Information processing device, information processing method, and program

Abstract

Information can be effectively utilized within the voting platform system. [Solution] The information processing device 100 includes a voting reception unit 13 that accepts votes from users using points for a voting target having multiple candidates, a reward unit 14 that rewards points to users who have voted for the selected candidate for the voting target, and an analysis unit 16 that analyzes the relationship between user information or voting target information and user voting results using a trained model.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] Patent Document 1 discloses a network voting system that conducts voting via a network, including a voting system that creates competition data related to a game and outputs the created competition data, a communication terminal capable of executing a game application, and a virtual voting execution means that is connected to the communication terminal via communication means and receives voting data from the communication terminal when the communication terminal is executing the game application, and outputs the received voting data to the voting system. In this system, information related to a game or competition, such as information on the game schedule, entry list, odds, payouts, popularity, venue, etc., is provided, votes from users are received, the votes are reflected in the odds in real time, and the winning / losing results and payouts of the game are provided.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a voting system as disclosed in Patent Document 1, although information on which target a user voted for is used for reflecting odds and the like, there is room for improvement in effectively utilizing such information.

[0005] The present invention has been made in view of the above problems, and an object thereof is to provide an information processing apparatus, an information processing method, and a program capable of effectively utilizing information in a voting platform system.

Means for Solving the Problems

[0006] According to one aspect of the present invention, the information processing device includes: a reception unit that accepts votes from users using points for a voting target having multiple candidates; a reward unit that rewards points to users who have voted for the selected candidate for the voting target; and an analysis unit that analyzes the relationship between user information or voting target information and user voting results using a trained model.

[0007] According to one aspect of the present invention, an information processing method includes the steps of: accepting votes from users using points for a voting target having multiple candidates; returning points to users who voted for the selected candidate for the voting target; storing information about the user, information about the voting target, and the voting results by the user; and analyzing the relationship between the stored information about the user or the voting target and the voting results using a trained model.

[0008] According to one aspect of the present invention, a computer-executable program causes the computer to perform the following steps: accept votes from users using points for a voting target having multiple candidates; reward points to users who voted for the selected candidate for the voting target; store information about the user, information about the voting target, and the voting results by the user; and analyze the relationship between the stored information about the user or the voting target and the voting results using a trained model. [Effects of the Invention]

[0009] According to the present invention, information can be effectively utilized in a voting platform system. [Brief explanation of the drawing]

[0010] [Figure 1] This is a schematic diagram showing the configuration of an information processing system according to an embodiment of the present invention. [Figure 2] This is a block diagram showing the configuration of an information processing device according to an embodiment of the present invention. [Figure 3] This figure shows an example of a user information database stored in the storage unit of an information processing device according to an embodiment of the present invention. [Figure 4] This figure shows an example of a voting target information database stored in the storage unit of an information processing device according to an embodiment of the present invention. [Figure 5] This figure shows an example of a voting results information database stored in the storage unit of an information processing device according to an embodiment of the present invention. [Figure 6] This figure shows an example of a point information database stored in the storage unit of an information processing device according to an embodiment of the present invention. [Figure 7] This is a flowchart showing the voting process using the information processing device according to an embodiment of the present invention. [Figure 8] This is a flowchart showing the process of user trend analysis using an information processing device according to an embodiment of the present invention. [Figure 9] This is a flowchart showing the process of personal trend analysis using an information processing device according to an embodiment of the present invention. [Figure 10] This is a flowchart showing the process of heat scoring analysis by an information processing device according to an embodiment of the present invention. [Figure 11] This flowchart shows the process of high-value information analysis using an information processing device according to an embodiment of the present invention. [Figure 12] This flowchart shows the processing of target group behavior analysis by an information processing device according to an embodiment of the present invention. [Figure 13] This flowchart shows the process of insight fit analysis by an information processing device according to an embodiment of the present invention. [Modes for carrying out the invention]

[0011] Hereinafter, an information processing apparatus 100, an information processing method, a program, and an information processing system 1000 according to an embodiment of the present invention will be described with reference to the drawings.

[0012] First, the configuration of the information processing system 1000 will be described with reference to FIGS. 1 to 6. FIG. 1 is a schematic diagram showing the configuration of the information processing system 1000. FIG. 2 is a block diagram showing the configuration of the information processing apparatus 100. FIGS. 3 to 6 are diagrams showing an example of a database stored in the storage unit 19 of the information processing apparatus 100.

[0013] The information processing system 1000 is a system for receiving votes from users for a voting target having a plurality of candidates and providing value (points) to users who voted for the selected candidate. In the information processing system 1000, a user can receive voting and value provision through the user terminal 110 from a voting platform provided by the information processing apparatus 100.

[0014] As shown in FIG. 1, the information processing system 1000 includes an information processing apparatus 100 and a user terminal 110.

[0015] The information processing apparatus 100 is composed of a computer including a CPU 101 (central processing unit), a storage device 102 composed of a ROM (read only memory), a RAM (random access memory), etc., and an I / O interface 103 (input / output interface), a communication device, etc. The RAM stores data in the processing of the CPU 101, the ROM stores a control program of the CPU 101 in advance, and the I / O interface 103 is used for input / output of information with connected devices. By the CPU 101 executing the control program, various processes of the information processing apparatus 100 described in this specification are executed. Note that the information processing apparatus 100 may be configured as one device, or may be divided into a plurality of devices and configured such that each control is distributedly processed by the plurality of devices.

[0016] In the information processing apparatus 100, in response to operation inputs from operators such as the business operator (system administrator) of the voting platform input through the input device 104, display of information by the display device 105, execution of processing by the information processing apparatus 100, input / output of data to / from the information processing apparatus 100, etc. are performed. The information processing apparatus 100 is configured to be communicable with the user terminal 110 through the network NW.

[0017] The information processing apparatus 100 has a function as a server that provides the voting platform to users through the Internet. The information processing apparatus 100 causes the user terminal 110 to display a web page that serves as a graphical user interface (GUI) of the voting platform in response to a request from a user through the user terminal 110. Note that the GUI is not limited to the web page transmitted from the information processing apparatus 100, and may be an image saved and displayed on the user terminal 110 (for example, an image displayed by an application installed on the user terminal 110), etc.

[0018] As shown in FIG. 2, the information processing apparatus 100 includes a user management unit 10, a vote management unit 11, a point management unit 12, a vote reception unit 13, a reduction unit 14, a recording unit 15, an analysis unit 16, an information providing unit 17, a display control unit 18, and a storage unit 19. Note that each component of the information processing apparatus 100 shown in FIG. 2 shows each function of the information processing apparatus 100 as a virtual unit (functional unit), and does not necessarily mean that it physically exists.

[0019] The user management unit 10 manages information regarding users who use the voting platform. Specifically, the user management unit 10 performs processes such as new user registration processing, authentication processing at the time of login, and update / deletion processing of user information. The user information managed by the user management unit 10 includes a user ID for identifying the user, authentication information (for example, a hashed password), basic attributes (for example, age, gender, family composition, residential area, occupation, etc.), and membership rank, title, etc.

[0020] Membership rank is an indicator that changes according to the user's system usage, and is represented, for example, as Platinum, Gold, Silver, Bronze, etc. Membership rank can be set based on indicators such as "usage," which is an indicator based on the period elapsed since user registration or the active period, which is whether the user has been continuously logged in within a specified period; "activity," which is an indicator based on the number of votes and / or logins per unit period; and "accuracy," which is an indicator based on the correct answer rate in past votes.

[0021] Titles are indicators (statuses) awarded based on a user's voting behavior. Titles can be awarded, for example, based on the themes a user votes on and their accuracy. Specifically, users whose accuracy in a particular category (e.g., sports) exceeds a predetermined threshold can be awarded titles such as "Sports Master." Furthermore, titles are managed as tokens on the blockchain, as described below. Specifically, titles are issued as SBTs (Soul Bound Tokens) linked to the user's account and are prohibited from being transferred to other users. This ensures that the title is guaranteed and managed as a title for that user.

[0022] The voting management unit 11 manages the questions and events that are the subject of voting (hereinafter referred to as "voting subjects"). Voting subjects include any event from which one or more options can be selected, such as predictions of the outcome of future events (e.g., election results, sports wins and losses, fluctuations in economic indicators), and expressions of opinion on specific themes (e.g., selection of new product designs, support or opposition to social issues).

[0023] Specifically, the voting management unit 11 sets and manages voting information such as the voting theme, question content, multiple candidates (options), and voting period (start date and end date and time). Voting information is entered, for example, by a business operator (or system administrator) providing the voting platform through the input device 104. Voting information may also be entered into the information processing device 100 from an external information processing terminal, or it may be automatically generated based on a pre-stored program.

[0024] The voting management unit 11 also assigns category information, topicality scores, and insight tags to the voting targets. Category information indicates the classification of the content of the voting target, such as "politics," "economics," "entertainment," and "sports." The topicality score is an indicator of the level of social attention it has received. Insight tags represent characteristics, such as "price-conscious," "environmentally conscious," and "trend-oriented." The category information, topicality scores, and insight tags may be determined by the system administrator (manual determination) or by the voting management unit 11 (automatic determination by the system). When the voting management unit 11 determines the category information, topicality scores, and insight tags, it can, for example, use a trained model that has learned the relationship between the category information, topicality scores, and insight tags and the voting targets to determine this information. After the voting period ends, the voting management unit 11 registers the candidates selected based on the actual results.

[0025] The point management unit 12 manages the points owned by the user. In this embodiment, there are two types of points: a first type of point that can be used for voting (hereinafter also referred to as "Point A") and a second type of point that cannot be used for voting and can be exchanged for gift certificates, etc. (hereinafter also referred to as "Point B"). In other words, Point A is a point that does not have monetary value, and Point B is a point that has monetary value. Furthermore, Point B is not limited to simply having monetary value; it may also be a point that is managed as a crypto asset that is directly traded, processed, or recorded on the blockchain and has monetary value that can be grasped by the blockchain (value traded as a crypto asset). In addition, Point B may be in the form of other forms of media or rewards (e.g., tickets, etc.) that have monetary value and can be used on the voting platform provided by this system, rather than in the form of points.

[0026] The point management unit 12 manages the balance of these two types of points for each user. Specifically, the point management unit 12 awards points A to users in response to reasons such as login bonuses, viewing advertisements displayed on the voting platform, or completing a predetermined task (for example, playing games on the voting platform). Points A may also be awarded to users by transferring points used outside the voting platform provided by this system (in other words, existing points on other platforms or services managed on servers outside the system). The point management unit 12 also awards points B to users based on instructions from the redemption unit 14, which will be described later. Furthermore, the point management unit 12 deducts points B held by users in response to requests from users to exchange them for gift certificates, etc.

[0027] The voting reception unit 13 receives votes from the user terminal 110 for voting targets that have multiple candidates. Voting is performed using points A held by the user. The voting reception unit 13 receives voting information from the user terminal 110, including information identifying the user, information identifying the voting target, information indicating the candidate selected by the user, and the number of points A used for voting. The voting reception unit 13 also instructs the point management unit 12 to subtract the number of points consumed by the vote from the user's point A balance.

[0028] The reward unit 14 rewards users who voted for the selected candidates for the voting target (i.e., users whose predictions were correct) with points B. The number of points to be rewarded is calculated, for example, based on the odds (reward multiplier) set for each voting target and the number of points A used by the user for voting. The odds may be determined based on the distribution of the number of votes and / or voting points for each candidate, or they may be set as a fixed value in advance. The reward unit 14 notifies the point management unit 12 of the calculated number of reward points and adds it to the point balance of the target user's points B.

[0029] The recording unit 15 associates user information with user voting results and records them on a blockchain (distributed ledger). This ensures the transparency, immutability, and difficulty of tampering with voting data. The recording unit 15 writes voting data (e.g., voting date and time, anonymized user identifier, voting target identifier, selected candidate, number of points consumed, etc.) stored in the storage unit 19 to the blockchain at predetermined times (e.g., once a day or at each voting deadline). Various recording methods can be employed, such as storing the voting data itself in a smart contract, or storing the voting data in a distributed storage system such as IPFS (InterPlanetary File System) and recording the hash value of that data on the blockchain. The recorded data can be verified by third parties via a blockchain explorer, etc.

[0030] The recording unit 15 manages point B as a token on the blockchain. On the other hand, point A is not managed as a token on the blockchain, but is managed on the point management DB of the memory unit 19, which will be described later. In other words, point A is centrally managed in the information processing device 100, while point B is distributedly managed on the blockchain. By managing only point B, which has monetary value, using the blockchain in this way, it is possible to prevent fraud and ensure transparency regarding the acquisition and consumption of point B.

[0031] The analysis unit 16 analyzes the relationship between user information or voting target information and user voting results using a trained model. Voting results represent details of the vote, such as which user voted, at what time, for which voting target, how many points were spent, and which candidates were selected. The analysis unit 16 obtains the data to be analyzed from various databases of the memory unit 19 (user information DB, voting target information DB, voting result information DB, point management DB, etc.), which will be described later, and performs various analyses using the trained model. From another perspective, the analysis unit 16 performs various analyses based on highly transparent, tamper-proof information recorded on the blockchain by the recording unit 15. The analyses performed by the analysis unit 16 will be described in detail later.

[0032] Furthermore, the analysis unit 16 can utilize statistical methods (rule-based, algorithmic analysis) in addition to analysis using pre-trained models. For example, simple aggregation of vote counts and calculation of distributions based on attributes such as age and region can be adequately analyzed using statistical methods. On the other hand, analysis of comments including natural language, trend prediction by linking with external data, and high-dimensional trend analysis combining multiple user attributes and behavioral indicators can be achieved with high accuracy through analysis using pre-trained models (artificial intelligence, AI). By using analysis using pre-trained models and analysis using statistical methods in combination as needed, both flexibility and efficiency in analysis can be achieved.

[0033] A pre-trained model is constructed in advance using machine learning or deep learning techniques based on a large amount of data accumulated in the past. The type of model varies depending on the purpose of analysis and can be a model that utilizes supervised learning, unsupervised learning, reinforcement learning, etc. The pre-trained model is configured to perform tasks such as regression or classification, clustering, time series analysis, and natural language processing, depending on the purpose of analysis. In other words, the analysis unit 16 is not limited to one pre-trained model, but can use multiple models. The pre-trained model may be retrained periodically or whenever new data is accumulated. Since known configurations can be used for the pre-trained model, a detailed explanation is omitted.

[0034] The regression model can be used, for example, to estimate a user's "voting enthusiasm score," which will be described later, by taking user behavior history (number of votes, number of points used for voting, activity time, whether or not it was shared on social media, etc.) as input.

[0035] The classification model can be used, for example, to classify users into categories such as "insight tags" based on their attribute information and behavioral history, according to their voting tendencies or preference patterns.

[0036] Clustering models can be used, for example, to "group" users in a non-hierarchical or hierarchical manner based on similarities such as individual users' voting preference patterns or voting timing.

[0037] Time series models can be used to detect trends (peaks or sharp declines) by analyzing changes over time, such as the number of votes or points consumed for each voting target.

[0038] Natural language processing models can be used, for example, to take natural language data such as comments made during voting or posts made on social networking services (SNS) as input and extract tags (insight tags) that indicate specific opinion trends, emotions, or topics.

[0039] The trained model may be stored in the information processing device 100, or it may be stored in a computer separate from the information processing device 100 (for example, an external server device). The analysis unit 16 retrieves the trained model stored in the internal or external server device of the information processing device 100 and performs the various analyses described herein.

[0040] The information provision unit 17 converts the analysis results from the analysis unit 16 into a predetermined format suitable for output and provides (outputs) them to an external device, such as a user terminal 110. The output format can be any format and may include, for example, a real-time display in a dashboard format on a web page, a standardized report in PDF format, an image file containing graphs and tables, a data file in CSV (Comma Separated Values) format, or an API (Application Programming Interface) for linking with other systems.

[0041] The display control unit 18 controls the display mode (display content) of the GUI displayed on the user terminal 110. The GUI includes, for example, a list of voting targets, a voting screen, a point balance confirmation screen, a screen for viewing past voting history, and a display screen (dashboard) of analysis results provided by the analysis unit 16 of the information processing device 100. The user performs various operations such as voting via the GUI displayed on the user terminal 110.

[0042] The storage unit 19 is configured, for example, as a database, and stores in advance the information necessary for the processing performed by the information processing device 100. Specifically, the storage unit 19 has a database (DB) consisting of a user information DB, a voting target information DB, a voting result information DB, and a points information DB.

[0043] The user information database stores information about users, as shown in Figure 3. This user information includes user attribute information and behavioral pattern information. Attribute information is stored in the attribute information table. Attribute information is static information obtained during user registration, such as name (account name), email address, age, gender, family structure, residential area, occupation, and interest categories. Interest categories are the categories of voting topics that the user is interested in.

[0044] Behavioral pattern information is stored in the behavioral pattern information table. Behavioral pattern information includes dynamic historical information associated with the use of this platform, such as member rank (including time-series information), usage level, activity level, voting history, comment posting on the voting platform, whether or not it was shared on social media, insight tags, titles, etc.

[0045] The behavior pattern table stores voting history IDs corresponding to users, and details of the voting history are managed in the voting history table by associating the voting history ID with the user ID. The voting history includes information such as the voted target, choice, points consumed, win / loss, and date and time. Comment posting and SNS sharing are represented, for example, by the number of times a comment or share was made about the voted target and the content of those comments or shares. Details of comment posting and SNS sharing may be managed in a table separate from the behavior pattern information table. Insight tags are assigned by the voting management unit 11 based on the voting history.

[0046] The voting target information database stores information about the voting targets, as shown in Figure 4. This information includes, for example, the question content, multiple candidates (options) to vote on, the voting target category (similar to the interest category), the voting period, odds, a popularity score representing the topicality of related news articles or social media, and insight tags.

[0047] The topicality score is calculated based on factors such as the number of news articles, mentions on social media, or search volume. Insight tags are set based on the content of the questions.

[0048] The voting results database stores information related to the voting results, as shown in Figure 5. This information includes the ID of the target of the vote, the user ID of the user who voted, the identifier of the selected candidate, the number of points A consumed, the date and time of the vote, and the accuracy of the prediction.

[0049] As shown in Figure 6, the points information database manages the points used for voting. Specifically, the points information database manages the point balances and increase / decrease history of each user. For example, the points information database includes a points management table that holds the point balances of each user for points A and points B, and a points log table that records the point increase / decrease history (user ID, increase / decrease date and time, reason, point type, increase / decrease amount).

[0050] In addition to the database mentioned above, the memory unit 19 also stores the analysis results from the analysis unit 16, and blockchain data generated and managed by the recording unit 15.

[0051] The user terminal 110 is an information processing terminal operated by a user of this system. The user operates the user terminal 110 to participate in the voting platform provided by the information processing device 100, and to cast votes for the target of the vote, view analysis results, etc. Note that although only one user terminal 110 is shown in Figure 1 for illustrative purposes, multiple user terminals 110 can be connected to the voting platform (information processing device 100).

[0052] The user terminal 110 is comprised of a CPU 111 (central processing unit), a storage device 112 consisting of ROM (read-on memory) and RAM (random access memory), and a computer equipped with an I / O interface 113 (input / output interface), an input device 114, and a display device 115. The RAM stores data from the CPU 111's processing, the ROM stores control programs for the CPU 111 in advance, and the I / O interface 113 is used for inputting and outputting information with connected devices. The CPU 111 executes various processes of the user terminal 110 as described herein by executing the control program. The user terminal 110 may be, for example, a mobile terminal such as a smartphone, or a personal computer (PC).

[0053] In the user terminal 110, information is displayed on the display device 115 and communication is performed with the information processing device 100, etc., in response to user operation inputs entered through the input device 114.

[0054] Next, the operation (information processing method) of the information processing device 100 and the information processing system 1000 will be described. Although each process performed by the information processing device 100 is performed by each functional unit, in the following description, it will be described as being performed by the information processing device 100.

[0055] [Voting Process] First, let's explain the voting process. The information processing device 100 executes the process shown in Figure 7 as the voting process.

[0056] In step S100, access to the voting platform is accepted. When an access request is received via the user terminal 110, a GUI for the user to cast a vote is displayed on the user terminal 110, and the process proceeds to step S101.

[0057] In step S101, a list of eligible voting targets is sent to the user terminal 110. This list is generated by referencing the voting target information DB and creating a list of voting targets that are currently in the voting period. The list includes information such as the question content, category, and voting end date and time for each voting target.

[0058] Step S102 accepts the user's input to select one item from the list of items to vote for. Once a voting item is selected, the process proceeds to step S103.

[0059] In step S103, the vote for the selected voting target is accepted. Specifically, the user inputs the candidate to be voted for and the number of points A to be used. Once the candidate and the number of points A to be used are entered on the user terminal 110 and the voting operation is performed, the voting information (user ID, ID of the selected voting target, identifier of the selected candidate, and number of points A to be used) is stored in step S104.

[0060] In step S105, the balance of points A belonging to the user stored in the memory unit 19 is subtracted by the number of points used for voting.

[0061] Step S106 determines whether the voting period for the subject has ended. If the predetermined voting end date and time have passed, it is determined that the voting period has ended and the process proceeds to step S107. If the voting period has not ended, the process in step S106 is repeated (in other words, the process waits until the voting period ends).

[0062] In step S107, the selection results (candidates selected from multiple candidates) for the voting target are registered in the database. The registration of the selection results is done, for example, by the system administrator manually entering the information from the administration screen.

[0063] In step S108, the point redemption process is performed for users who correctly predicted the selection results. Specifically, the number of points B to be redeemed is calculated for all users who voted for the selected candidates (hereinafter referred to as "correct users"). The number of redeemed points is calculated, for example, by multiplying the odds set in advance for each voting target by the number of points A used by the correct users for their votes. Then, the calculated number of redeemed points is added to the point B balance of each correct user and stored in the storage unit 19. In this way, points B are redeemed for the correct users.

[0064] In step S109, the process ends after notifying all users who voted for the target of the vote of the selection results, the number of points awarded, etc. Voting results can be notified by means of GUI notification, pop-up notification on the user's terminal, email, or SNS notification.

[0065] In this manner, the voting process for the target of the vote is executed.

[0066] [Analysis Processing] Next, the analysis process performed by the analysis unit 16 will be explained.

[0067] The analysis processes performed by the analysis unit 16 include user trend analysis, individual trend analysis, energy scoring analysis, high-value information analysis, target audience behavior analysis, and insight fit analysis. It is sufficient that the analysis unit 16 is configured to perform at least one of these processes; it is not required that it be able to perform all of them.

[0068] [User trend analysis] User trend analysis analyzes the overall voting trends of users for a given voting target. This process analyzes voting results based on user information, such as user attributes (e.g., age, gender, family structure, region, membership rank, title, etc.) and / or behavioral patterns (e.g., voting frequency, average points spent, etc.). Behavioral patterns such as voting frequency and average points spent can be calculated based on information managed in a behavioral pattern information database. User trend analysis only requires the use of at least one of user attributes and behavioral patterns; the use of both is not mandatory.

[0069] For example, the percentage of votes for each candidate or the accuracy rate of predictions can be aggregated and visualized by age group or membership rank. Such processing can be achieved, for example, by a machine learning model that performs clustering. Clustering automatically classifies user groups with similar voting tendencies, making it possible to understand the structure of what kind of opinions different types of users hold.

[0070] Furthermore, it is possible to analyze, for example, the voting trends of users who possess a specific title, or conversely, the voting trends of users who do not possess a title.

[0071] As a result of user trend analysis, voting preferences by age group (preferences and biases in voting choices) can be output in the form of distribution graphs (e.g., bar graphs, pie charts, heatmaps). In addition, indicators of the "splitting" of votes (whether choices are scattered or concentrated on one side) can be output in the form of distribution graphs or scatter plots.

[0072] Figure 8 is a flowchart showing the user trend analysis process. When the information processing device 100 receives a signal requesting the execution of user trend analysis, for example, through an operation input by a system administrator, it executes the process shown in Figure 8.

[0073] In step S200, the selection of voting targets to be analyzed is accepted. The selection of voting targets to be analyzed is performed, for example, by input by the system administrator through the input device 104 of the information processing device 100, or by automatic selection based on predetermined conditions. Once the targets for analysis are selected, the process proceeds to step S201.

[0074] In step S201, the voting result data for the voting target to be analyzed is obtained from the storage unit 19. The data obtained includes the user IDs of all users who voted for the voting target, the identifier of the candidate selected by each user, and the number of points consumed.

[0075] In step S202, attribute information and behavioral pattern information for each user are obtained from the storage unit 19 based on the list of user IDs obtained in step S202.

[0076] Step S203 analyzes the relationship between user attributes and / or behavioral patterns and voting results. The relationships between the data to be analyzed (which items to use as input and what to output as results) may be entered by the system administrator or may be pre-configured. The analysis is performed using a trained model (e.g., clustering).

[0077] For example, in clustering, each user is represented by a feature vector based on their voting behavior (tendencies in selected candidates and points spent), and users with similar vectors are classified into the same cluster. Algorithms such as k-means and hierarchical clustering are used for this process.

[0078] Next, the characteristics of each cluster formed by clustering are analyzed. Specifically, the analysis examines which user attributes (age group, membership rank, etc.) are most prevalent within each cluster. This allows for analysis of voting trends based on user attributes (which options are more likely to be chosen), correct answer trends by attribute, and factors influencing the correct answer rate (which elements contribute to the correct answer rate). More specifically, it is possible to understand voting trends such as, "Users with membership rank Gold or higher tend to concentrate on high-reward options," "Users with membership rank Bronze tend to participate lightly (with low points) and lean towards specific genres," and "Users with membership rank Platinum tend to cast heavier points just before the deadline." In this way, the relationship between user attributes and / or behavioral patterns and the voting results is analyzed.

[0079] In step S204, the analysis results are output and the process ends. The analysis results are represented, for example, by a graph showing trends or by indicator values ​​(such as variance), and are output in a dashboard format displayed on the GUI or in a file format.

[0080] As described above, user trend analysis is performed. User trend analysis allows for an intuitive understanding of the overall voting trends of users regarding the voting target and the structure of user attitudes in the market.

[0081] [Personal trend analysis] Personal trend analysis delves into and analyzes the voting tendencies of individual users. This process analyzes the relationship between user information, specifically the user's attribute information, information about the voting target, and voting behavior history (past voting dates and times, voting target, selected candidates, points spent, etc.) in a time-series manner. This allows for the evaluation of individual characteristics, such as whether the user's preferences are consistent (e.g., biased towards voting targets in a particular category) or whether they flexibly change their voting behavior in response to changes in the external environment (e.g., a sudden increase in popularity or fluctuations in odds). It should be noted that using information about the voting target is not a mandatory component of personal trend analysis.

[0082] Preference consistency is assessed by analyzing information about the voting target (information attached to the questions and options, specifically categories and insight tags, etc.) and the individual user's past voting history, and scoring the consistency of specific trends. In other words, the degree of fluctuation (stability) in the candidate preference patterns over time is calculated and evaluated as a "preference consistency score." This allows for the understanding of user voting tendencies, such as "always tending to choose trends" or "having a bias towards economic topics."

[0083] Flexibility to external changes involves analyzing and evaluating whether user selection trends change before and after changes in environmental factors. External changes include, for example, event-driven changes within society, markets, and services. Specifically, external changes include changes in the trend of correct choices (i.e., changes in the situation of the voting target), changes in topicality scores (sudden increases or decreases), changes in the concentration or dispersion of votes in specific genres, and changes in the incentive structure (changes in reward multipliers, etc.).

[0084] Individual tendency analysis can be achieved using trained models that perform processes such as regression, clustering, principal component analysis, and time series analysis. This allows for pattern analysis of user voting tendencies.

[0085] The results of the individual trend analysis can be output as, for example, a "deviation score" comparing the individual's trends with average trends by age group or membership rank, and a "trend report" summarizing the individual's tendencies. This allows for an understanding of a user's voting trends, such as "after which month did the voting tendency shift from A to B?" or "quantifying and understanding the frequency of deviations and changes from past trends." Furthermore, individual trend analysis allows for an understanding of a specific user's voting trends, such as whether the user is moving towards new trends without clinging to past trends, or in other words, whether they possess flexibility in change, the opposite of consistency and stability.

[0086] Figure 9 is a flowchart showing the process of personal trend analysis. When the information processing device 100 receives a signal requesting the execution of personal trend analysis, for example, through operation input by a system administrator, it executes the process shown in Figure 9.

[0087] In step S300, the system accepts the selection of a specific user to be analyzed. The selection of the user to be analyzed can be done, for example, by input by the system administrator through the input device 104 of the information processing device 100, or by automatic selection based on predetermined conditions. Once the target for analysis is selected, the system proceeds to step S301.

[0088] In step S301, the attribute information and voting history as time-series data of the selected user are read. The voting history data to be obtained includes the voting date and time for each vote, the target ID, the selected candidate, the number of points consumed, and the accuracy of the prediction.

[0089] In step S303, additional information related to the user's voting history is retrieved from the database. Specifically, information such as the category of each voted item, the topicality score, and the insight tag included in the voting history is retrieved.

[0090] Step S304 analyzes the relationship between user information, information about the voting target, and the user's voting history (voting results). Specifically, it analyzes the consistency of user preferences, the rate of response to external changes, and the deviation from the average in attributes.

[0091] Preference consistency is assessed, for example, by analyzing the series of voted genres or insight tags selected by users using a trained time-series model. This analysis evaluates the stability of the insight tag selection patterns and quantifies it as a "preference consistency score."

[0092] Responsiveness to external changes is assessed, for example, by using a trained time-series model to test whether there are significant changes in user voting behavior (distribution of selected insight tags, number of points consumed, etc.) before and after an external change occurs. The greater the change, the more likely the user is to flexibly change their voting behavior in response to the external environment, and this degree is quantified as a "flexibility score."

[0093] The deviation from the mean in an attribute is calculated by comparing each score of the target user with the average score of a group of users with the same attributes, for example, using a trained regression model or an algorithm employing statistical methods. The difference between the target user and the group of users with the same attributes is calculated as the "deviation score." This score serves as an indicator of how unique the target user's voting behavior is compared to users of the same age or rank. The average score of the group of users with the same attributes may be calculated and stored in advance, or it may be calculated together with the target user's score.

[0094] The analysis results are generated as a "Personal Trend Report," which integrates various scores calculated so far (consistency score, flexibility score, deviation score, etc.), the user's predicted accuracy rate, average points spent, etc., to summarize individual trends. This report includes text explaining the characteristics of the user's voting style and graphs showing the trend over time.

[0095] In step S304, the analysis results are output and the process ends. The analysis results are output in either a dashboard format displayed on the GUI or in a file format.

[0096] As described above, individual tendency analysis is performed. This analysis can have various effects, such as encouraging or improving the accuracy of voting by providing incentives to users with high scores, or identifying potential influencers.

[0097] Furthermore, by integrating the results of individual trend analyses of multiple users, it becomes possible to understand not only individual preferences but also trends in preference consistency and responsiveness to external changes by age group and membership rank. For example, "Users with Platinum membership rank tend to have stable preferences (continuing to make similar choices)," and "Users with Bronze membership rank are more likely to switch choices due to external factors (topics or changes in odds)."

[0098] [Energy scoring analysis] Energy scoring analysis quantifies the level of user interest (energy) in a voting target. This process scores not only the direct voting results but also the actions associated with them. Indicators used to calculate the score include, for example, the number of points spent on voting, the timing of voting (higher ratings for votes cast closer to the deadline, etc.), whether or not comments about the voting target were posted, the number of characters in the comments, the sentiment analysis results of the comments (positive / negative, etc.), and the number of times the topic was shared on social media.

[0099] The sentiment analysis results of comments can be obtained, for example, by performing sentiment analysis on the comment's text data using a natural language processing model. Since publicly available techniques can be used for sentiment analysis, a detailed explanation will be omitted. In sentiment analysis, the natural language processing model determines whether the comment evokes positive emotions (expectation, enthusiasm, etc.), negative emotions (anxiety, criticism, etc.), or neutral emotions based on the words and expressions in the text. Then, for example, a sentiment score for the comment (a value within the range of -1 for negative to +1 for positive) is calculated. Sentiment analysis may be performed, for example, when a comment is posted and stored in the memory unit 19 along with the comment content, or it may be performed on the comment when it is used for scoring in the enthusiasm scoring analysis.

[0100] Energy scoring analysis can be implemented, for example, by using a trained regression model that has learned the relationship between indicators and user interest levels from past data. By weighting and integrating these multiple indicators using the regression model, a "voting energy score" is calculated. Energy scoring analysis can be calculated as the strength of interest of multiple users towards a specific voting target, or as the strength of interest of a specific user towards multiple voting targets.

[0101] Figure 10 is a flowchart illustrating the process of heat scoring analysis. When the information processing device 100 receives a signal requesting the execution of heat scoring analysis, for example, through an operation input by a system administrator, it executes the process shown in Figure 10.

[0102] Step S400 accepts the selection of voting targets or users to be scored. The selection of voting targets to be analyzed can be done, for example, by input by the system administrator through the input device 104 of the information processing device 100, or by automatic selection based on predetermined conditions. For example, analysis may be performed in real time for all voting targets. Once the analysis targets are selected, the process proceeds to step S401.

[0103] Step S401 retrieves user attribute information, voting information, and associated action information for the selected analysis target.

[0104] Step S402 calculates and analyzes the voting enthusiasm score. Specifically, multiple indicators obtained in Step S401 (number of points spent, voting timing, presence or absence of comments, number of characters in comments, sentiment score, presence or absence of SNS sharing, etc.) are input into a regression model as features. In the regression model, the input features (indicators) are integrated according to their importance and weights to calculate a single voting enthusiasm score. Then, based on the user's attribute information, data on voting enthusiasm scores by age group or membership rank (for example, the average value) is calculated. This allows us to understand trends such as, "Users with a membership rank of Gold or higher tend to cast votes with higher point consumption just before the deadline, resulting in higher enthusiasm," and "Users with a membership rank of Bronze tend to show increased enthusiasm when they share on SNS."

[0105] In step S403, the analysis results are output and the process ends. The analysis results are displayed on the GUI along with the voting targets on the voting target selection screen, for example. Alternatively, the voting targets and analysis results can be listed and output in file format.

[0106] As described above, enthusiasm scoring analysis is performed. By scoring the strength of interest in a voting target, it becomes possible to, for example, place advertisements on voting targets with high interest, conduct marketing based on the strength of interest according to attributes, and set up new voting targets based on voting targets with high interest.

[0107] [High-Value Information Analysis] High-value information analysis involves obtaining information on evaluations (value levels) based on predetermined evaluation criteria, for example, in response to the needs of companies advertising on voting platforms. This process retrieves information with high evaluations on predetermined criteria from current or past voting data regarding a specified user group (target group) or voting subject (specific topic, category). The evaluation criteria (value score) that define the level of evaluation are calculated by combining multiple indicators.

[0108] These metrics include, for example, the response rate, which indicates how much the target group participated in the voting theme; consistency, which indicates whether participants continued to participate in a particular theme; and a voting enthusiasm score. Consistency is determined, for example, based on whether voting behavior continued over a predetermined period for voting targets related to a specific theme. High-value information analysis calculates a value score based on these metrics and outputs rankings of voting themes with high scores, as well as profiles of user groups that showed particularly good response. Note that the metrics (evaluation axes) used to calculate the value score are not limited to those mentioned above; other metrics may also be used.

[0109] High-value information analysis can be achieved, for example, by using a pre-trained regression model that has learned the relationship between indicators and evaluation values ​​from historical data. The regression model then weights and integrates these multiple indicators to calculate a value score.

[0110] Figure 11 is a flowchart showing the process of high-value information analysis. When the information processing device 100 receives a signal requesting the execution of high-value information analysis, for example, through an operation input by a system administrator, it executes the process shown in Figure 11.

[0111] Step S500 accepts input for analysis conditions. These conditions include attributes of the target user group (age, region, gender, occupation, etc.) and the topics or categories of interest for voting. Once the analysis conditions are entered, the process proceeds to step S501.

[0112] In step S501, voting data to be analyzed is extracted based on the entered conditions. Specifically, the memory unit 19 is referenced to identify users belonging to the entered target group and voting targets corresponding to the specified topic, and the voting results related to them are obtained.

[0113] In step S502, the extracted data is used to calculate and analyze a value score. First, response rate, consistency, and enthusiasm scores are calculated as indicators that make up the value score. Response rate represents the proportion of the specified target group that participated in the relevant voting themes during the analysis period. Consistency is an indicator that shows whether the target group participated in a particular category of themes continuously rather than just once, and is similar to the consistency of preferences in personal tendency analysis. Enthusiasm is an indicator that shows the strength of the target group's interest in the theme in question, and is the voting enthusiasm score in enthusiasm scoring analysis. Then, each indicator is input as a feature into the regression model. In the regression model, the input features (indicators) are integrated according to their importance and weight to calculate a single value score. The higher the value score, the more valuable the voting theme is considered to be to the specified company.

[0114] In step S503, high-value information is analyzed based on the target audience, voting target, and value score. The analysis results include, for example, a list of high-response themes in the target age group, a ranking of themes with high value scores, and target profiles that value specific topics (which user attributes value them highly), as well as value information associated with user information or voting target information.

[0115] In step S504, the analysis results are output and the process is terminated. The analysis results can be output in various formats, such as a dashboard displayed as a GUI, or as a list or report file.

[0116] High-value information analysis is performed in the manner described above. High-value information analysis can contribute to marketing decisions, such as determining which demographics to target when placing advertisements on voting platforms.

[0117] [Target audience behavior analysis] Target audience behavior analysis is a function that analyzes in detail the voting behavior of a specific user group (target audience) designated by, for example, a company advertising on a voting platform. The target audience is defined by combining attribute information such as age, gender, region, occupation, interest category, and membership rank. Target audience behavior analysis extracts data only from users belonging to the specified target audience and analyzes and outputs items such as the response rate for each voting theme within that group, the ranking of frequently selected candidates, the distribution of the average number of points consumed and enthusiasm score during voting, and the trend of the retention rate of voting participation.

[0118] Target audience behavior analysis is performed, for example, by machine learning models that perform clustering. Statistical methods can also be used in conjunction with target audience behavior analysis.

[0119] Figure 12 is a flowchart showing the process of target audience behavior analysis. When the information processing device 100 receives a signal requesting the execution of target audience behavior analysis, for example, through an operation input by a system administrator, it executes the process shown in Figure 12.

[0120] Step S600 accepts input for the target group to be analyzed. Once the target group is entered, the process proceeds to step S601.

[0121] In step S601, a list of users that match the input target group criteria is extracted.

[0122] In step S602, based on the extracted user list, voting behavior data for the user group belonging to the target segment is obtained. Voting behavior data includes, for example, posting date and time, choices, number of points consumed, comments and SNS shares, retention rate, voting timing (the actual timing of voting at the start or end of the voting period), and voting enthusiasm score.

[0123] In step S603, based on the data obtained in steps S601 and S602, the voting behavior trends of the target user group are analyzed. Specifically, the response rate (voting participation rate) for each voting theme, the ranking of candidates that are frequently selected within each theme, the average number of points spent when voting, the distribution of the average voting enthusiasm score, and the trend of the participation retention rate are analyzed. This makes it possible to understand what themes the target group is participating in and at what cost. For example, it is possible to identify trends such as, "Users with Platinum membership rank and aged 25 to 29 show a strong response to new product themes," and "Users with Bronze membership rank and residing in rural areas show a good response to campaign announcements."

[0124] In step S604, the analysis results are output and the process is terminated. The analysis results include the results of the analysis items listed above. The analysis results can be output in various formats, such as a dashboard displayed as a GUI, or as a list or report file.

[0125] As described above, target audience behavior analysis is conducted. Through target audience behavior analysis, it is possible to understand the trends of user groups that meet their needs, and this can contribute to marketing effectiveness, such as selecting the target audience for advertising and promotion.

[0126] [Insight Fit Analysis] Insight fit analysis analyzes the relationship between user information (attributes, voting history) and voting results, based on insights into specific users (consumers) that companies advertising on voting platforms are interested in. In this process, it is first determined which of the predetermined insights each user fits. This determination is made based on the user's voting behavior (e.g., preference for options with high discount rates or environmental themes) and the content of their posted comments (keyword extraction and contextual analysis using natural language processing). Then, the group of users determined to fit a specific insight is extracted, and the voting behavior trends unique to that group (e.g., preferred theme categories, point consumption trends, prediction win rate, etc.) are analyzed and output.

[0127] Insight fit analysis based on user attributes and voting behavior can be performed using, for example, a regression model to calculate the degree of fit to an insight, a classification model to classify which insight a user belongs to, a clustering model to group users into similar clusters based on predetermined evaluation axes, or a natural language processing model.

[0128] Figure 13 is a flowchart showing the process of insight fit analysis. When the information processing device 100 receives a signal requesting the execution of insight fit analysis, for example, through an operation input by a system administrator, it executes the process shown in Figure 13.

[0129] Step S700 accepts input for the insights to be analyzed. Insights refer to the deep psychology and motivations of consumers, such as "price-conscious," "environmentally conscious," and "brand-oriented," as mentioned above. Once the insights are entered, the process proceeds to Step S701.

[0130] In step S701, a scoring process is initiated to determine the "suitability" of all users registered on the voting platform to a specified insight. This process is based on at least one of the user's voting history and comments on the voting platform. The insight may be determined each time an insight suitability analysis is performed, or it may not be determined during the analysis but rather periodically separately.

[0131] The goodness of fit can be determined, for example, by calculating a goodness of fit score for each insight using a regression model trained on behavioral history corresponding to each insight as training data, and then determining whether the insight is a good fit based on that score. For example, if a user exhibits voting behavior such as "high frequency of voting on topics related to options with high discount rates or inexpensive products" for the "price-conscious" insight, then the goodness of fit score for the "price-conscious" insight will be calculated to be high.

[0132] Furthermore, when determining relevance based on comments, for example, a natural language processing model is used to analyze the relationship between the content of the comments and the specified insight. For example, for a "price-conscious" insight, the frequency and context of keywords such as "cost-effective," "cheap," and "good deal" are evaluated. For an "environmentally conscious" insight, keywords such as "eco," "sustainable," and "recyclable" are evaluated. Based on the results of this analysis, a comment-based relevance score can be calculated.

[0133] When both behavioral history and comments are used to determine the degree of fit, the fit scores calculated from each can be combined using a predetermined weighting to calculate a single fit score.

[0134] In step S702, a group of users whose relevance score to the insight exceeds a predetermined threshold is extracted as the "Insight-Compatible User Group".

[0135] Step S703 focuses on the extracted insight-matching user group and analyzes the voting behavior trends specific to that group. Specifically, it calculates the categories of voting themes that this user group particularly prefers, the average number of points spent, the accuracy of predictions, and the average voting enthusiasm score. Furthermore, by filtering these analysis results with user attribute information, it becomes possible to clearly understand user trends. For example, it becomes possible to identify trends such as, "Platinum users who fit the 'environmentally conscious' category respond well to sustainability-related themes," and "Bronze users who fit the 'price-conscious' category tend to respond well to options that appeal to discounts."

[0136] Step S703 outputs the analysis results and terminates the process. The analysis results can be output in various formats, such as a dashboard displayed as a GUI, or as a list or report file.

[0137] As described above, insight fit analysis is performed. Through insight fit analysis, advertising companies can gain a deep understanding of the characteristics and responses of users with the desired insights, thereby achieving greater marketing effectiveness.

[0138] Although embodiments of the present invention have been described above, these embodiments only represent a part of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments.

[0139] For example, the program according to the above embodiment is a program configured to be executable by the information processing device 100 as a computer, and is to be executed by the computer.

[0140] Furthermore, the program for executing the series of processes described above is provided on a storage medium readable by the information processing device 100. Alternatively, the program may be provided to the information processing device 100 via a network NW. Also, a portion of the processing performed by the information processing device 100 as described in the above embodiment may be executed on the user terminal 110.

[0141] The effects and advantages of this embodiment will be described below.

[0142] The information processing device 100 includes a voting reception unit 13 that accepts votes from users using points for voting targets that have multiple candidates, a reward unit 14 that rewards points to users who have voted for the selected candidate for the voting target, and an analysis unit 16 that analyzes the relationship between user information or voting target information and user voting results using a trained model.

[0143] The information processing method performed by the information processing device 100 as a computer includes the steps of: accepting votes from users using points for a voting target having multiple candidates; returning points to users who voted for the selected candidate for the voting target; storing information about the user, information about the voting target, and the voting results by the user; and analyzing the relationship between the stored information about the user or the voting target and the voting results using a trained model.

[0144] A program configured to be executable by the information processing device 100 as a computer includes the steps of: accepting votes from users using points for a voting target having multiple candidates; rewarding points to users who voted for the selected candidate for the voting target; storing information about the user, information about the voting target, and the voting results by the user; and analyzing the relationship between the stored information about the user or the voting target and the voting results using a trained model; and causing the computer to execute the program.

[0145] With these configurations, information obtained through the voting platform can be effectively utilized by analyzing information about users and voting targets in the voting process, along with the voting results.

[0146] In particular, analyzing information obtained through voting platforms can yield valuable insights for marketing.

[0147] Furthermore, in the information processing device 100, the analysis unit 16 analyzes the voting results for a specific voting target based on the attributes or behavioral patterns of the users who voted, as information about the users.

[0148] This configuration allows us to understand the overall voting trends of users and the structure of user attitudes in the market.

[0149] Furthermore, in the information processing device 100, the analysis unit 16 analyzes the user's voting history based on information about the user or information about the voting target.

[0150] This configuration can provide marketing benefits such as encouraging or improving the accuracy of voting by offering incentives to users with high scores, and identifying them as potential influencers.

[0151] Furthermore, in the information processing device 100, the analysis unit 16 scores the degree of interest in the voting subject based on the relationship between user actions associated with voting as information about the user and the voting results.

[0152] This configuration allows for marketing effects such as placing advertisements on voting targets of high interest, conducting marketing based on the level of interest according to attributes, and setting up new voting targets based on the voting targets of high interest.

[0153] Furthermore, in the information processing device 100, the analysis unit 16 acquires information regarding evaluations on predetermined evaluation axes based on the relationship between information about the user or information about the voting target and the voting results.

[0154] This configuration allows for analysis to contribute to identifying the target audience, thereby achieving effective marketing results.

[0155] Furthermore, in the information processing device 100, the analysis unit 16 analyzes the voting results among user groups having predetermined attributes.

[0156] This configuration allows for understanding the trends of user segments that meet specific needs, contributing to the selection of target audiences for advertising and promotion, thus demonstrating effective marketing results.

[0157] Furthermore, in the information processing device 100, the analysis unit 16 analyzes the voting results based on predetermined insights as information about the users, and analyzes the voting behavior trends of user groups that match the insights.

[0158] This configuration allows for a deep understanding of the characteristics and responses of users with the desired insights, enabling effective marketing.

[0159] Furthermore, the information processing device 100 includes a recording unit 15 that associates information about the user or information about the voting target with the voting results and records them in a distributed ledger (blockchain) at a predetermined time, and an analysis unit 16 performs analysis based on the information recorded in the distributed ledger.

[0160] This configuration prevents the falsification of voting results and ensures transparency regarding voting.

[0161] Although embodiments of the present invention have been described above, these embodiments only represent a part of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments. [Explanation of Symbols]

[0162] 10: User Management Department 11: Voting Management Department 12: Point Management Department 13: Voting Department 14: Reduction section 15: Records Department 16:Analysis Department 17: Information provision department 18: Display Control Unit 19: Storage part 100: Information Processing Device 101: CPU (Central Processing Unit) 102: Storage device 103: I / O Interface 104: Input device 105:Display device 110: User terminal 111: CPU (Central Processing Unit) 112: Storage device 113: I / O Interface 114: Input device 115:Display device 1000: Information Processing Systems

Claims

1. A point management unit that manages a first point that does not have monetary value and can be used for voting, and a second point that has monetary value, Regarding a voting target with multiple candidates, a voting reception unit accepts votes from users using the first point mentioned above, A reward unit that rewards the user who voted for the candidate selected for the voting target with the second points, The system includes an analysis unit that analyzes the relationship between the user information or the voting target information and the user's voting results using a trained model, The analysis unit performs analyses including user trend analysis, which analyzes the overall voting trends of the users regarding the voting target based on the user's attributes or behavioral patterns, and individual trend analysis, which analyzes the voting behavior history of a specific user in chronological order. Information processing device.

2. An information processing apparatus according to claim 1, The analysis unit analyzes the voting results for a specific voting target based on the attributes or behavioral patterns of the user who voted as information about the user. Information processing device.

3. An information processing apparatus according to claim 1, The analysis unit analyzes the voting behavior history of a specific user based on the information about the user. Information processing device.

4. An information processing apparatus according to claim 1, The analysis unit scores the degree of interest in the voting target based on the relationship between the user's actions associated with the vote and the voting results, as information about the user. Information processing device.

5. An information processing apparatus according to claim 1, The analysis unit calculates a value score that integrates multiple indicators, including the response rate, consistency, and enthusiasm for the vote, based on the relationship between the user information or the voting target information and the voting results. Information processing device.

6. An information processing apparatus according to claim 1, The analysis unit analyzes the voting results among the user group having predetermined attributes. Information processing device.

7. An information processing apparatus according to claim 1, The analysis unit analyzes the voting results based on predetermined insights as information about the users, and analyzes the voting behavior trends of the user group that fit the insights. Information processing device.

8. An information processing apparatus according to claim 1, The system includes a recording unit that associates the user information or the voting target information with the voting results and records them in a distributed ledger at a predetermined timing. The analysis unit performs analysis based on the information recorded in the distributed ledger. Information processing device.

9. An information processing device manages a first point that does not have monetary value and a second point that has monetary value, The information processing device receives votes from users using the first point for a voting target that has multiple candidates, The information processing device provides the second points to the user who voted for the selected candidate for the voting target, The information processing device stores information about the user, information about the voting target, and the voting results by the user. The information processing device includes the step of analyzing the relationship between the stored information about the user or the voting target and the voting result using a trained model, The analysis step includes a user trend analysis that analyzes the overall voting trends of the users regarding the voting target based on the user's attributes or behavioral patterns, and an individual trend analysis that analyzes the voting behavior history of a specific user in chronological order. Information processing methods.

10. A program configured to be executable by a computer, A step of managing a first set of points that do not have monetary value and can be used for voting, and a second set of points that do have monetary value, For a voting target with multiple candidates, the first step involves accepting votes from users using the aforementioned points, The steps include: rewarding the user who voted for the candidate selected for the voting target with the second points; A step of storing information about the user, information about the voting target, and the voting results by the user, The computer is instructed to perform the following steps: analyze the relationship between the stored information about the user or the voting target and the voting results using a trained model; The analysis step includes a user trend analysis that analyzes the overall voting trends of the users regarding the voting target based on the user's attributes or behavioral patterns, and an individual trend analysis that analyzes the voting behavior history of a specific user in chronological order. program.

Images