Information processing device, information processing method, and program

Abstract

An information processing device and an information processing method are provided that can appropriately analyze the characteristics of content that a user likes. [Solution] The information processing device 10 includes a clustering unit 20 that classifies content into multiple clusters based on the viewing histories of multiple users, a cluster profiling unit 21 that determines, from among multiple pieces of accompanying information of content included in each cluster, at least one piece of accompanying information that represents the characteristics of a group of content belonging to each cluster as a label for that cluster, and a preference determination unit 22 that determines, based on the viewing history of a single user, the labels of at least one or more clusters that include content viewed by the single user as labels for content preferred by the single user.

Description

[Technical Field]

[0001] The present invention relates to a technique for determining a user's preferences for content such as anime and dramas. [Background technology]

[0002] Methods for analyzing user preferences for content such as anime and dramas have been known for some time. Patent Document 1 discloses an invention related to a program recommendation device that automatically searches for and recommends programs that match a user's preferences. The invention in Patent Document 1 classifies programs into multiple attributes and assigns evaluation values ​​to multiple attribute values ​​contained in each attribute, thereby analyzing a user's preferences for program content. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-54942 Summary of the Invention [Problem to be solved by the invention]

[0004] Conventional methods such as those described in Patent Document 1 only identify broad categories such as "variety," "foreign drama," and "period drama," making it difficult to understand the characteristics of content that users truly like.

[0005] In view of the above background, an object of the present invention is to provide an information processing device and the like that can appropriately analyze the characteristics of content that a user likes. [Means for solving the problem]

[0006] The information processing device of the present invention includes a clustering unit that classifies content into a plurality of clusters based on the viewing histories of a plurality of users; a cluster profiling unit that determines, from among a plurality of pieces of accompanying information of content included in each cluster, at least one piece of accompanying information that represents the characteristics of a group of content belonging to each cluster as a label for that cluster; and a preference determination unit that, based on the viewing history of a single user, determines the labels of at least one or more clusters that include content viewed by the single user as labels for content preferred by the single user.

[0007] The information processing method of the present invention is a method for analyzing user preferences by an information processing device, comprising: The method includes the steps of: the information processing device classifying content into a plurality of clusters based on the viewing histories of a plurality of users; the information processing device determining, from among a plurality of pieces of accompanying information of content included in each cluster, at least one piece of accompanying information that represents characteristics of a group of content belonging to each cluster as a label for that cluster; and the information processing device determining, based on the viewing history of a single user, labels for at least one or more clusters that include content viewed by the single user as labels for content preferred by the single user. [Effects of the Invention]

[0008] According to the present invention, it is possible to appropriately analyze the characteristics of content that a user prefers. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 illustrates a functional configuration of an information processing apparatus according to an embodiment. [Figure 2] FIG. 1 illustrates a hardware configuration of an information processing apparatus according to an embodiment. [Figure 3] FIG. 10 is a diagram showing an example of part of data stored in a content-associated data storage unit. [Figure 4] FIG. 2 is a diagram showing an example of data stored in a viewing history data storage unit. [Figure 5] FIG. 10 is a diagram illustrating an example of labels assigned by cluster profiling. [Figure 6] FIG. 10 is a diagram illustrating a word vectorization process. [Figure 7] FIG. 10 is a diagram illustrating processing by a cluster profiling unit. [Figure 8] FIG. 1 is a diagram illustrating an operation of an information processing device. DETAILED DESCRIPTION OF THE INVENTION

[0010] The information processing device according to the present embodiment will be described below with reference to the drawings. Note that the following description merely shows an example of a preferred embodiment, and is not intended to limit the scope of the invention as defined in the claims.

[0011] Fig. 1 is a diagram showing the functional configuration of an information processing device 10 according to an embodiment, and Fig. 2 is a diagram showing the hardware configuration of the information processing device 10. The information processing device 10 is connected to a plurality of user terminals 30 so as to be able to communicate with them.

[0012] First, the hardware configuration of the information processing device 10 will be described with reference to FIG. 2. The information processing device 10 is physically configured as a computer including a processor 101, a memory 102, a storage 103, a communication device 104, an input device 105, an output device 106, and a bus connecting these devices. Each of these devices operates using power supplied from a battery (not shown). In the following description, the term "device" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the information processing device 10 may be configured to include one or more of the devices shown in FIG. 2, or may be configured without including some of the devices. Furthermore, the information processing device 10 may be configured by communicating with multiple devices each having a different housing.

[0013] Each function of the information processing device 10 is realized by loading predetermined software (programs) onto hardware such as the processor 101, memory 102, etc., so that the processor 101 performs calculations, controls communication via the communication device 104, and controls at least one of reading and writing of data in the memory 102 and storage 103.

[0014] The processor 101 controls the entire computer by running, for example, an operating system. The processor 101 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. Furthermore, for example, a baseband signal processing unit, a call processing unit, etc. may be realized by the processor 101.

[0015] The processor 101 reads programs (program codes), software modules, data, etc. from at least one of the storage 103 and the communication device 104 into the memory 102, and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described below. The functional blocks of the information processing device 10 may be implemented by a control program stored in the memory 102 and running on the processor 101. Various processes may be executed by one processor 101, or may be executed simultaneously or sequentially by two or more processors 101. The processor 101 may be implemented by one or more chips. The programs may be transmitted to the information processing device 10 via a telecommunications line.

[0016] The memory 102 is a computer-readable recording medium and may be configured by, for example, at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 102 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 102 can store executable programs (program codes), software modules, etc. for implementing the method according to this embodiment.

[0017] Storage 103 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray® disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy disk, a magnetic strip, etc. Storage 103 may also be referred to as an auxiliary storage device.

[0018] The communication device 104 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module. Each device, such as the processor 101 and the memory 102, is connected by a bus for communicating information. The bus may be configured using a single bus, or different buses may be used between each device.

[0019] The information processing device 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 101 may be implemented using at least one of these pieces of hardware.

[0020] Next, the functional configuration of the information processing device 10 will be described. The information processing device 10 has a function of distributing content such as anime and dramas, and a function of analyzing user preferences based on the user's viewing history. The information processing device 10 of this embodiment has both a function of distributing content and a function of analyzing user preferences, but a content distribution server that distributes content may be separately arranged, and the content may be distributed from the content distribution server to the user terminal 30. In this case, the information processing device 10 acquires the user's viewing history from the content distribution server and analyzes the user's preferences. The user terminal 30 has a function of receiving and playing content. The user terminal 30 is, for example, a smartphone, a tablet terminal, a PC, etc.

[0021] The information processing device 10 has a calculation unit 11, a content data storage unit 12, a content associated data storage unit 13, a viewing history data storage unit 14, and a communication unit 15. The communication unit 15 is configured to communicate with the user terminal 30. The content data storage unit 12 stores data on content such as anime and dramas to be distributed. If a content distribution server is separately provided, the content distribution server will have the content data storage unit 12. The content associated data stores various data associated with the content.

[0022] The following description will be given using an example of content, anime. The content-associated data includes, for example, data on voice actors, original publisher, anime production company, original author, and anime production year. The content-associated data does not necessarily have to include all of the data listed here, and may include some of the data listed here. Furthermore, content-associated data other than the data listed here may also be included.

[0023] 3 is a diagram showing an example of part of the data stored in the content-associated data storage unit 13. In this embodiment, the content-associated data storage unit 13 stores data such as voice actors, original publishers, etc. in association with the titles of works that identify the content.

[0024] FIG. 4 is a diagram showing an example of data stored in the viewing history data storage unit 14. The viewing history data storage unit 14 stores the titles of works viewed by a user, their episode numbers, viewing dates, and viewing times, in association with a user ID that identifies the user. In this example, the titles of works are used as data for identifying works, but work IDs may also be used. The episode numbers are data for identifying episodes viewed by a user when a work consists of multiple episodes. The viewing dates are data indicating the dates on which the work was viewed. The viewing times are data indicating the duration of time the work was viewed. In some cases, a user may stop watching midway through. In such cases, the elapsed time up to the time the user stopped watching is stored as viewing time data. As can be seen from the viewing date data, if a user watches the same episode of the same work multiple times, viewing history data is stored each time. By knowing the number of times the same episode has been viewed, it can be determined that a user has a strong preference for that episode or that work if the user has watched the same episode multiple times.

[0025] Next, a description will be given of the configuration of the calculation unit 11. The calculation unit 11 includes a clustering unit 20, a cluster profiling unit 21, a preference determination unit 22, a preference information presentation unit 23, and a content information presentation unit 24.

[0026] The clustering unit 20 has a function of clustering content based on the viewing histories of multiple users. As an example, the clustering unit 20 clusters content using item-based clustering technology. The clustering unit 20 creates a matrix indicating which user has viewed each piece of content based on the viewing history data, and generates a vector representing the viewing pattern of each piece of content. The clustering unit 20 clusters content using the vector. As a clustering algorithm, k-means, DBSCAN, or hierarchical clustering can be used.

[0027] The cluster profiling unit 21 has a function of analyzing the clusters generated by the clustering unit 20 and assigning a label that characterizes each cluster to each cluster.

[0028] FIG. 5 is a diagram showing examples of labels assigned by cluster profiling. In FIG. 5, contents AAA to HHH are classified into three clusters C1 to C3. The cluster profiling unit 21 assigns labels that characterize each of the three clusters C1 to C3. For example, the labels "Voice actor aaa" and "X publisher" are assigned to cluster C1. The labels are selected from the accompanying data of the contents. Characteristic content accompanying data that is common to multiple contents included in a cluster becomes the label.

[0029] 6 and 7 are diagrams for explaining the processing of the cluster profiling unit 21. As shown in FIG. 6, the cluster profiling unit 21 converts content-associated data into word vectors. That is, the cluster profiling unit 21 converts the content-associated data, which is text data, into numerical data. In the example shown in FIG. 6, the name of voice actor aaa is converted to 0.12, and the name of voice actor bbb is converted to 0.56. As a word vectorization method, for example, Bag of Words (BoW) can be used. BoW is a method of counting the number of times a word appears and vectorizing it. In this embodiment, the cluster profiling unit 21 may perform vectorization based on the frequency of appearance in the data stored in the content-associated data storage unit 13.

[0030] Next, the cluster profiling unit 21 creates a co-occurrence network for each cluster. Specifically, feature quantities of content are extracted from the cluster to be processed. The extracted feature quantities are content-associated data of the content within the cluster, and in this embodiment, they are data on the voice actors, original publisher, animation production company, original author, and animation production year of the anime content. Note that these feature quantities are data vectorized by the above-mentioned word vectorization.

[0031] The cluster profiling unit 21 calculates the Jaccard coefficient between the content-associated data, which is a feature, and generates a co-occurrence network of the content-associated data. For example, the Jaccard coefficient of content-associated data A and content-associated data B can be calculated by dividing the set (co-occurrence frequency) of contents having both content-associated data A and B by the set (union) of contents having either content-associated data A or B.

[0032] Figure 7 shows an example of the Jaccard coefficients between content-related data and a co-occurrence network generated based on the Jaccard coefficients. The Jaccard coefficients shown in the upper part of Figure 7 indicate the Jaccard coefficients between content-related data for content within a cluster in each matrix. For example, the Jaccard coefficient between voice actor aaa and voice actor bbb is 0.7, and the Jaccard coefficient between voice actor aaa and publisher X is 0.9.

[0033] The co-occurrence network shown in the lower part of Figure 7 has content-associated data as nodes, and the content-associated data are connected to each other by edges. The thickness of the edge represents the Jaccard coefficient. In the example shown in Figure 7, the Jaccard coefficient for voice actor aaa and publisher X is 0.9, and the edge in the co-occurrence network is the thickest. This cluster has the highest co-occurrence between voice actor aaa and publisher X, and is characterized by voice actor aaa and publisher X. Therefore, the cluster profiling unit 21 labels this cluster with voice actor aaa and publisher X. Through the above processing, the cluster profiling unit 21 assigns labels that characterize each cluster to the clusters classified by the clustering unit 20.

[0034] In the example shown in FIG. 7, the Jaccard coefficient of voice actor bbb and publisher X is 0.8, indicating a high degree of co-occurrence. Content-associated data with a Jaccard coefficient equal to or greater than a predetermined threshold may also be used as a label. In this case, labels may be prioritized according to the degree of co-occurrence. In this example, publisher X and voice actor aaa may be the first label, and voice actor bbb may be the second label. Also, although an example of generating a co-occurrence network using the Jaccard coefficient has been given as the processing of cluster profiling unit 21, another method may also be used for the processing of cluster profiling unit 21.

[0035] In the above, an example was given in which content-associated data that serves as a label is obtained based on the thickness of the edges connecting nodes, but it is also possible to detect groups of closely related nodes in a co-occurrence network (called "communities") and obtain the nodes that make up the community as labels that characterize the cluster. As a method for detecting communities, for example, known methods such as the Louvain method can be used.

[0036] The preference determination unit 22 determines the user's preference based on the labels assigned to each cluster by the cluster profiling unit 21 and the user's viewing history data. The preference determination unit 22 reads the viewing history data of the user to be determined from the viewing history data storage unit 14 and identifies the cluster to which the content viewed by the user belongs. For example, in the example shown in FIG. 5, if the user viewed "AAA," "BBB," and "CCC," the cluster of the content viewed by the user is cluster C1. The preference determination unit 22 determines, as the user's preference information, what label has been assigned to the cluster of the content viewed by the user. In the above example, the preference information of the user who viewed cluster C1 is "voice actor aaa" and "X publisher."

[0037] If the content viewed by a user belongs to multiple clusters, weighting may be applied to the clusters based on the user's viewing history, and the label of the user's preferred content may be determined based on the weighting. For example, if a user viewed "AAA," "BBB," "CCC," and "EEE," the content viewed by the user spans clusters C1 and C2. However, since the user viewed three contents belonging to cluster C1 and only one content belonging to cluster C2, it is considered that the user has a stronger preference for cluster C1 than for cluster C2. A higher weight is applied to the label of cluster C1 than to the label of cluster C2. That is, the preference determination unit 22 first determines "voice actor aaa" and "publisher XXX" as labels indicating the user's preferences, followed by "Publisher Z" and "Production Company α" as labels indicating the user's preferences. If the number of contents belonging to cluster C1 and cluster C2 is significantly different—for example, if the user viewed nine contents belonging to cluster C1 and one content belonging to cluster C2—only the label of cluster C1 may be determined as the user's preferences.

[0038] Although the example given here uses the number of contents viewed by a user to weight clusters, parameters such as the total viewing time of the contents or the number of times the same content has been viewed repeatedly may also be used. If a user repeatedly views the same content, it is considered that the user particularly likes that content, so the weight of a cluster that includes such content may be increased.

[0039] The preference information presentation unit 23 has a function of presenting to the user a label indicating the user's preference determined by the preference determination unit 22. Specifically, the preference information presentation unit 23 transmits data of the label indicating the user's preference to the user terminal 30, and presents the information to the user by, for example, displaying the information on the user terminal 30. A user who is presented with the label can discover preferences that he or she was not even aware of. Conventionally, a user may have only recognized that his or her preference was comedy or teenage stories, but according to this embodiment, he or she can discover that he or she actually likes works by voice actor aaa and publisher X.

[0040] The content information presentation unit 24 presents content information based on labels indicating the user's preferences determined by the preference determination unit 22. That is, the content information presentation unit 24 recommends content that matches the user's preferences. For example, if the user's preferences are "voice actor aaa" and "publisher XXX," the content information presentation unit 24 extracts content having content associated data for "voice actor aaa" and "publisher XXX," and transmits information about the extracted content to the user terminal 30. The user terminal 30 displays the content information received from the information processing device 10 and presents it to the user.

[0041] When transmitting content information, the content information presenting unit 24 may also transmit label data as a reason for recommending the content. By configuring content to be recommended based on labels in this way, it is possible to present and appropriately recommend content that users prefer. Since the content information presenting unit 24 extracts content based on content-associated data, unlike methods such as collaborative filtering, content that is not viewed by many users can also be subject to information presentation.

[0042] As mentioned above, there are cases where a user's viewing history spans multiple clusters and multiple labels are assigned based on the multiple clusters. In such cases, it is possible to present content information that precisely reflects the user's preferences.

[0043] FIG. 8 is a flowchart showing the operation of the information processing device 10. The information processing device 10 first performs clustering of content based on user viewing history data (S10). The viewing history data used here is viewing history data of multiple users, and viewing history data of all users may be used. The information processing device 10 creates a matrix indicating which user has viewed each piece of content based on the viewing history data, and generates a vector representing the viewing pattern of each piece of content. The information processing device 10 performs clustering based on the similarity of the vectors representing the viewing patterns. As a result, content classified into the same cluster will have some common characteristics.

[0044] The information processing device 10 profiles each of the generated clusters (S11). That is, commonalities between the contents included in the same cluster are analyzed. The cluster profiling method executed by the information processing device 10 of this embodiment is as described with reference to FIGS. 5 to 7, and a label that characterizes the cluster is assigned to each cluster through cluster profiling. The processing up to this point, in which contents are clustered and labeled based on the viewing history data of multiple users, is a preparatory stage for analyzing the preferences of individual users, and does not need to be performed every time, and may be performed in advance. For example, the clusters and labels may be updated periodically.

[0045] Next, the information processing device 10 analyzes the preferences of individual users. Based on the viewing history data of a user to be analyzed, content viewed by that user within a predetermined period is extracted (S12). The predetermined period is a relatively long span, such as one year, three years, or five years. The reason for using viewing history data within the predetermined period in this way is that user preferences are likely to change over time, and viewing history data from the distant past is omitted to appropriately analyze user preferences. Here, the viewing history data used for analyzing preferences is limited to within the predetermined period, but such a period limitation may also be omitted.

[0046] Next, the information processing device 10 identifies a cluster that includes the extracted content (S13). Using Fig. 5 as an example, if the user is viewing works AAA and BBB, cluster C1 is identified, and if the user is viewing works DDD and EEE, cluster C2 is identified. If the user is viewing works AAA, BBB, and DDD, cluster C1 and cluster C2 are identified.

[0047] The information processing device 10 determines whether the number of identified clusters is one (S14). If one cluster is identified (YES in S14), the information processing device 10 determines the label of the identified cluster as the label indicating the user's preference (S15). If more than one cluster is identified (NO in S14), the information processing device 10 weights the clusters based on the user's viewing history data and determines the label indicating the user's preference (S16). A larger weight is set for a cluster that includes more viewing content, and the label of the cluster with the larger weight is preferentially used to determine the user's preference label.

[0048] The information processing device 10 transmits label information indicating the user's preferences to the user terminal 30, and causes the user terminal 30 to present the information (S17). This allows the user to discover his or her preferences. The information processing device 10 also recommends content to the user using the user preference label data (S18). In this embodiment, the labels are content-associated data (e.g., voice actor, original publisher, animation production company, original author, or animation production era), and the information processing device 10 searches for content having content-associated data that is the same as the label indicating the user's preferences, and transmits information about the searched content to the user terminal 30, and causes the content to be presented on the user terminal 30.

[0049] The configuration of the information processing device of this embodiment has been described above, and an example of the hardware of the information processing device is a computer equipped with a CPU, RAM, ROM, hard disk, display, keyboard, mouse, communication interface, etc. The information processing device is realized by storing a program having modules that realize each of the above functions in RAM or ROM and executing the program by the CPU. Such programs are also included within the scope of the present invention.

[0050] The following additional notes are provided regarding the above-described embodiment. [Appendix 1] The information processing device of Supplementary Note 1 includes a clustering unit that classifies content into a plurality of clusters based on the viewing histories of a plurality of users; a cluster profiling unit that determines, from among a plurality of pieces of accompanying information of content included in each cluster, at least one piece of accompanying information that represents characteristics of a group of content belonging to the cluster as a label for that cluster; and a preference determination unit that, based on the viewing history of a single user, determines labels of at least one or more clusters that include content viewed by the single user as labels for content preferred by the single user.

[0051] With this configuration, it is possible to appropriately analyze the characteristics of content that a user prefers based on the label of a cluster that includes content that the user has viewed.

[0052] [Appendix 2] In the information processing device of Supplementary Note 1, the content may be an anime, and the accompanying information may include at least information relating to any one of voice actors, original publisher, anime production company, original author, and anime production year.

[0053] This configuration allows for analysis of user preferences for anime content from a perspective that differs from conventional broad categories such as genre.

[0054] [Appendix 3] The information processing device of Supplementary Note 1 or 2 may include a preference information presentation unit that presents information on labels of content that is preferred by the one user to the one user.

[0055] With this configuration, it is possible to make the user aware of his or her preferences by presenting the preference information to the user.

[0056] [Appendix 4] The information processing device of any one of Supplementary Notes 1 to 3 may include a content information presenting unit that presents to the one user content extracted based on a label of content that the one user likes.

[0057] This configuration makes it possible to present content that matches the user's preferences.

[0058] [Appendix 5] In the information processing device according to any one of Supplementary Notes 1 to 4, the cluster profiling may obtain, as the label, associated information that characterizes the cluster based on a co-occurrence network generated based on the associated information.

[0059] This configuration makes it possible to identify content-associated information that is common to the contents included in the cluster.

[0060] [Appendix 6] In any of the information processing devices of Supplementary Notes 1 to 5, when the content viewed by the one user belongs to multiple clusters, the preference determination unit may determine a label for the content preferred by the one user based on weighting of the multiple clusters determined based on the viewing history of the one user.

[0061] This configuration allows for more detailed analysis of user preferences.

[0062] [Appendix 7] An information processing method according to a seventh aspect is a method for analyzing user preferences by an information processing device, The method includes the steps of: the information processing device classifying content into a plurality of clusters based on the viewing histories of a plurality of users; the information processing device determining, from among a plurality of pieces of accompanying information of content included in each cluster, at least one piece of accompanying information that represents characteristics of a group of content belonging to each cluster as a label for that cluster; and the information processing device determining, based on the viewing history of a single user, labels for at least one or more clusters that include content viewed by the single user as labels for content preferred by the single user.

[0063] With this configuration, it is possible to appropriately analyze the characteristics of content that a user prefers based on the label of a cluster that includes content that the user has viewed. [Explanation of symbols]

[0064] 10. Information processing equipment 11 Arithmetic section 12 Content data storage unit 13 Content accompanying data storage unit 14 Viewing history data storage unit 15 Communications Department 20 Clustering Department 21 Cluster Profiling Department 22 Preference determination unit 23 Preference information presentation section 24 Content information presentation unit 30 User terminals

Claims

1. a clustering unit that classifies content into a plurality of clusters based on the viewing histories of a plurality of users; a cluster profiling unit that determines, from among a plurality of pieces of associated information of the contents included in each cluster, at least one piece of associated information that represents a feature of a group of contents belonging to each cluster as a label of the cluster; a preference determination unit that determines, based on a viewing history of a user, labels of at least one cluster including content viewed by the user as labels of content preferred by the user; An information processing device comprising:

2. the content is an anime, The information processing device according to claim 1 , wherein the accompanying information includes at least information regarding a voice actor, an original publisher, an animation production company, an original author, and an animation production year.

3. The information processing device according to claim 1 , further comprising a preference information presentation unit that presents information on labels of content that the one user likes to the one user.

4. The information processing device according to claim 1 , further comprising a content information presenting unit that presents to the one user content extracted based on a label of content that the one user likes.

5. The information processing apparatus according to claim 1 , wherein the cluster profiling unit obtains, as the label, associated information that characterizes the cluster based on a co-occurrence network generated based on the associated information.

6. The information processing device according to claim 1, wherein, when the content viewed by the user belongs to multiple clusters, the preference determination unit determines a label for the content preferred by the user based on weighting of the multiple clusters determined based on the viewing history of the user.

7. A method for analyzing user preferences by an information processing device, comprising: classifying content into a plurality of clusters based on viewing histories of a plurality of users by the information processing device; a step by the information processing device determining, from among a plurality of pieces of associated information of the contents included in each of the clusters, at least one piece of associated information that represents a feature of a group of contents belonging to each of the clusters as a label of the cluster; The information processing device determines, based on a viewing history of a user, labels of at least one or more clusters including content viewed by the user as labels of content preferred by the user; An information processing method comprising:

Images