Selection system and selection program

Abstract

To provide a selection system and a selection program with which recommended content information can be appropriately selected.SOLUTION: An information processing system 100 for selecting content information recommended to a user from among multiple pieces of content information includes: generation means which generates, based on action log information indicating an action log of a user, third score ratio information of a user corresponding to the action log information; and selection means which selects recommended content information based on the third score ratio information generated by the generation means and second score ratio information corresponding to each of the pieces of content information. The third score ratio information is determined according to the number of predetermined keywords appearing in the action log information. The second score ratio information is determined according to the number of the predetermined keywords appearing in each of the pieces of content information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a selection system and a selection program. [Background technology]

[0002] Conventionally, a technique for selecting content to be recommended to a user has been known (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-170433 Summary of the Invention [Problem to be solved by the invention]

[0004] Meanwhile, there has been a demand for a technique for appropriately selecting recommended content information to be recommended to a user.

[0005] The present invention has been made in view of the above, and has an object to provide a selection system and a selection program that enable appropriate selection of recommended content information. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the objective, the selection system described in claim 1 is a selection system that selects recommended content information, which is content information to recommend to a user, from a plurality of content information, and includes: a generation means that generates log-side score information for the user corresponding to action log information that indicates the user's action log, based on the action log information; and a selection means that selects the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information, wherein the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information, and the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of content information.

[0007] The selection system described in claim 2 is the selection system described in claim 1, wherein there are a plurality of the specified keywords, and the generation means generates the log-side score information of the user based on first weight information indicating the weight assigned to each of the plurality of specified keywords and the behavior log information.

[0008] The selection system described in claim 3 is the selection system described in claim 1, wherein there are multiple types of user behavior logs, and the generation means generates the user's log-side score information based on the behavior log information and second weight information indicating weights determined for the multiple types of behavior log information depending on the type of the user's behavior log.

[0009] The selection system described in claim 4 is the selection system described in claim 1, wherein the user's behavior log includes multiple types of behavior logs that are performed sequentially, and the generation means generates the user's log-side score information based on third weight information that indicates weights determined for the multiple types of behavior log information according to the order in which the user's behavior logs are performed, and the behavior log information.

[0010] The selection program described in claim 5 is a selection program that selects recommended content information, which is content information to recommend to a user, from a plurality of content information, and causes a computer to function as: a generation means that generates log-side score information for the user corresponding to action log information that indicates the user's action log, based on the action log information; and a selection means that selects the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information, wherein the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information, and the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of content information. [Effects of the Invention]

[0011] According to the selection system described in claim 1 and the selection program described in claim 5, by selecting recommended content information based on log-side score information (information generated based on behavior log information) and content-side score information, it is possible to appropriately select recommended content information, for example.

[0012] According to the selection system described in claim 2, by generating a user's log-side score information based on first weight information indicating the weight assigned to each of a plurality of specified keywords and on the behavioral log information, it becomes possible to reflect, for example, the degree of importance of the keywords, thereby enabling the appropriate selection of recommended content information.

[0013] According to the selection system described in claim 3, by generating the user's log-side score information based on the second weight information indicating weights determined for multiple types of behavior log information depending on the type of user's behavior log and the behavior log information, it is possible to reflect, for example, the type of behavior log, thereby making it possible to appropriately select recommended content information.

[0014] According to the selection system described in claim 4, by generating the user's log-side score information based on the third weight information indicating weights determined for multiple types of behavior log information according to the order in which the user's behavior logs are performed, and the behavior log information, it is possible to reflect, for example, the order in which the behavior logs are performed, thereby making it possible to appropriately select recommended content information. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a block diagram of an information processing system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating keyword master information. [Figure 3] FIG. 10 is a diagram illustrating an example of medical dictionary master information. [Figure 4] FIG. 10 is a diagram illustrating an example of content master information. [Figure 5] FIG. 10 is a diagram illustrating content-specific keyword information. [Figure 6] FIG. 10 is a diagram illustrating an example of search log information. [Figure 7] FIG. 10 is a diagram illustrating an example of chat log information. [Figure 8] FIG. 10 is a diagram illustrating an example of content viewing log information. [Figure 9] 10 is a flowchart of a statistical process. [Figure 10] FIG. 10 is a diagram illustrating a processing result. [Figure 11] FIG. 10 is a diagram illustrating a processing result. [Figure 12] FIG. 10 is a diagram illustrating a processing result. [Figure 13] FIG. 10 is a diagram illustrating a processing result. [Figure 14] FIG. 10 is a diagram illustrating a processing result. [Figure 15] FIG. 10 is a diagram illustrating an example of log type weight information. [Figure 16] FIG. 10 is a diagram illustrating a processing result. [Figure 17] FIG. 10 is a diagram illustrating an example of execution order weight information. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, a selection system and a selection program according to an embodiment of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to the embodiment.

[0017] (Basic concept) First, the basic concept will be described. The selection system according to the present invention is a system for selecting recommended content information, which is content information recommended to a user from among a plurality of pieces of content information, and the concept includes, for example, a dedicated system for selecting recommended content information, or a system realized by installing a selection program and implementing a function for selecting recommended content information in a general-purpose system (for example, a general-purpose computer, a server computer, or a plurality of computers distributed over a network (i.e., so-called cloud computers)).

[0018] "Content information" is information indicating the content of the information to be transmitted, and is a concept that includes, for example, information intended to be transmitted to the user, such as advertising information, educational information, and guidance information. The format of content information is arbitrary, and includes, for example, text information, image information (video, still images), and audio information. In the following, "content information" will also be simply referred to as "content."

[0019] A "user" is a person to whom content information is recommended, and is a concept that includes, for example, medical professionals such as doctors, nurses, and laboratory technicians, as well as anyone other than medical professionals (for example, office workers, teachers, students, etc.).

[0020] In this embodiment, the case where the "content information" is advertising information and the format of the "content information" is text information or image information will be described as an example, and the case where the "user" is a doctor will be described as an example.

[0021] (composition) First, an information processing system according to the present embodiment will be described. Fig. 1 is a block diagram of an information processing system according to the present embodiment.

[0022] The information processing system 100 is a system including a selection system, and includes, for example, a terminal device 1 and a server device 2 that are capable of communicating with each other. Note that the number of elements included in the information processing system 100 is arbitrary, but in this embodiment, the description will focus on the elements shown in FIG.

[0023] (Configuration - Terminal Device) The terminal device 1 is a device used by each user, and is, for example, a personal computer, and includes, for example, a communication unit 11, an input unit 12, an output unit 13, a recording unit 14, and a control unit 15.

[0024] As a variation, the terminal device 1 may be any other terminal device including, for example, a tablet terminal, a smartphone, or the like.

[0025] (Configuration - Terminal Device - Communication Unit) The communication unit 11 is a communication means for communicating with an external device (for example, the server device 2). The specific type and configuration of the communication unit 11 are arbitrary, but it can be configured using, for example, a known communication circuit or the like.

[0026] (Configuration - Terminal Device - Input Section) The input unit 12 is an operation means for accepting various operation inputs from the user. The specific configuration of the input unit 12 is arbitrary, but it can be configured using information input devices such as a keyboard and a mouse, for example.

[0027] (Configuration - Terminal Device - Output Unit) The output unit 13 is an output means that outputs various information under the control of the control unit 15. The specific configuration of this output unit 13 is arbitrary, but it can be configured, for example, using a display that displays and outputs information.

[0028] (Configuration - Terminal Device - Recording Unit) The recording unit 14 is a recording means for recording programs and various data necessary for the operation of the terminal device 1, and is configured using, for example, a flash memory (not shown) as an external recording device. However, instead of or together with the flash memory, any other recording medium including a magnetic recording medium such as a hard disk or a magnetic disk, or an optical recording medium such as a DVD or a Blu-ray disc can be used (the same applies to the recording units of other devices).

[0029] (Configuration - Terminal Device - Control Unit) The control unit 15 is a control means for controlling the terminal device 1, and specifically, is a computer including a CPU, various programs interpreted and executed on the CPU (including basic control programs such as an OS and application programs that are started on the OS and realize specific functions), and an internal memory such as a RAM for storing programs and various data (this also applies to control units of other devices). In particular, the program according to the embodiment is installed in the terminal device 1 via an arbitrary recording medium or a network, thereby substantially configuring each unit of the control unit 15 (this also applies to control units of other devices).

[0030] (Configuration - Server Device) The server device 2 is a selection system, and includes, for example, a communication unit 21, a recording unit 22, and a control unit 23.

[0031] (Configuration - Server Device - Communication Unit) The communication unit 21 is a communication means for communicating with an external device (for example, the terminal device 1). The specific type and configuration of the communication unit 21 are arbitrary, but it can be configured in the same way as the communication unit 11, for example.

[0032] (Configuration - Server Device - Recording Unit) The recording unit 22 is a recording means for recording programs and various data necessary for the operation of the server device 2. The recording unit 22 stores, for example, keyword master information, medical dictionary master information, content master information, content-specific keyword information, search log information, chat log information, and content viewing log information.

[0033] (Configuration - Server Device - Recording Unit - Keyword Master Information) 2 is a diagram illustrating an example of keyword master information. "Keyword master information" is information related to keywords (predetermined keywords) that serve as the basis for processing, and for example, the information of each item shown in FIG. 2 is associated with each other.

[0034] "Company information" in Fig. 2 is information indicating companies related to keywords (in Fig. 2, it is the name of the company that defines the keyword, such as "XX Pharmaceutical"). Although not shown in Fig. 2, this company information may also include other pharmaceutical companies besides "XX Pharmaceutical," such as "XX Pharmaceutical."

[0035] "Medical department information" in FIG. 2 is information indicating a medical department related to a keyword (such as "cardiology" in FIG. 2).

[0036] "Keyword information" in FIG. 2 is information indicating a keyword (predetermined keyword) (in FIG. 2, "heart failure" is a keyword in the medical field related to the doctor who is the user, etc.).

[0037] The "weight information" in Fig. 2 is first weight information indicating the weight assigned to each keyword, and more specifically, information indicating the degree of importance of each keyword. In Fig. 2, the weight information used is numerical information that becomes larger as the weight increases (i.e., the degree of importance increases and the more important a keyword is). Note that the specific numerical values ​​exemplified in this embodiment are for the sake of convenience.

[0038] The "first score information" in Fig. 2 is information indicating the score that expresses the weight of each keyword as a number based on a predetermined reference value ("100"), and specifically, information indicating the score that relatively represents the weight of keywords related to common company information and medical department information. In Fig. 2, the weight information related to "XX Pharmaceuticals" and "Cardiology" is "3," "1," "2," and so on, which are "50," "16.6," "33.3," etc., obtained by dividing "100" proportionally based on the ratios of "50," "16.6," and "33.3."

[0039] In the keyword master information of Figure 2, for example, the information at the top indicates that the keyword "heart failure" related to "cardiology" was defined by "XX Pharmaceuticals," and that the weight and score of the keyword are "3" and "50."

[0040] 2 may be stored by any method, but for example, the company information, medical department information, keyword information, and weight information are stored by a person in charge of a company such as a pharmaceutical company inputting the information for any purpose (for example, for the purpose of advertising their own products (medicines, etc.) through content information). Also, for example, the first score information is stored by executing an information storage process described later (the same applies to each piece of information described later).

[0041] As an example, it is assumed that a company representative will input keyword information related to the company's products, and will input a larger value for weight information corresponding to keyword information that is more strongly related to the company's products that have a high sales priority.

[0042] (Configuration - Server Device - Recording Unit - Medical Dictionary Master Information) 3 is a diagram showing an example of medical dictionary master information. "Medical dictionary master information" is information related to keywords (predetermined keywords) that serve as the basis for processing, and for example, the information of each item shown in FIG. 3 is mutually associated.

[0043] "Medical department information" in FIG. 3 is information indicating the medical department related to the medical word (that is, keyword) (in FIG. 3, this information is common to the information of the same name in FIG. 2).

[0044] "Medical word information" in Figure 3 is medical word information indicating medical words. "Medical words" are the same as the keywords indicated by the keyword information in Figure 2. In Figure 3, the medical word information stored is information that is the same as the keyword information in Figure 2.

[0045] In such medical dictionary master information of FIG. 3, for example, the information at the top indicates that the medical word (that is, keyword) related to "cardiology" is "heart failure."

[0046] (Configuration - Server Device - Recording Unit - Content Master Information) 4 is a diagram showing an example of content master information. The "content master information" is information relating to content information that serves as a processing standard, and for example, the information items shown in FIG. 4 are mutually associated.

[0047] "Company information" in FIG. 4 is information indicating companies related to the content information. In FIG. 4, this is the name of the company that advertises its own products through the corresponding content information, such as "XX Pharmaceuticals." Although not shown in FIG. 4, this company information also stores other pharmaceutical companies besides "XX Pharmaceuticals," such as "XX Pharmaceuticals." Furthermore, companies related to the content information also include companies that are related to the content information through elements other than advertising their own products, but in this embodiment, advertising will be mainly used as an example.

[0048] "Medical department information" in FIG. 4 is information indicating the medical department related to the content information ("cardiology department" in FIG. 4, etc.).

[0049] "Content type information" in FIG. 4 is information indicating the type of content information ("product information" in FIG. 4, etc.).

[0050] "Content name information" in FIG. 4 is information indicating the name of the content information ("Content A" in FIG. 4, etc.).

[0051] In the content master information of Figure 4, for example, the information at the top indicates that the company and medical department associated with content information "Content A" are "XX Pharmaceuticals" and "Cardiology," and that the type of content information is "Product Information" (i.e., content that introduces product information).

[0052] (Configuration - Server Device - Recording Unit - Keyword Information by Content) Fig. 5 is a diagram showing an example of content-specific keyword information. "Content-specific keyword information" is information related to keyword information (i.e., medical word information) included in each piece of content information, and for example, the information items shown in Fig. 5 are mutually associated.

[0053] The "content name information" and "keyword information" in FIG. 5 are the same as the information with the same names in FIGS.

[0054] 5 is information indicating the number of times the keyword indicated by the keyword information appears in the content information indicated by the content name information (i.e., the number of keywords included in the content information). In Fig. 5, the number is "2", which indicates that "heart failure" appears twice in the content information of "content A" (i.e., the keyword "heart failure" is included twice in "content A").

[0055] The "second score information" in Fig. 5 is information indicating the score that numerically represents the number of times each keyword appears and its weight. In Fig. 5, it is the result of multiplying the first score information in Fig. 2 by the number of times information in Fig. 5, such as "100" which is the multiplication result of 50 x 2.

[0056] The "second score ratio information" in Figure 5 is content-side score information corresponding to each piece of content information, and is information determined according to the number of times a predetermined keyword appears in each piece of content information. This "second score ratio information" is information indicating a score in which the second score information, which indicates the number of times each keyword appears and its weight in each piece of content, is expressed as a numerical value based on a predetermined reference value ("100"). Specifically, it is information indicating a score that relatively represents the second score information of keywords related to common content name information. In Figure 5, the second score information for "Content A" is "66.7", "11.1", "22.2", etc., which are obtained by dividing "100" proportionally at the ratios of "100", "16.6", and "33.3".

[0057] 5, for example, the information in the top row indicates that "heart failure" appears twice in the content information of "content A," and that the second score information for "heart failure" is "100," which is the calculation result of 50 (first score information corresponding to "heart failure" in FIG. 2) x 2 (number of occurrences information corresponding to "heart failure" in FIG. 5). It also indicates that the second score ratio information for "heart failure" in the content information of "content A" is "66.7."

[0058] (Configuration - Server Device - Recording Unit - Search Log Information) Fig. 6 is a diagram illustrating search log information. The "search log information" is behavior log information showing a user (doctor) behavior log, and more specifically, is information showing a log (history) of search words searched for via information search application software accessible by the user via the terminal device 1 (software that searches for and displays information such as web pages containing information corresponding to a search word when the search word is entered), and for example, the information items shown in Fig. 6 are associated with each other.

[0059] The information search application software is a concept that includes, for example, known browser-based Internet search engines.

[0060] The "Department Information" and "Doctor Information" in Figure 6 are the department to which the user doctor belongs (i.e., the department in which the doctor specializes) and information that identifies the doctor (name, medical registration number, etc.) (in Figure 6, "Cardiology", "AAAA", etc.).

[0061] The "date information" in FIG. 6 is the date indicating the date on which the user performed the search (eg, "October 28th" in FIG. 6).

[0062] "Keyword information" in FIG. 6 is information indicating keywords (medical word information in FIG. 3) included in search words input by the user via the terminal device 1 for information search.

[0063] In the search log information in Figure 6, for example, the information at the top shows that the search words entered by "AAAA," a doctor in the cardiology department, to search for information on October 28th included "heart failure."

[0064] (Configuration - Server Device - Recording Unit - Chat Log Information) Fig. 7 is a diagram illustrating chat log information. "Chat log information" is behavior log information showing a user (doctor) behavior log, and more specifically, it is information showing a log (history) of information (hereinafter referred to as "transmitted information") that the user has transmitted to others or received from others via chat application software (software for transmitting information to others using text information, etc.) accessible via the terminal device 1. For example, the information items shown in Fig. 7 are associated with each other.

[0065] The "medical department information" and "doctor information" in FIG. 7 are the same as the information with the same names in FIG.

[0066] The "Date Information" in Figure 7 is information indicating the date on which a user transmitted "Communication Information" to another person via chat application software, or the date on which the "Communication Information" was transmitted from another person (e.g., "October 28th" in Figure 7).

[0067] The "chat body information" in FIG. 7 indicates the aforementioned "transmission information" (in FIG. 7, "Transferred to CCU due to post-PCI" etc.).

[0068] In the chat log information in Figure 7, for example, the information at the top indicates that "AAAA", a doctor in the cardiology department, sent or was sent the message "Transferred to CCU after PCI" on October 28th.

[0069] (Configuration - Server Device - Recording Unit - Content Viewing Log Information) Fig. 8 is a diagram illustrating content viewing log information. The "content viewing log information" is action log information showing the action log of a user (doctor), and more specifically, is information showing the log (history) of content information viewed by the user via content information viewing application software (software for allowing the user to view content information) accessible via the terminal device 1, and for example, the information items shown in Fig. 8 are associated with each other.

[0070] The "medical department information" and "doctor information" in FIG. 8 are the same as the information with the same names in FIG.

[0071] The "date information" in FIG. 8 is information indicating the date on which the user viewed the content information via the content information viewing application software (such as "October 28th" in FIG. 7).

[0072] The "viewed content name information" in Figure 8 is information indicating the name of the content information viewed by the user (i.e., the content information that the content information viewing application software allowed the user to view) (e.g., "Content A" in Figure 8).

[0073] In the content viewing log information of Figure 8, for example, the information at the top indicates that "AAAA", a doctor in the cardiology department, viewed the content information for "Content A" on October 28th.

[0074] (Configuration - Server Device - Control Unit) The control unit 23 is a control means that controls the server device 2. The control unit 23 conceptually includes, for example, a generating means and a selecting means.

[0075] ===Generation means=== The "generation means" is a means for generating log-side score information of a user corresponding to action log information based on action log information indicating the user's action log. The "generation means" generates the log-side score information of a user based on, for example, first weight information indicating weights assigned to each of a plurality of predetermined keywords and the action log information.

[0076] ===Selection Method=== The "selection means" is a means for selecting recommended content information based on the log score information generated by the generation means and the content score information corresponding to each of the plurality of content information. The processing performed by each unit of the control unit 23 will be described later.

[0077] (process) Next, the information storage process, statistical process, and recommendation process performed by the information processing system 100 configured as above will be described.

[0078] (Processing - Information storage processing) First, the information storage process will be described. The "information storage process" is a process for storing part of the information in the recording unit 22 of FIG.

[0079] ===Keyword Master Information=== The process of storing the first score information of the keyword master information in FIG. 2 is started when information other than the first score information is input and stored.

[0080] The control unit 23 acquires weight information that has the company information and medical department information in Figure 2 in common, and identifies and stores the first score information corresponding to each weight information by dividing the reference value of "100" proportionately based on the proportion of each acquired weight information.

[0081] Here, for example, the weight information for "XX Pharmaceutical" and "Cardiology" is obtained as "3," "1," and "2," and the reference value "100" is apportioned in proportion to the obtained "3," "1," and "2," thereby identifying and storing the first score information as "50" (the calculation result of "100 x 3 / (3 + 1 + 2)"), "16.6" (the calculation result of "100 x 1 / (3 + 1 + 2)"), and "33.3" (the calculation result of "100 x 2 / (3 + 1 + 2)") corresponding to each weight information.

[0082] In addition, similar processing is performed on the weight information relating to "XX Pharmaceuticals" and "Hematology" and the information is stored as shown in the example of Figure 2.

[0083] ===Medical Dictionary Master Information=== The process of storing the medical dictionary master information in FIG. 3 is repeatedly started at predetermined time intervals.

[0084] The control unit 23 compares the combination of medical department information and keyword information in the keyword master information of Figure 2 with the combination of medical department information and medical word information in Figure 3, and if there is a combination of medical department information and keyword information (Figure 2) in the medical dictionary master information of Figure 3 that is not stored as a combination of medical department information and medical word information, it stores that combination of medical department information and keyword information as a combination of medical department information and medical word information in Figure 3.

[0085] Here, for example, unlike the example in Figure 3, if "cardiology" and "heart failure" in Figure 3 have not yet been stored, "cardiology" and "heart failure" exist as medical department information and keyword information in Figure 2, so "cardiology" and "heart failure" are stored as medical department information and medical word information in Figure 3, as shown in the top row of Figure 3.

[0086] By performing this processing, information common to the medical treatment information and keyword information in FIG. 2 is stored as the medical dictionary master information in FIG.

[0087] ===Content master information and content-specific keyword information=== We will explain an example in which the above-mentioned content information viewing application software is implemented in the server device 2, and the content information recorded in the recording unit 22 of the server device 2 can be viewed by the user via the terminal device 1.

[0088] The process of storing the content master information in Figure 4 and the content-specific keyword information in Figure 5 is initiated, for example, when a company representative or any other person inputs the content information itself (for example, advertising information indicating the content of the information to be transmitted), company information corresponding to the content information, medical department information, content type information, and content name information in order to record the content information in the recording unit 22.

[0089] First, the control unit 23 stores the content information itself input when the process is started in the recording unit 22 (not shown in Figure 1), and also stores information corresponding to the company information, medical department information, content type information, and content name information input when the process is started as information of the same name in Figure 4.

[0090] Here, for example, when the content information of "Content A," "XX Pharmaceuticals," "Cardiology," "Product Information," and "Content A" are input, the content information of "Content A" itself is recorded in the recording unit 22, and the information in the top row of Figure 4 is also stored.

[0091] Next, the control unit 23 stores information corresponding to the content name information input when the process is started as information of the same name in FIG.

[0092] Here, for example, "Content A" in the top row of FIG. 5 is stored.

[0093] Next, the control unit 23 identifies the number of times the keyword indicated by the medical word information in the medical dictionary master information of Figure 3 appears in the content information input when the process is started, and stores the keyword and the number of times the keyword appears as keyword information and number of times information in Figure 5.

[0094] Here, for example, in the content information of "Content A," the occurrence counts of the medical word information "heart failure," "arrhythmia," and "PCI" in the medical dictionary master information of Figure 3 are identified as 2, 1, and 1, and the information from the top to the third rows of Figure 4 is stored.

[0095] Next, the control unit 23 refers to the keyword master information in Figure 2, identifies the first score information associated with the same keyword information as the keyword information in Figure 5, multiplies the identified first score information by the aforementioned stored count information in Figure 5, and stores the multiplication result as the second score information in Figure 5.

[0096] Here, for example, by referring to the keyword master information in Fig. 2, "50", "16.6", and "33.3" are identified as the first score information associated with "heart failure", "arrhythmia", and "PCI". Next, for "heart failure", the identified "50" is multiplied by "2" (the number of times information in Fig. 5), and the multiplication result "100" is stored as the second score information in Fig. 5. Furthermore, by performing the same process for "arrhythmia" and "PIC", "16.6" and "33.3" are stored as the second score information in Fig. 5.

[0097] Next, the control unit 23 acquires second score information that shares the content name information of Figure 5 (i.e., the content name information input and stored when the process is started), and identifies and stores second score ratio information corresponding to each piece of second score information by dividing the reference value of "100" proportionately based on the ratio of each piece of acquired second score information.

[0098] Here, for example, the second score information for "Content A" is obtained as "100", "16.6", and "33.3", and by dividing the reference value "100" proportionately based on the obtained ratios of "100", "16.6", and "33.3", the second score ratio information corresponding to each piece of second score information is identified and stored as "66.7" (the calculation result of "100 x 100 / (100 + 16.6 + 33.3)"), "11.1" (the calculation result of "100 x 16.6 / (100 + 16.6 + 33.3)"), "22.2" (the calculation result of "100 x 33.3 / (100 + 16.6 + 33.3)")

[0099] ===Search log information=== Each user, i.e., a doctor, will log in by entering his / her user ID (identification information that allows his / her department and name to be uniquely identified on the server device 2 side) into the terminal device 1, and then perform an information search via the information search application software.

[0100] The process of storing search log information in Figure 6 is initiated, for example, when a user inputs a search word, the terminal device 1 transmits the user ID of the input user and the search word to the server device 2, and the process is initiated when the server device 2 receives the information.

[0101] First, the control unit 23 acquires the user ID and search word received when the process is started, and identifies the medical department and name of the user corresponding to the acquired user ID using any method (for example, a method in which information relating the user ID, medical department, and name to each other is stored in the recording unit 22, and the information is referenced to identify the user).

[0102] Here, for example, if a user "AAAA" inputs his / her user ID and the search word "cases of heart failure," "cardiology" and "AAAA" are identified.

[0103] Next, the control unit 23 identifies words (i.e., keywords) from the acquired search words that correspond to the medical word information in the medical dictionary master information of Fig. 3. In this case, the control unit 23 may be configured to identify keywords regardless of the medical department information of Fig. 3, or may be configured to identify keywords that are associated with the same medical department information (Fig. 3) as the identified medical department of the user and that correspond to the medical word information (Fig. 3).

[0104] Here, for example, since "heart failure" in the search word "cases of heart failure" is stored as medical word information in FIG. 3, "heart failure" is identified.

[0105] Next, the control unit 23 stores the identified user's medical department, name, current date (the current date identified by a timing means such as a timer in the server device 2), and the identified keywords as the medical department information, doctor information, date information, and keyword information in Figure 6.

[0106] For example, if the current date is October 28th, the information in the top row of FIG. 6 is stored.

[0107] ===Chat log information=== Each user, i.e., each doctor, logs into the chat application software using his / her own user ID and communicates information with others.

[0108] Regarding the process of storing chat log information in Figure 7, for example, when a user inputs "communication information" to another person, the user ID of the input user and the "communication information" are sent to the server device 2 (directly from the terminal device 1 or via any other computer), and the process is initiated when the server device 2 receives the information.

[0109] First, the control unit 23 acquires the user ID and "transmission information" received when the process is started, and identifies the medical department and name of the user corresponding to the acquired user ID by any method.

[0110] Here, for example, if a user "AAAA" inputs his / her own user ID and "transferred to CCU due to post-PCI" as "transfer information," "cardiology" and "AAAA" are identified.

[0111] Next, the control unit 23 stores the identified user's medical department, name, current date, and the acquired "communication information" as the medical department information, doctor information, date information, and chat text information in Figure 7.

[0112] For example, if the current date is October 28th, the information in the top row of FIG. 7 is stored.

[0113] As mentioned above, the chat log information in Figure 7 stores not only the "communication information" that the user has communicated to others, but also information corresponding to the "communication information" that others have communicated to the user. However, the latter information can also be stored by applying any processing, including the processing similar to that described above.

[0114] ===Content Viewing Log Information=== Regarding the operation of the content information viewing application software, the server device 2 allows a user who has logged in by entering a user ID into the terminal device 1 to view the content information recorded in the recording unit 22 using any viewing method.

[0115] As an optional viewing method, for example, a method may be adopted in which one or more of the content information recorded in the recording unit 22 is selected, and a viewing image (for example, a button image corresponding to a so-called banner advertisement image) for viewing the selected one or more content information is displayed on the screen that the user is viewing via the terminal device 1 (for example, a part of the screen displayed on the display that is the output unit 13 (for example, a part of the lower right side of the screen)), and the content information corresponding to the viewing image selected by a click operation or the like via the input unit 12 is viewed.

[0116] The one or more pieces of content information may be selected arbitrarily or by other methods.

[0117] Alternatively, for example, a method may be adopted in which the user is prompted to input a search keyword and is then allowed to view content information corresponding to the search keyword input by the user.

[0118] The process of storing content viewing log information in FIG. 8 is started, for example, when the server device 2 displays content information via the terminal device 1 and allows the content information to be viewed.

[0119] First, the control unit 23 uses an arbitrary method to identify the medical department and name of the user corresponding to the input user ID, and also identifies the name of the content information that has been browsed.

[0120] Here, for example, "cardiology" and "AAAA" are specified, and "content A" is specified as the name of the content information that the user is allowed to view.

[0121] Next, the control unit 23 stores the identified user's medical department, name, current date, and the name of the content information that the user has viewed as the medical department information, doctor information, date information, and viewed content name information in Figure 8.

[0122] Here, for example, if the current date is October 28th, the information in the top row of Fig. 8 is stored. This concludes the explanation of the information storage process.

[0123] (Processing - Statistical Processing) Next, the statistical processing will be described. Fig. 9 is a flowchart of the statistical processing (hereinafter, each step will be referred to as "S"). The statistical processing is processing performed by the server device 2, and is generally processing for grasping the characteristics and tendencies of user behavior, and is processing performed, for example, mainly based on behavior log information. The timing for executing this statistical processing is arbitrary; for example, it may be executed when the behavior log information of Figs. 6 to 8 is newly stored, or it may be executed repeatedly at predetermined time intervals.

[0124] 10 to 14 are diagrams showing examples of the processing results, and the following description will be given with reference to these figures as appropriate.

[0125] ===SA1=== At SA1 in FIG. 9, the control unit 23 identifies the number of times each piece of keyword information appears for each doctor based on the search log information in FIG.

[0126] Specifically, in the search log information of Figure 6, the number of times each piece of keyword information appears is identified by identifying the number of times each piece of keyword information is stored for a combination of common medical department information and doctor information.

[0127] For example, in this case, for "AAAA," who is a doctor in the "cardiology" department in Fig. 6, only one piece of keyword information, "heart failure," is stored, so the number of occurrences of "heart failure" is identified as "1," as shown in the top row of Fig. 10. In other words, it is identified that "heart failure" has appeared once in the search words for "AAAA."

[0128] Note that the search log information in Figure 6 is assumed to be accumulated information from a relatively long period of time in the past, but from the perspective of understanding user characteristics and trends based on behavior log information for a predetermined period, processing may be configured to be performed based on search log information within a predetermined period. Note that the predetermined period is arbitrary, and may be, for example, a period designated by an administrator (for example, a person in charge of advertising the deadline related to the content information), or a predetermined period such as the most recent month. Note that the same applies to SA2 and SA3 described below.

[0129] ===SA2=== At SA2 in FIG. 9, the control unit 23 identifies the number of times each piece of keyword information appears for each doctor based on the chat log information in FIG.

[0130] Specifically, in the chat log information of Figure 7, all chat text information is obtained for the combination of common medical department information and doctor information, and in the obtained chat text information, words (i.e., keywords) that correspond to the medical word information in the medical dictionary master information of Figure 3 and the number of times that each word appears are identified.

[0131] In this case, the system may be configured to identify keywords, etc., regardless of the medical department information in Figure 3, or it may be configured to identify keywords, etc., that are associated with the same medical department information (Figure 3) as the corresponding doctor's medical department and that correspond to the medical word information (Figure 3).

[0132] Here, for example, for "AAAA", a doctor in the "cardiology department" in Figure 7, the chat text information "Transferred to CCU after PCI", "Please carry out the necessary handover", and "Arrhythmia detected, XX is being considered" are obtained, and since "PCI" and "arrhythmia" (the medical word information in Figure 3) are stored in these, the occurrence counts of "PCI" and "arrhythmia" are identified as "1" and "1", as shown in the top and second rows of Figure 11. In other words, it is identified that "PCI" and "arrhythmia" each appeared once in the "transmission information" of "AAAA".

[0133] ===SA3=== In SA3 of FIG. 9, the control unit 23 identifies the number of times each piece of keyword information appears for each doctor based on the content viewing log information of FIG.

[0134] Specifically, in the content viewing log information of Figure 8, all viewed content name information is obtained for the combination of common medical department information and doctor information, and the viewed content name information, keyword information, and number of times information in the content-specific keyword information of Figure 5 are referenced to identify the keyword information contained in the content information indicated by the aforementioned obtained viewed content name information and the number of times the keyword information appears.

[0135] In addition, in the content viewing log information of Figure 8, it is expected that there may be multiple pieces of the same viewed content name information (for example, ``Content A'' in Figure 8) for a combination of common medical department information and doctor information, so the number of times that keyword information corresponding to that viewed content name information appears is determined by multiplying the number of times information in Figure 5 by the number of pieces of that viewed content name information.

[0136] Here, for example, for "AAAA", who is a doctor in the "cardiology department" of Fig. 8, "viewed content name information" of "content A", "content B", and "content A" is obtained. Next, in the content-specific keyword information of Fig. 5, "content A" is associated with "2" for "heart failure", "1" for "arrhythmia", and "1" for "PCI", and "content B" is associated with "heart failure" and "2" for "arrhythmia", so as shown in the third row from the top of Fig. 12, the occurrence counts of "heart failure", "arrhythmia", and "PCI" are identified as "5", "4", and "2".

[0137] Note that the number of occurrences in the top row of Figure 12, "5", is the result of calculating "2" (number of occurrences information for "heart failure" in "content A" in Figure 5) x "2" (the number of contents A acquired as described above) + "1" (number of occurrences information for "heart failure" in "content B" in Figure 5) x "1" (the number of contents B acquired as described above).

[0138] Furthermore, the number of occurrences in the second row of Figure 12, "4", is the result of calculating "2" (number of occurrences information for "arrhythmia" in "content A" in Figure 5) x "2" (the number of contents A acquired as described above) + "2" (number of occurrences information for "arrhythmia" in "content B" in Figure 5) x "1" (the number of contents B acquired as described above).

[0139] Furthermore, the number of occurrences in the third row of Figure 12, "2", is the result of calculating "1" (number of occurrences information for "PCI" of "Content A" in Figure 5) x "2" (the number of Content A acquired as described above).

[0140] ===SA4=== In SA4 of FIG. 9, the control unit 23 counts the number of occurrences identified in SA1 to SA3 for each doctor.

[0141] Specifically, in FIGS. 10 to 12, the total number of occurrences of each piece of mind word information is calculated and tallied for the combination of common medical department information and doctor information.

[0142] Here, for example, for "AAAA", a doctor in the "cardiology department", as shown in the top row of Figure 13, "6" (the calculation result of "1 + 5") for "heart failure" is calculated and totaled, "6" ("1 + 1 + 4") for "arrhythmia" is calculated and totaled, and "3" (the calculation result of "1 + 2") for "PCI" is calculated and totaled.

[0143] ===SA5=== 9, the control unit 23 generates third score ratio information for each doctor. The "third score ratio information" is log-side score information generated based on the behavior log information, and is, for example, information indicating the characteristics and tendencies of the user's behavior.

[0144] Specifically, by referring to the keyword master information in Figure 2, the first score information associated with the same keyword information as the keyword information corresponding to the number of occurrences tallied in SA4 (keyword information in Figure 13) is identified, the identified first score information is multiplied by the number of occurrences of the corresponding keyword information tallied in SA4, and the result of the multiplication is identified as the third score information (Figure 14).

[0145] Here, for example, by referring to the keyword master information in Fig. 2, "50", "16.6", and "33.3" are identified as the first score information associated with "heart failure", "arrhythmia", and "PCI". Next, for "heart failure", the identified "50" is multiplied by "6" (the number of occurrences in Fig. 13), and the multiplication result "300" is identified as the third score information in Fig. 14. Furthermore, by performing the same process for "arrhythmia" and "PIC", "99.6" and "99.9" are identified as the third score information (Fig. 14).

[0146] Next, third score information that has common department information and doctor information (i.e., for the same doctor) is obtained, and the reference value of "100" is divided proportionately based on the proportion of each piece of obtained third score information to identify and generate third score ratio information corresponding to each piece of third score information.

[0147] Here, for example, the third score information for "cardiology" and "AAAA" (i.e., for the doctor "AAAA"), "300," "99.6," and "99.9," is obtained, and the reference value of "100" is apportioned based on the obtained ratios of "300," "99.6," and "99.9." As shown in the third row from the top of Figure 14, "60.1" (the calculation result of "100 x 300 / (300+99.6+99.9)"), "19.9" (the calculation result of "100 x 99.6 / (300+99.6+99.9)"), and "20" (the calculation result of "100 x 99.9 / (300+99.6+99.9)") corresponding to each third score information are identified and stored as the third score ratio information.

[0148] For example, as shown in the leftmost column of Fig. 14 (the column "Company Information"), the keyword master information of Fig. 2 may be referenced to identify company information corresponding to the combination of each medical department information and keyword information of Fig. 14. This concludes the explanation of the statistical processing.

[0149] (Processing - Recommended Processing) Next, the recommendation process will be described. The "recommendation process" is a process of selecting and presenting content information (recommended content information) recommended to each user from the content information recorded in the recording unit 22 in Fig. 1 (the content information itself is not shown, but is information shown in the content-specific keyword information in Fig. 5), and is generally a process executed by the server device 2.

[0150] The timing for executing this recommendation process is arbitrary, and for example, it may be executed repeatedly at predetermined time intervals.

[0151] Here, an example will be described in which, for example, a doctor "AAAA" (see FIG. 6, etc.) logs in by entering his / her own user ID.

[0152] In general, the control unit 23 selects content information corresponding to the characteristics and tendencies of the user's behavior based on the processing results of the statistical processing in FIG. 9, and presents the content information to the user.

[0153] Specifically, first, the control unit 23 identifies the combination of keyword information and third score ratio information in Figure 14 identified by the processing results of SA4 and SA5 in Figure 9 that corresponds to the logged-in user (i.e., the user to whom content information is to be recommended).

[0154] Here, for example, since the doctor "AAAA" is logged in, the combination of "heart failure" and "60.1", the combination of "arrhythmia" and "19.9", and the combination of "PCI" and "20" shown in the third row from the top of Figure 14 are identified.

[0155] Next, the control unit 23 refers to the keyword information and second score ratio information of the content-specific keyword information in Figure 5, and identifies, using any identification method, content name information among the content name information in Figure 5 whose corresponding keyword information and second score ratio information are similar to the keyword information and third score ratio information in Figure 14 identified above, and then selects the content information corresponding to the identified content name information as recommended content information.

[0156] In addition, in the method for identifying content name information, for example, the more keyword information by content in Figure 5 contains that is the same as the keyword information in Figure 14 (i.e., information based on user behavior log information), the easier it is to identify it as similar, and also, the closer the second score ratio information is to the third score information in Figure 14 (i.e., information based on user behavior log information), the easier it is to identify it as similar.

[0157] Furthermore, the specification method may be configured to specify only one piece of content name information that is most similar, or may be configured to specify a plurality of pieces of content name information.

[0158] In addition, the identification method may employ a method of identification using a trained model generated by performing machine learning using a predetermined method.

[0159] Here, for example, after identifying "Content A" in Figure 5, content information corresponding to the identified "Content A" is selected as recommended content information from among multiple content information (not shown) stored in the recording unit 22.

[0160] Next, the control unit 23 presents the content information of "content A" selected as recommended content information to the logged-in user. The specific method is arbitrary, but for example, a button image (image for viewing) corresponding to an image of a banner advertisement corresponding to the recommended content information may be displayed on the screen viewed by the user via the terminal device 1, and when the user selects the image for viewing by a click operation or the like, the recommended content information may be displayed and presented. In other words, the recommended content information may be presented via the content information viewing application software described above, but is not limited to this and may be presented by any other method. This concludes the description of the recommendation process.

[0161] (Effects of the embodiment) Thus, according to this embodiment, by selecting recommended content information based on the third score ratio information (log-side score information) (information generated based on behavior log information) and the second score ratio information (content-side score information), it is possible to appropriately select recommended content information, for example.

[0162] Furthermore, by generating a user's log-side score information based on weight information (first weight information) indicating the weight assigned to each of a plurality of predetermined keywords, and search log information, chat log information, and content viewing log information (behavior log information), it becomes possible to reflect, for example, the degree of importance of keywords, thereby enabling appropriate selection of recommended content information.

[0163] [Modifications to the embodiment] Although the embodiments of the present invention have been described above, the specific configurations and means of the present invention can be modified and improved as desired within the scope of the technical concept of the present invention as set forth in the claims. Such modifications will be described below.

[0164] (About the problem to be solved and the effects of the invention) First, the problems that the invention aims to solve and the effects of the invention are not limited to those described above, and may vary depending on the implementation environment of the invention and the details of the configuration, and may solve only some of the problems described above or achieve only some of the effects described above.

[0165] (Regarding decentralization and integration) Furthermore, the electrical components described above are functional concepts and do not necessarily have to be physically configured as shown in the drawings. In other words, the specific form of distribution or integration of each part is not limited to that shown in the drawings, and all or part of them can be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc. Furthermore, the term "device" in this application is not limited to a single device, but includes a device configured from multiple devices.

[0166] (shape, numbers, structure, time series) The components illustrated in the embodiments and drawings may be modified and improved as desired within the scope of the technical concept of the present invention in terms of shape, numerical value, or the structure or chronological relationship of multiple components.

[0167] (About types of behavior logs) Also, the log-side score information of a user may be generated taking into consideration the type of action log (i.e., the type of action log information). Fig. 15 is a diagram illustrating an example of log type weight information, and Fig. 16 is a diagram illustrating an example of the processing result.

[0168] The log type weight information in Fig. 15 is stored in the recording unit 22 in Fig. 1. "Log type weight information" is information indicating a weight corresponding to the type of action log information, and for example, the information of each item shown in Fig. 15 is associated with each other.

[0169] The "log type information" in Figure 15 is information indicating the type of action log information (in Figure 15, "search" indicates the type corresponding to search log information, "chat" indicates the type corresponding to chat log information, and "content viewing" indicates the type corresponding to content viewing log information).

[0170] The "weight information" in Fig. 15 is second weight information indicating weights determined for multiple types of action log information according to the type of user's action log, and specifically, information indicating the degree of importance determined for the type of action log information. In Fig. 15, numerical information that becomes larger as the weight becomes higher (i.e., the higher the degree of importance, the more important) is used as the weight information.

[0171] 15, for example, the information at the top indicates that the weight of the action log information corresponding to "search" (i.e., search log information) is 3. The log type weight information may be stored based on information input to the server device 2 (the same applies to the information in FIG. 17 described below).

[0172] ===Processing=== Regarding the process, for example, in SA4 of Fig. 9, in the process of aggregating the occurrence counts identified in SA1 to SA3, the log type weight information of Fig. 15 may be used. Specifically, the occurrence counts identified in SA1 to SA3 may be multiplied by the corresponding weight information of Fig. 15, and the multiplication results may be aggregated.

[0173] For example, for "AAAA", a doctor in the "cardiology department", as shown in the top row of Figure 16, "13" for "heart failure" (the calculation result of "1 (number of occurrences of search log information) x 3 (weight information for "search" in Figure 15) + 5 (number of occurrences of content viewing log information) x 2 (weight information for content viewing in Figure 15)") is calculated and tallied. Also, "12" for "arrhythmia" ("1 x 3 + 1 x 1 + 4 x 2") is calculated and tallied, and "PCI" (5) (the calculation result of "1 x 1 + 2 x 2") is calculated and tallied.

[0174] 13. Thereafter, in SA5 of FIG. 9, the information in the "number of occurrences x weight information" column of FIG. 16 is used for processing instead of the information in the "number of occurrences" column of FIG.

[0175] By configuring in this manner, the user's log-side score information is generated based on the behavior log information and weight information (second weight information) indicating weights determined for multiple types of behavior log information depending on the type of user's behavior log, thereby making it possible to reflect, for example, the type of behavior log, and therefore to appropriately select recommended content information.

[0176] (Regarding the order of action logs) Furthermore, the log-side score information of the user may be generated taking into consideration the execution order of the action log (i.e., the execution order of each action corresponding to the action log). Fig. 17 is a diagram illustrating an example of execution order weight information.

[0177] The performance order weight information in Fig. 17 is stored in the recording unit 22 in Fig. 1. The "performance order weight information" is information indicating a weight corresponding to the performance order (execution order) of the actions corresponding to the action log information, and for example, the information of each item shown in Fig. 17 is mutually associated.

[0178] The "implementation information" in FIG. 17 is information indicating the order of the actions corresponding to the action log information (in FIG. 15, "1" indicating the first, etc.).

[0179] The "weight information" in FIG. 15 is third weight information indicating a weight determined for the action log information according to the execution order of the user's action logs (that is, the execution order of the actions corresponding to the action logs), and is numerical information.

[0180] In the log type weight information in FIG. 15, for example, the information at the top indicates that the weight of the action log information corresponding to the first execution order is "3".

[0181] ===Processing=== The processing will be described by taking as an example a case where the execution order is specified by the type of action log indicated by the action log information (i.e., the type of action log information). For example, the search log information in Fig. 6, the chat log information in Fig. 7, and the content browsing log information in Fig. 8 store not only the date but also the hour and minute in the date information of each piece of information.

[0182] First, the control unit 23 identifies the execution order of each piece of action log information after executing SA1 to SA3 in Fig. 9. Any method of identification may be used, but for example, the control unit 23 may identify a tendency of the execution order of each piece of action log information by referring to the date information (information indicating the date and time when the action was taken) in Fig. 6 to Fig. 8, and identify the execution order corresponding to the identified tendency as the execution order of the action log information.

[0183] Here, for example, it is thought that a doctor who places importance on doing his or her own research in the course of his or her normal work will often conduct a search, then view the content information, and then chat, in that order. In the behavioral log information of such a doctor, within any given period of time (for example, within one hour), the date information in Figure 6 (time of search) will tend to be earlier than the date information in Figure 8 (time of viewing), and it is also expected that the date information in Figure 8 (time of viewing) will be earlier than the date information in Figure 7 (time of transmission, etc.).

[0184] In such a case, the type corresponding to the search log information (i.e., "search") is identified as the first type to be performed, the type corresponding to the content viewing log information (i.e., "content viewing") is identified as the second type to be performed, and the type corresponding to the chat log information (i.e., "chat") is identified as the third type to be performed. This case will be described as an example.

[0185] First, in SA4 of FIG. 9, the control unit 23 may be configured to multiply the occurrence counts identified in SA1 to SA3 by the corresponding weight information in FIG. 17, and to tally up the multiplication results.

[0186] Here, for example, for "AAAA", a doctor in the "cardiology department", as shown in the top row of Figure 16, "13" for "heart failure" ("1 (number of times search log information appears) x 3 (weight information "3" in Figure 17 corresponding to number 1 in the execution order of the search log information) + 5 (number of times content viewing log information appears) x 2 (weight information "2" in Figure 17 corresponding to number 2 in the execution order of the content viewing log information)) is calculated and tallied. Also, "12" for "arrhythmia" ("1 x 3 + 1 x 1 + 4 x 2") is calculated and tallied, and "PCI" (5 (the calculation result of "1 x 1 + 2 x 2") is calculated and tallied).

[0187] 13. Thereafter, in SA5 of FIG. 9, the information in the "number of occurrences x weight information" column of FIG. 16 is used for processing instead of the information in the "number of occurrences" column of FIG.

[0188] By configuring in this manner, the user's log-side score information is generated based on the weight information (third weight information) indicating weights determined for multiple types of action log information according to the order in which the user's action logs are performed, and the action log information, thereby making it possible to reflect, for example, the order in which the action logs are performed, thereby making it possible to appropriately select recommended content information.

[0189] (About combinations) Furthermore, the processes (including variations) described in the above embodiments and the processes described as modified examples may be combined in any manner, or some of the processes may be omitted.

[0190] (Addendum) The selection system of Appendix 1 is a selection system that selects recommended content information, which is content information to recommend to a user, from a plurality of content information, and includes: a generation means that generates log-side score information for the user corresponding to action log information that indicates the user's action log, based on the action log information; and a selection means that selects the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information, wherein the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information, and the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of content information.

[0191] The selection system of Appendix 2 is the selection system described in Appendix 1, wherein there are a plurality of the predetermined keywords, and the generation means generates the log-side score information of the user based on first weight information indicating the weight assigned to each of the plurality of predetermined keywords and the behavior log information.

[0192] The selection system of Appendix 3 is the selection system described in Appendix 1, in which there are multiple types of user behavior logs, and the generation means generates the user's log-side score information based on the behavior log information and second weight information indicating weights determined for the multiple types of behavior log information depending on the type of the user's behavior log.

[0193] The selection system of Appendix 4 is the selection system described in Appendix 1, wherein the user's behavior log includes multiple types of behavior logs performed sequentially, and the generation means generates the user's log-side score information based on the behavior log information and third weight information indicating weights determined for the multiple types of behavior log information according to the order in which the user's behavior logs are performed.

[0194] The selection program of Appendix 5 is a selection program that selects recommended content information, which is content information to recommend to a user, from a plurality of content information, and causes a computer to function as a generation means that generates log-side score information for the user corresponding to action log information that indicates the user's action log, based on the action log information, and a selection means that selects the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information, wherein the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information, and the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of content information.

[0195] (Effect of supplementary notes) According to the selection system described in Appendix 1 and the selection program described in Appendix 5, by selecting recommended content information based on log-side score information (information generated based on behavior log information) and content-side score information, it is possible to appropriately select recommended content information, for example.

[0196] According to the selection system described in Appendix 2, by generating a user's log-side score information based on first weight information indicating the weight assigned to each of a plurality of specified keywords and behavioral log information, it becomes possible to reflect, for example, the degree of importance of the keywords, thereby making it possible to appropriately select recommended content information.

[0197] According to the selection system described in Appendix 3, by generating the user's log-side score information based on the second weight information indicating weights determined for multiple types of behavior log information depending on the type of user's behavior log and the behavior log information, it is possible to reflect, for example, the type of behavior log, thereby making it possible to appropriately select recommended content information.

[0198] According to the selection system described in Appendix 4, by generating a user's log-side score information based on the third weight information, which indicates weights determined for multiple types of behavior log information according to the order in which the user's behavior logs are performed, and the behavior log information, it is possible to reflect, for example, the order in which the behavior logs are performed, thereby making it possible to appropriately select recommended content information. [Explanation of symbols]

[0199] 1. Terminal equipment 2. Server device 11 Communications Department 12 Input section 13 Output section 14 Recording section 15 Control Unit 21 Communications Department 22 Recording section 23 Control Unit 100 Information Processing Systems

Claims

1. A selection system for selecting recommended content information, which is content information recommended to a user, from a plurality of pieces of content information, a generation means for generating log-side score information of the user corresponding to the action log information based on the action log information indicating the action log of the user; a selection means for selecting the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information, the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information; the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of pieces of content information; Selection system.

2. There are a plurality of the predetermined keywords, the generation means generates log-side score information of the user based on first weight information indicating weights assigned to the plurality of predetermined keywords and the action log information; The selection system of claim 1 .

3. There are multiple types of user action logs, The generation means generates log-side score information of the user based on second weight information indicating weights determined for the plurality of types of action log information according to the type of the action log of the user and the action log information. The selection system of claim 1 .

4. The user's action log includes a plurality of types of action logs that have been sequentially performed, The generation means generates log-side score information of the user based on third weight information indicating weights determined for the plurality of types of action log information according to an execution order of the action log of the user and the action log information. The selection system of claim 1 .

5. A selection program for selecting recommended content information, which is content information recommended to a user, from a plurality of pieces of content information, Computer, a generation means for generating log-side score information of the user corresponding to the action log information based on the action log information indicating the action log of the user; a selection means for selecting the recommended content information based on the log-side score information generated by the generation means and content-side score information corresponding to each of the plurality of content information; the log-side score information is determined according to the number of times a predetermined keyword appears in the action log information; the content-side score information is determined according to the number of times the predetermined keyword appears in each of the plurality of pieces of content information; Selection program.

Images