a related attribute of the item,
Using the selected related attribute, a relevance calculation unit that calculates relevance indicating the strength of relevance between the reference item and other items,
Based on the calculated relevance, a related item selection unit that selects an item related to the reference item,
To function as,
The relevance calculation unit creates a related candidate item set whose elements are items corresponding to any of the related attributes of the reference item in the attribute correspondence information, and the related candidates that are elements of the reference item and the related candidate item set Calculate the relevance of the item,
An information processing program characterized by that.

Images