JP2023061295A - Information processing method, device, and program - Google Patents

Abstract

To suitably recommend an object to a user even when there is a little user information.SOLUTION: An information processing method according to one embodiment includes: acquisition processing which acquires user information on a plurality of users, action history information on the plurality of users, and object information on a plurality of objects including a categorical data number; similarity calculation processing which calculates a degree of similarity between values of categorical data on the basis of the action history information on the plurality of users and the object information on the plurality of objects; processing which adds the categorical data to the user information on one of the users on the basis of the action history information on the one of the users and the object information on the plurality of objects and revises the categorical data on the one of the users on the basis of the degree of similarity; recommendation degree calculation processing which calculates a degree of recommendation of each of the objects for the one of the users on the basis of the user information on the one of the users and the object information on the plurality of objects; and recommendation processing which recommends some of the objects to the one of the users on the basis of the degree of recommendation of each of the objects.SELECTED DRAWING: Figure 8

Description

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

Cited Document 1 describes a job offer and job application information providing device that provides job offer information that matches the job aptitude of job seekers. According to this device, it is possible to estimate the type of industry and job type suitable for the job seeker based on the content etc. published by the job seeker on a blog site, etc., and present the corresponding job offer information to the job seeker.

JP 2013-186813 A

With the conventional technology described above, it is difficult to present appropriate job offer information to a job seeker when there is little information about the job seeker.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and an object of the present invention is to appropriately recommend a target to a user even when there is little user information.

An information processing method according to an embodiment includes an acquisition process of acquiring user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including the number of categorical data; a similarity calculation process for calculating a similarity between values of the categorical data based on action history information of a plurality of users and target information of the plurality of targets; and the action history information of one user. and target information of the plurality of targets, and a processing process of adding the categorical data to the user information of the one user and correcting the categorical data of the one user based on the similarity. , a recommendation degree calculation process for calculating a recommendation degree of each target for the one user based on user information of the one user and target information of the plurality of targets; and recommendation processing for recommending a target to the one user.

According to one embodiment, it is possible to appropriately recommend a target to a user even when there is little user information.

It is a figure which shows an example of a structure of an information processing system. It is a figure which shows an example of the hardware constitutions of an information processing apparatus. It is a figure which shows an example of the functional structure of an information processing apparatus. It is a figure which shows an example of user information. It is a figure which shows an example of object information. It is a figure which shows an example of action log information. 4 is a flowchart illustrating an example of processing executed by a server; It is a figure which shows the specific example of the process which a server performs. It is a figure explaining the specific example of the calculation method of similarity.

Hereinafter, each embodiment of the present invention will be described with reference to the accompanying drawings. Regarding the descriptions of the specifications and drawings according to each embodiment, components having substantially the same functional configuration are denoted by the same reference numerals, thereby omitting redundant description.

<System Overview>
First, an outline of an information processing system according to this embodiment will be described. The information processing system according to this embodiment is a recommendation system that appropriately recommends a target to a user using categorical data. A recommendation system provides a service of recommending (matching) an object to a user. A user is a person who receives a target recommendation using the recommendation system, and is a user of the recommendation service. The target is something that is recommended to the user on the recommendation system, and can be anything that can be recommended to the user. The target is, for example, a job, a person, a travel plan, a car, miscellaneous goods, a household appliance, food, a store, an article, music, a moving image, or an image, but is not limited thereto. Categorical data (categorical variables) are data that take preset categorical values (discrete values). The categorical data is, for example, region, industry, age, sex, or employment status, but is not limited to these. In this embodiment, the categorical data is included in the target information. Note that the target information will be described later.

FIG. 1 is a diagram showing an example of the configuration of an information processing system according to this embodiment. As shown in FIG. 1, the information processing system according to this embodiment includes a user terminal 1, a registrant terminal 2, and a server 3, which are connected to each other via a network N so as to be able to communicate with each other. The network N is, for example, a wired LAN (Local Area Network), a wireless LAN, the Internet, a public line network, a mobile data communication network, or a combination thereof. In the example of FIG. 1, the information processing system includes one user terminal 1, one registrant terminal 2, and one server 3, but a plurality of each may be provided.

The user terminal 1 is an information processing device used by a user. The user terminal 1 is, for example, a PC (Personal Computer), a smart phone, or a tablet terminal, but is not limited thereto. A user registers his/her own user information 321 via the user terminal 1 . In addition, the user displays the target information 322 on the user terminal 1, searches for a target that matches a desired condition from among the registered targets, or confirms a target recommended by the server 3. , to find suitable subjects for themselves. Also, the user sends an application (requests matching) to a target determined to be suitable for the user via the user terminal 1 .

The registrant terminal 2 is an information processing device used by a registrant who registers an object to be recommended to the user. The registrant terminal 2 is, for example, a PC, a smartphone, or a tablet terminal, but is not limited thereto. The registrant registers the target information 322 via the registrant terminal 2 . In addition, the registrant receives, via the registrant terminal 2, a notice of an application from the user for the object registered by the registrant.

The server 3 is an information processing device that recommends an object to the user. The server 3 stores user information 321 and target information 322 received from the user terminal 1 and the registrant terminal 2, respectively. Also, the server 3 transmits the target information 322 to the user terminal 1 in response to a request from the user terminal 1 . In addition, the server 3 accepts an application for the object from the user terminal 1 and notifies the corresponding registrant terminal 2 to that effect. Further, the server 3 selects an appropriate target for the user based on the user information 321 and the target information 322, and transmits the target information 322 of the selected target to the user terminal 1, thereby selecting the target appropriate for the user. to recommend.

<Hardware configuration>
Next, the hardware configuration of the information processing device 100 will be described. FIG. 2 is a diagram showing an example of the hardware configuration of the information processing apparatus 100. As shown in FIG. As shown in FIG. 2, the information processing apparatus 100 includes a processor 101, a memory 102, a storage 103, a communication I/F 104, an input/output I/F 105, and drives, which are interconnected via a bus B. a device 106;

The processor 101 controls each component of the information processing apparatus 100 and implements the functions of the information processing apparatus 100 by loading the program stored in the storage 103 into the memory 102 and executing the program. The server 3 according to this embodiment is implemented by the processor 101 executing the recommendation program. The processor 101 is, for example, a CPU (Central Processing Unit), MPU (Micro Processing Unit), GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit), DSP (Digital Signal Processor), or a combination thereof.

The memory 102 is, for example, ROM (Read Only Memory), RAM (Random Access Memory), or a combination thereof. ROM is, for example, PROM (Programmable ROM), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), or a combination thereof. RAM is, for example, DRAM (Dynamic RAM), SRAM (Static RAM), MRAM (Magnetoresistive RAM), or a combination thereof.

The storage 103 stores an OS, recommendation programs, and various data. The storage 103 is, for example, flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), SCM (Storage Class Memories), or a combination thereof.

The communication I/F 104 is an interface for connecting the information processing device 100 to an external device via the network N and controlling communication. The communication I/F 104 is, for example, an adapter conforming to Bluetooth (registered trademark), Wi-Fi (registered trademark), ZigBee (registered trademark), Ethernet (registered trademark), or optical communication (for example, Fiber Channel). , but not limited to this.

The input/output I/F 105 is an interface for connecting the input device 107 and the output device 108 to the information processing apparatus 100 . The input device 107 is, for example, a mouse, keyboard, touch panel, microphone, scanner, camera, various sensors, operation buttons, or a combination thereof. Output device 108 is, for example, a display, projector, printer, speaker, vibrator, or a combination thereof.

The drive device 106 reads and writes data on the disk media 109 . Drive device 106 is, for example, a magnetic disk drive, an optical disk drive, a magneto-optical disk drive, or a combination thereof. The disk medium 109 is, for example, a CD (Compact Disc), DVD (Digital Versatile Disc), FD (Floppy Disk), MO (Magneto-Optical disk), BD (Blu-ray (registered trademark) Disc), or a combination thereof. is.

In this embodiment, the recommendation program may be written in the memory 102 or the storage 103 at the manufacturing stage of the server 3, may be provided to the server 3 via the network N, or may be stored in the disk media 109 or the like. It may be provided to the server 3 via a non-transitory computer-readable recording medium.

<Functional configuration>
Next, the functional configuration of the server 3 will be described. FIG. 3 is a diagram showing an example of the functional configuration of the server 3. As shown in FIG. As shown in FIG. 3 , the server 3 includes a communication section 31 , a storage section 32 and a control section 33 .

Communication unit 31 is realized by communication I/F 104 . The communication unit 31 transmits and receives information to and from the user terminal 1 and the registrant terminal 2 via the network N. FIG. The communication unit 31 receives user information 321 and target information 322 from the user terminal 1 and the registrant terminal 2, respectively. Also, the communication unit 31 transmits target information to the user terminal 1 .

The storage unit 32 is implemented by the memory 102 and storage 103 . The storage unit 32 stores user information 321 , object information 322 and action history information 323 .

The user information 321 is information about users. The user information 321 is registered by the user via the user terminal 1, for example. The storage unit 32 stores user information 321 of a plurality of users. FIG. 4 is a diagram showing an example of the user information 321. As shown in FIG. The user information 321 in FIG. 4 has “UID”, “age” and “gender” as information items. “UID” is user identification information. "Age" is the age of the user. “Gender” is the gender of the user. "Age" and "Gender" correspond to categorical data. In the example of FIG. 4, for example, the "age" of the user whose "UID" is U001 is 18, and the "gender" is M (male). This indicates that user U001 is an 18-year-old male. Note that the user information 321 is not limited to the example in FIG. The user information 321 may include some of the information items shown in FIG. 4, and information items other than those shown in FIG. , marital status, etc.).

The target information 322 is information about the target. The target information 322 is registered by the registrant via the registrant terminal 2, for example. The storage unit 32 stores target information 322 of a plurality of targets. FIG. 5 is a diagram showing an example of the target information 322. As shown in FIG. FIG. 5 shows target information 322 when the target is "work". The target information 322 in FIG. 5 has “JID”, “industry” and “area” as information items. “JID” is identification information of a job (target). The “industry” is the type of work. "Region" is the region where you work. "Industry" and "Region" correspond to categorical data. In the example of FIG. 5, for example, the "industry" of the job (target) whose "JID" is J001 is the restaurant business, and the "region" is Tokyo. This indicates that the job being offered is a restaurant in Tokyo. Note that the target information 322 is not limited to the example in FIG. The target information 322 may include some of the information items in FIG. 5, and may include information items other than those described in FIG. 5 (employment status, working hours, wages, shifts, etc.).

The action history information 323 is information about the user's action on the target on the online service provided by the information processing system. User actions are, for example, browsing, applying, evaluating, or sending messages, but are not limited to these. The action history information 323 is recorded, for example, each time the user takes action. The storage unit 32 stores action history information 323 of a plurality of users. Action history information 323 of a plurality of types of actions may be recorded in the storage unit 32 for each user. FIG. 6 is a diagram showing an example of the action history information 323. As shown in FIG. FIG. 6 shows the action history information 323 when the target is "work" and the action is "viewing target information". The action history information 323 of FIG. 6 has "J001", "J002" and "J003" as information items. "J001", "J002", and "J003" are the number of times the target information 322 of jobs J001, J002, and J003 has been browsed, respectively. In the example of FIG. 6, for example, "J001" of the user whose "UID" is U001 is 1, "J002" is 0, and "J003" is 2. This indicates that the user U001 browsed the target information 322 of the job J001 once, browsed the target information 322 of the job J002 zero times, and browsed the target information 322 of the job J003 twice. Note that the action history information 323 is not limited to the example in FIG. The action history information 323 may include some of the information items in FIG. 6, or may include information items other than those described in FIG.

The control unit 33 is implemented by the processor 101 reading and executing a program from the memory 102 and cooperating with other hardware configurations. The control unit 33 controls the overall operation of the server 3 . The control unit 33 includes an acquisition unit 331 , a similarity calculation unit 332 , a processing unit 333 , a recommendation calculation unit 334 , and a recommendation unit 335 .

The acquisition unit 331 acquires various types of information from an external device via the communication unit 31 . Specifically, the acquisition unit 331 acquires user information 321 of multiple users from multiple user terminals 1 . Also, the acquisition unit 331 acquires the target information 322 of multiple targets from multiple registrant terminals 2 . Further, the acquisition unit 331 acquires the action history information 323 of multiple users from the access log to the server 3 by the multiple user terminals 1 . The acquisition unit 331 may acquire the user information 321 of a plurality of users from one user terminal 1 or the target information 322 of a plurality of targets from one registrant terminal 2 .

The similarity calculation unit 332 calculates the similarity between values of categorical data included in the target information 322 based on the action history information 323 of multiple users and the target information 322 of multiple targets. Further, the similarity calculation unit 332 calculates the similarity between values of categorical data included in the user information 321 based on the action history information 323 of the plurality of users and the user information 321 of the plurality of users. may A method for calculating the degree of similarity will be described later in detail.

The processing unit 333 processes the user information 321 and the target information 322 . Specifically, based on the action history information 323 of one user and the target information 322 of a plurality of targets, the processing unit 333 processes the categorical data included in the target information 322 in the user information 321 of one user. is added, and the categorical data added to the user information 321 of one user is corrected based on the degree of similarity. Further, the processing unit 333 adds categorical data included in the user information 321 to the target information 322 of one target based on the action history information 323 of one user and the user information 321 of a plurality of users. , the categorical data added to the target information 322 of one target may be corrected based on similarity. A method of processing the user information 321 and the target information 322 by the processing unit 333 will be described later in detail.

The recommendation degree calculation unit 334 calculates the recommendation degree R of each target for one user based on the user information 321 of one user and the target information 322 of a plurality of targets. Further, the recommendation degree calculation unit 334 may calculate the recommendation degree R of each target for one user based on the target information 322 of one target and the user information 321 of a plurality of users. A method for calculating the recommendation level R will be described later in detail.

The recommendation unit 335 recommends a target to one user based on the recommendation degree R of each target calculated for the one user. Details of the recommendation method will be described later.

<Processing executed by the server device>
Next, processing executed by the server 3 will be described. FIG. 7 is a flow chart showing an example of processing executed by the server 3 . FIG. 8 is a diagram showing a specific example of processing executed by the server 3. As shown in FIG. In the example of FIG. 8, the object is work. FIG. 9 is a diagram illustrating a specific example of a similarity calculation method.

(Step S101)
The acquisition unit 331 acquires user information 321 of multiple users, target information 322 of multiple targets, and action history information 323 of multiple users, and stores them in the storage unit 32 . In the example of FIG. 8, the storage unit 32 stores user information 321 of users U001, U002, and U003, target information 322 of jobs J001, J002, and J003, and action history information 323 of users U001, U002, and U003. remembered. The acquisition unit 331 can acquire the user information 321, the target information 322, and the action history information 323 at arbitrary timing.

(Step S102)
The similarity calculation unit 332 calculates the similarity S between the values of the categorical data included in the target information 322 based on the action history information 323 of the plurality of users and the target information 322 of the plurality of targets, Save in the storage unit 32 . The similarity calculator 332 can calculate the similarity S at any timing. The similarity calculator 332 calculates the similarity S by, for example, the following method.

First, based on the action history information 323, the similarity calculation unit 332 aggregates the frequency (number of appearances) of the categorical data values included in the target information 322 for each user. In the example of FIG. 9, the frequencies of the values (Tokyo, Kanagawa, Chiba) of "Region", which is categorical data included in the target information 322, are aggregated for each user based on the number of views of each job. For example, the number of views of job J001 by user U001 is 1, and the "area" of job J001 is "Tokyo". Therefore, the frequency of the value of "Tokyo" for the user U001 is one.

Next, the similarity calculation unit 332 calculates the similarity s of frequency between values for each user. The degree of similarity s is the degree of similarity in frequency between values. The closer the frequencies, the higher the similarity s. As shown in FIG. 9, the similarity s is expressed in the form of a matrix.

For example, in the example of FIG. 9, since the frequencies of "Kanagawa" and "Chiba" of the user U002 are the same (both are 1), the similarity s of the frequencies of "Kanagawa" and "Chiba" of the user U002 is high. . On the other hand, since the frequencies of user U002's "Tokyo" and "Chiba" are different (2 and 1), the similarity s of the frequencies of user U002's "Tokyo" and "Chiba" is low. This means that from the user U002 point of view, “Kanagawa” and “Chiba” are similar (have the same degree of interest), and “Tokyo” and “Chiba” are not similar (the degree of interest is different). are doing.

Assuming that the frequencies of the aggregated values are p and q (where p+q=0, s=1), the similarity s is calculated by the following formula, for example.

The example of FIG. 9 shows the frequency similarity s between values calculated for each user by the above formula. For example, the similarity s between "Kanagawa" and "Chiba" for the user U002 is 1, and the similarity s between "Tokyo" and "Chiba" is 0.66.

After that, the similarity calculation unit 332 calculates the similarity S between values based on the similarity s of the frequency between values calculated for each user. The similarity S is, for example, an arithmetic mean or a weighted average of the frequency similarity s between values calculated for each user. As shown in FIG. 9, the similarity S is expressed in the form of a matrix.

The degree of similarity S between values obtained in this manner is a value that indicates the degree of similarity in the degree of interest in each value as viewed by a plurality of users. Therefore, for example, when the degree of similarity S between “Kanagawa” and “Chiba” is high, a user who browses the job of “Kanagawa” will be able to see the job of “Chiba” even if he or she has not browsed the job of “Chiba”. can be assumed to be of similar interest.

Note that the method of calculating the frequency of categorical data values, the method of calculating the degree of similarity s, and the method of calculating the degree of similarity S are not limited to the above examples. For example, the frequency of values of categorical data may be weighted for each user, or may be weighted for each user's behavior.

Further, when the target information 322 includes a plurality of categorical data, the similarity calculator 332 may calculate the similarity S for each categorical data. For example, in the example of FIG. 8, it is conceivable to calculate the similarity S1 of "industry" and the similarity S2 of "area".

Further, when the user information 321 includes one or a plurality of categorical data, the similarity calculator 332 may calculate the similarity S for each categorical data. For example, in the example of FIG. 8, it is conceivable to calculate the "age" similarity S3 and the "sex" similarity S4. When "age" is treated as categorical data, the age itself may be treated as a categorical data value, or an age division (20 years old or older/under 20 years old, etc.) may be treated as a categorical data value. good too.

(Step S103)
The processing unit 333 adds target information 322 to the user information 321 of one user based on the action history information 323 of one user and the target information 322 of a plurality of targets. Specifically, the processing unit 333 adds the categorical data included in the target information 322 aggregated based on the action history information 323 of one user to the user information 321 of one user. As a result, target information 322 corresponding to one user's behavior (interest) is added to the user information 321 .

In the example of FIG. 8, "Region" is added to the user information 321 of each user. The value of “area” for each user is the number of views of jobs for each area, which is aggregated based on the action history information 323 of each user. For example, the number of views of job J001 by user U001 is 1, and the "area" of job J001 is "Tokyo". Therefore, the frequency of the value of "Tokyo" for the user U001 is one.

(Step S104)
The processing unit 333 corrects the user information 321 processed in step S103. Specifically, the processing unit 333 corrects the categorical data added to the user information 321 of one user based on the similarity S of the categorical data. Specifically, the processing unit 333 multiplies the categorical data added to the user information 321 of one user by the corresponding similarity S. The processing unit 333 preferably normalizes the corrected categorical data.

In the example of FIG. 9, the "area" added to the user information 321 is multiplied by the similarity S of the "area". When "industry" is added to the user information 321, the "industry" added to the user information 321 may be multiplied by the similarity S of the "industry".

The processing unit 333 also replaces the values of the categorical data originally included in the user information 321 with numerical values. In the example of FIG. 9, "age" is replaced with 1 (20 years old or older) and 0 (less than 20 years old). Also, "sex" is replaced with 1 (M) and 0 (F).

(Step S105)
The processing unit 333 adds the user information 321 to the target information 322 . In the example of FIG. 8, "age" and "sex" are added to the object information 322 of each object. The values of "age" and "sex" of each user are set to 0 for convenience.

(Step S106)
The recommendation degree calculation unit 334 calculates the recommendation degree R based on the user information 321 processed in steps S103 and S104 and the target information 322 processed in step S105. More specifically, the recommendation degree calculation unit 334 calculates the recommendation degree R of each target for one user based on the user information 321 of one user and the target information 322 of a plurality of targets. The degree of recommendation R corresponds to the degree of similarity between the user information 321 and the target information 322 . It means that the closer the user information 321 and the target information 322 are to each other (recommendation level R), the closer the target is to the user's high interest.

上記の式において、kiは重み係数である。kiが全て１の場合、推薦度Ｒは、ユーザ情報３２１と対象情報３２２との内積となる。 Since the user information 321 and the target information 322 are information including common parameters through the processing up to step S105, both can be represented by n-dimensional vectors. When the user information 321 is (u1, u2, ..., un) and the target information 322 is (j1, j2, ..., jn), the recommendation level R is calculated by, for example, the following formula.

In the above formula, ki is a weighting factor. When all ki are 1, the recommendation degree R is the inner product of the user information 321 and the target information 322 .

For example, in the example of FIG. 8, the recommendation level R of the target J001 to the user U001 is 0.7, the recommendation level R of the target J002 is 0.1, and the recommendation level R of the target J003 is 0.2.

(Step S107)
The recommendation unit 335 recommends a target to one user based on the recommendation degree R of each target calculated for the one user. Specifically, the recommendation unit 335 transmits the target information 323 of the target selected based on the recommendation level R to the user terminal 1 and displays it on the display of the user terminal 1 .

For example, the recommendation unit 335 recommends a target with the highest recommendation degree R to the user. In this case, in the example of FIG. 8, the target J001 is recommended to the user U001, the target J001 is recommended to the user U002, and the target J002 is recommended to the user U003.

In addition, the recommendation unit 335 may recommend a target whose recommendation degree R to the user is equal to or higher than a threshold to the user. When the threshold value of the recommendation degree R is 0.8, in the example of FIG. 8, no target is recommended to the user U001, the target J001 is recommended to the user U002, and the targets J001 and J002 are recommended to the user U003.

It should be noted that the method of selecting a target for recommendation is not limited to the above example. The recommendation unit 335 can select a target by an arbitrary method based on the recommendation level R and recommend it to the user.

<Summary>
As described above, according to this embodiment, the target information 322 corresponding to the user's interest is added to the user information 321 based on the action history information 323 and the target information 322 . By recommending a target to the user based on this user information 321, even if the user information 321 is small, the target can be appropriately recommended to the user.

Also, the target information 322 added to the user information 321 is corrected by the similarity S. FIG. This makes it possible to estimate the degree of interest of the user in the object for which the object information 322 cannot be obtained from the action history information 323, calculate the degree of recommendation R based on the degree of interest, and recommend the object. As a result, even if the amount of target information 322 added to the user information 321 is small, that is, even if the user's action history information 323 is small, the target can be appropriately recommended to the user.

For example, in the example of FIG. 8, the user U001 has not viewed the target information 322 of the target J002 even once, and has viewed the target information 322 only three times in total. Therefore, it is not known whether the user U001 has not browsed the target information 322 of the target J002 by chance or has no interest in the target J002. That is, it is unknown whether the user U001 is interested in "Kanagawa". However, by correcting the target information 322 of the user U001 with the degree of similarity S, it is possible to estimate that the unknown degree of interest of the user U001 in "Kanagawa" is 0.1. By using the degree of interest estimated in this way, it is possible to recommend an object according to the user U001's interest to the user U001.

Also, the degree of recommendation R can be calculated by Σki·ui·ji. By adjusting the weighting factor ki, the degree of contribution of each piece of target information 322 can be easily adjusted when the degree of recommendation R is calculated. For example, when it is found that the contribution of certain target information 322 to the recommendation level R is small, the weighting coefficient ki of the target information 322 is reduced, and the target information 322 is reduced without modifying the target information 322 itself. It is possible to calculate the degree of recommendation R that reflects the degree of contribution.

In the above example, the user information 321 is added to the target information 322 in step S105 for convenience. However, the user information 321 may be added to the target information 322 by the same method as adding the target information 322 to the user information 321, and the added user information 321 may be corrected with the similarity S. FIG. Also in this case, the recommendation level R can be calculated in the same manner as described above.

<Appendix>
This embodiment includes the following disclosure.

(Appendix 1)
An information processing method in which an information processing device recommends a target to a user,
Acquisition processing for acquiring user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation process for calculating a similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
adding target information to the user information of the one user based on the action history information of the one user and the target information of the plurality of targets, and determining the target information of the one user based on the similarity A processing process for correcting by
a recommendation degree calculation process for calculating a recommendation degree of each target for the one user based on user information of the one user and target information of the plurality of targets;
recommendation processing for recommending a target to the one user based on the recommendation level of each target;
Information processing methods, including

(Appendix 2)
The degree of recommendation of one target for the one user is based on an inner product of the categorical data included in the target information of the one user and the categorical data included in the target information of the one target. The information processing method according to Supplementary Note 1, which is calculated by

(Appendix 3)
The information processing method according to appendix 1 or appendix 2, wherein the similarity between values of the categorical data is calculated based on a similarity in frequency between values calculated for each user.

(Appendix 4)
3. The information processing method according to any one of appendices 1 to 3, wherein the categorical data is region, industry, age, gender, or employment status.

(Appendix 5)
An information processing device that recommends a target to a user,
an acquisition unit that acquires user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation unit that calculates the similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
adding target information to the user information of the one user based on the action history information of the one user and the target information of the plurality of targets, and determining the target information of the one user based on the similarity a processing unit that corrects by
a recommendation degree calculation unit that calculates a recommendation degree of each target for the one user based on the target information of the one user and the target information of the plurality of targets;
a recommendation unit that recommends a target to the one user based on the recommendation level of each target;
Information processing device.

(Appendix 6)
information processing equipment,
Acquisition processing for acquiring user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation process for calculating a similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
adding target information to the user information of the one user based on the action history information of the one user and the target information of the plurality of targets, and determining the target information of the one user based on the similarity A processing process for correcting by
a recommendation degree calculation process for calculating a recommendation degree of each target for the one user based on the target information of the one user and the target information of the plurality of targets;
recommendation processing for recommending a target to the one user based on the recommendation level of each target;
A program for executing an information processing method including

The embodiments disclosed this time are illustrative in all respects and should be considered not restrictive. The scope of the present invention is indicated by the scope of the claims rather than the above-described meaning, and is intended to include all modifications within the scope and meaning equivalent to the scope of the claims. In addition, the present invention is not limited to the above-described embodiments, but can be modified in various ways within the scope of the claims, and can be obtained by appropriately combining technical means disclosed in different embodiments. Embodiments are also included in the technical scope of the present invention.

1: User terminal 2: Registrant terminal 3: Server device 31: Communication unit 32: Storage unit 33: Control unit 321: User information 322: Target information 323: Action history information 331: Acquisition unit 332: Similarity calculation unit 333: Processing unit 334: Recommendation degree calculation unit 335: Recommendation unit 101: Processor 102: Memory 103: Storage 104: Communication I/F
105: Input/output I/F
106: Drive Device 107: Input Device 108: Output Device 109: Disk Media

Claims (6)

An information processing method in which an information processing device recommends a target to a user,
Acquisition processing for acquiring user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation process for calculating a similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
Based on the action history information of the one user and the target information of the plurality of targets, the categorical data is added to the user information of the one user, and the categorical data of the one user is added to the similar A processing process that corrects based on the degree,
a recommendation degree calculation process for calculating a recommendation degree of each target for the one user based on user information of the one user and target information of the plurality of targets;
recommendation processing for recommending a target to the one user based on the recommendation level of each target;
Information processing methods, including The degree of recommendation of one target for the one user is based on an inner product of the categorical data included in the target information of the one user and the categorical data included in the target information of the one target. 2. The information processing method according to claim 1, wherein the calculation is performed by 3. The information processing method according to claim 1, wherein the similarity between values of the categorical data is calculated based on a frequency similarity between values calculated for each user. 4. The information processing method according to any one of claims 1 to 3, wherein said categorical data is area, industry, age, sex, or employment status. An information processing device that recommends a target to a user,
an acquisition unit that acquires user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation unit that calculates the similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
Based on the action history information of the one user and the target information of the plurality of targets, the categorical data is added to the user information of the one user, and the categorical data of the one user is added to the similar a processing unit that corrects based on the degree;
a recommendation degree calculation unit that calculates a recommendation degree of each target for the one user based on the user information of the one user and the target information of the plurality of targets;
a recommendation unit that recommends a target to the one user based on the recommendation level of each target;
Information processing device. information processing equipment,
Acquisition processing for acquiring user information of a plurality of users, action history information of the plurality of users, and target information of a plurality of targets including categorical data;
a similarity calculation process for calculating a similarity between values of the categorical data based on the action history information of the plurality of users and the target information of the plurality of targets;
Based on the action history information of the one user and the target information of the plurality of targets, the categorical data is added to the user information of the one user, and the categorical data of the one user is added to the similar A processing process that corrects based on the degree,
a recommendation degree calculation process for calculating a recommendation degree of each target for the one user based on user information of the one user and target information of the plurality of targets;
recommendation processing for recommending a target to the one user based on the recommendation level of each target;
A program for executing an information processing method including

Images