Information processing systems, information processing methods, and programs

Abstract

To provide more beneficial technology. [Solution] According to one aspect of the present invention, an information processing system is provided, comprising at least one processor, the processor configured to perform the following steps by reading a program, wherein in the acquisition step, open data published on the internet is acquired, the open data includes content information relating to one or more reference contents, in the identification step, at least one of an evaluation axis for evaluating a predetermined target content and weights for each evaluation axis is identified based on the open data and predetermined conditions, and in the output step, an evaluation result is output showing the result of evaluating the target content based on at least one of the identified evaluation axis and weights.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a technique for evaluating the content of a document from various viewpoints.

Prior Art Literature

Patent Literature

[0003]

Patent Document 1

Summary of the Invention

Problem to be Solved by the Invention

[0004] However, there is room for improvement in appropriately providing content evaluation results to users.

[0005] In view of the above circumstances, the present invention provides a more useful technology.

Means for Solving the Problem

[0006] According to one aspect of the present invention, there is provided an information processing system comprising at least one processor, wherein the processor is configured to execute the following steps by reading a program: in an obtaining step, open data published on the Internet is obtained, the open data includes content information related to one or more reference contents; in an identifying step, at least one of an evaluation axis for evaluating a predetermined target content and a weight for each evaluation axis is identified based on the open data and a predetermined condition; and in an output step, an evaluation result indicating a result obtained by evaluating the target content based on at least one of the identified evaluation axis and the identified weight is output.

[0007] This disclosure allows us to provide more beneficial technology. [Brief explanation of the drawing]

[0008] [Figure 1] This is a diagram showing the configuration of information processing system 1. [Figure 2] This is a block diagram showing the hardware configuration in this embodiment. [Figure 3] This is a block diagram showing the functional configuration of server device 2 according to one embodiment. [Figure 4] This is an activity diagram showing the information processing flow of this embodiment. [Figure 5] This is an activity diagram showing the information processing flow of this embodiment. [Figure 6] This figure shows an example of a data structure for information that is appropriately referenced in the information processing of this embodiment. [Figure 7] This figure shows an example of an evaluation screen 900 displayed on the display unit 34 of the user terminal 3. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the drawings. The various features shown in the embodiments below can be combined with each other.

[0010] Incidentally, the program for implementing the software appearing in one embodiment may be provided as a non-transitory computer-readable medium, or it may be provided as a downloadable medium from an external server, or it may be provided so that the program is launched on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0011] Furthermore, in various information processing according to one embodiment, an input and an output corresponding to the input can be realized. Here, as long as an output is obtained as a result of the input, the form of the information referenced in such information processing (hereinafter referred to as "reference information") is not limited. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a decision formula such as a regression equation constructed by a statistical method), or a trained model that has been pre-trained to learn the correlation between input and output, or a generative AI such as a large-scale language model that can output a desired result by inputting a prompt (these models include parameters that construct the correlation relationship between input and output) or a visual language model.

[0012] Furthermore, in one embodiment, "part" may include, for example, hardware resources implemented by a circuit in a broad sense, and the information processing of software that can be specifically realized by these hardware resources. Also, in one embodiment, various types of information are handled, and this information can be represented, for example, by the physical values ​​of signal values ​​representing voltage and current, the high or low values ​​of signal values ​​as a set of binary bits composed of 0s or 1s, or by quantum superposition (so-called qubits), and communication and calculations can be performed on a circuit in a broad sense.

[0013] Furthermore, a circuit in a broad sense is a circuit realized by combining at least a suitable combination of circuits, circuits, processors, and memory. The processor may be a general-purpose processor or a dedicated circuit. In other words, it includes application-specific integrated circuits (ASICs), programmable logic devices (for example, simple programmable logic devices (SPLDs), complex programmable logic devices (CPLDs), and field programmable gate arrays (FPGAs)), etc.

[0014] [Embodiment] 1. Hardware Configuration This section describes a hardware configuration according to one embodiment.

[0015] 1.1 Information Processing System 1 Figure 1 is a configuration diagram representing information processing system 1. Information processing system 1 comprises a server device 2 and a user terminal 3. The server device 2 and the user terminal 3 are configured to communicate with each other via a network 11. Here, the system exemplified in information processing system 1 consists of one or more devices or components.

[0016] 1.2 Server device 2 Figure 2(A) is a block diagram showing the hardware configuration of server device 2. Server device 2 comprises a communication bus 20, a communication unit 21, a storage unit 22, and a processor 23. The communication unit 21, the storage unit 22, and the processor 23 are electrically connected within server device 2 via the communication bus 20.

[0017] The communication unit 21 is a wired or wireless communication means. The server device 2 may communicate various information from outside via the communication unit 21 and a network.

[0018] The storage unit 22 stores various information defined by the above description. The storage unit 22 can function as various storage devices, or as a memory that stores temporarily necessary information (arguments, arrays, etc.) related to the operation of a program.

[0019] The processor 23 performs processing and control of overall operations related to the server device 2. The processor 23 is, for example, a Central Processing Unit (CPU). The processor 23 implements various functions related to the server device 2 by reading a predetermined program stored in the storage unit 22.

[0020] 1.3 User Terminal 3 FIG. 2(B) is a block diagram showing the hardware configuration of the user terminal 3. The user terminal 3 includes a communication bus 30, a communication unit 31, a storage unit 32, a processor 33, a display unit 34, and an input unit 35. The communication unit 31, the storage unit 32, the processor 33, the display unit 34, and the input unit 35 are electrically connected inside the user terminal 3 via the communication bus 30. Descriptions of the communication unit 31, the storage unit 32, and the processor 33 are omitted since they are the same as the descriptions of each unit in the server device 2. As shown in FIG. 1, the user terminal 3 may be a general-purpose computer 3a or a mobile terminal 3b.

[0021] The display unit 34 displays a screen of a Graphical User Interface (GUI) operable by a user. The display unit 34 may be included in the housing of the user terminal 3 or may be externally attached.

[0022] The input unit 35 receives operation inputs made by the user. These operation inputs are transmitted to the processor 33 via the communication bus 30 as command signals. The processor 33 can, if necessary, perform predetermined controls or calculations based on the transmitted command signals. The input unit 35 may be included in the casing of the user terminal 3 or it may be an external component.

[0023] 2. Functional Configuration Next, with reference to Figure 4, the functional configurations of the information processing system 1 will be described. Figure 4 is a block diagram showing the functional configuration of a server device 2 according to one embodiment. As shown in Figure 4, the processor 23 functions as each of the illustrated functional units by executing various programs stored in the storage unit 22.

[0024] The acquisition unit 231 is configured to acquire various types of information as an acquisition step. Specifically, the acquisition unit 231 is configured to acquire information via the communication unit 21 or the storage unit 22 and to make it readable into the working memory.

[0025] The reception unit 232 receives various inputs from the user. The reception unit 232 receives user inputs received at the input unit 35 of the user terminal 3 via the communication unit 21.

[0026] The identification unit 233 identifies evaluation axes or weights for each of them based on open data and predetermined conditions.

[0027] The output unit 234 outputs evaluation results regarding the content based on the specified evaluation axis.

[0028] The display control unit 235 performs a display control step which involves processing to display various information on a display medium in a manner that is recognizable to the user. When the phrase "display" is used, it is not necessary to know whether the display medium to be displayed is in the local environment or whether the processing to display it is performed via a network.

[0029] The calculation unit 236 is configured to perform various calculations related to the server device 2 as calculation steps. The type of calculation is not particularly limited.

[0030] The artificial intelligence unit 237 is configured to receive input from each functional unit and return the instructed output. The artificial intelligence unit 237 receives a predetermined input and instructs the artificial intelligence module to return the instructed output. The artificial intelligence module used by the artificial intelligence unit 237 may be stored in the memory unit 32 and memory unit 22, or it may be stored in an external artificial intelligence server different from the server device 2 and user terminal 3.

[0031] An artificial intelligence module is an AI (Artificial Intelligence) equipped with various language models and other learning models, and may be a generative AI or an AI agent. The language model may be not only one trained for a specific task, but also a general-purpose model that can be used for a wide range of tasks (for example, a large language model (LLM) that has learned from a vast amount of data).

[0032] Generative AI performs various tasks using language models. Specifically, it is configured to respond to given input (prompts) by processing a wide range of tasks such as understanding sentence patterns and context, answering questions, and generating text, images, and audio, and returning the instructed output. Prompts can be given in any format, including text, images, and audio. Tasks may also be processed interactively, with input reception and information generation and output alternating.

[0033] The AI ​​agent performs various tasks using a language model. Specifically, when a predetermined instruction is input, the AI ​​agent is configured to break down the processes necessary to achieve the specified task into subtasks, actions, etc., and to perform necessary data collection and analysis, program generation and execution, etc. The AI ​​agent may autonomously plan and execute tasks, and evaluate the results of its execution, thereby autonomously learning to achieve its goals.

[0034] 3. Information Processing Methods This section describes the information processing method performed by the information processing system 1. The order of the processes included in the information processing method can be rearranged as appropriate, multiple processes may be executed simultaneously, and some processes may be omitted. 3.1 Details of Information Processing

[0035] Figures 4 and 5 are activity diagrams showing the flow of information processing in this embodiment. Figure 6 is a diagram showing an example of the data structure of information that is referenced as appropriate in the information processing of this embodiment. The various information and data shown in Figure 6 may be stored in the storage unit 22 of the server device 2 or the storage unit 32 of the user terminal 3, or they may be stored in a predetermined storage area including an external storage device. In this embodiment, as an example, it is shown that each table T1 to T4 is stored in the storage unit 22 of the server device 2. Furthermore, each table T1 to T4 may be configured separately or as an integrated database.

[0036] In this embodiment, the information processing system 1 collects open data to identify at least one of the evaluation axes and weights for each evaluation axis (hereinafter referred to as "evaluation axes, etc.") for evaluating a predetermined target content, and outputs an evaluation result showing the result of the evaluation of the target content based on this.

[0037] The target content is subject to evaluation based on predetermined evaluation criteria. Open data includes content information about reference content that meets predetermined conditions, and the evaluation criteria are identified based on the content information. Content information is arbitrary information describing the content of the reference content and the target content. By selecting reference content and target content to meet predetermined conditions, the target content can be evaluated based on evaluation criteria set to show characteristics related to those predetermined conditions. In the following example, the case where the reference content and target content are stores is given. However, the reference content and target content are not limited to this and can be applied to any subject to evaluation.

[0038] In this embodiment, the reference content and the target content may be stores that provide a predetermined product. The predetermined condition is the region or location where the store provides the product. In this case, by collecting open data including user ratings and reviews of multiple stores located in the predetermined region, it is possible to identify the common demands and evaluation criteria that users have for stores in that region. By reflecting such information in the evaluation axes, the stores being evaluated can be evaluated taking into account the characteristics of the region.

[0039] Open data includes information that is publicly available on the internet and therefore accessible to anyone. Open data may include, for example, information published on any website, such as social media posts, word-of-mouth information, and user rating data. This allows for analysis that reflects real-time market trends and consumer opinions. Open data may also include information that is only accessible after fulfilling certain conditions, such as registering as a member. Open data can be automatically obtained through methods such as APIs provided by websites or through web crawling (scraping).

[0040] Referring to Figure 4, the processing flow for identifying evaluation axes, etc., will be explained. First, the acquisition unit 231 acquires information about the reference content (Activity A101). The acquisition unit 231 may acquire open data about content that meets predetermined conditions, for example, open data for each major category such as "stores," "offices," and "transportation," or for each minor category such as "restaurants," "convenience stores," and "drugstores." The acquisition unit 231 may also acquire open data for each region such as "Azabu-Juban" and "Roppongi," or for each language such as "Japanese," "English," and "Chinese." In other words, the acquisition unit 231 acquires open data that is publicly available on the internet. The open data also includes content information about one or more reference contents.

[0041] Next, the identification unit 233 identifies a provisional evaluation axis. The provisional evaluation axis is an index used to evaluate the reference content and identify a provisional evaluation result, and the evaluation axis may be identified based on the provisional evaluation result. Hereinafter, the provisional evaluation result for a store as reference content may be referred to as a shop summary. The provisional evaluation axis may be identified based on the category of the reference content or predetermined conditions, or it may be specified by the user. That is, the reception unit 232 may accept input from the user to set the evaluation axis. In this way, the acquisition unit 231 acquires information regarding the provisional evaluation axis.

[0042] Figure 6(A) shows the evaluation axis table T1, which manages information related to evaluation axes. The evaluation axis table T1 records information about, for example, a provisional evaluation axis. In the evaluation axis table T1, the type, evaluation axis, explanation of the evaluation axis, and weighting are recorded in association with the evaluation axis ID.

[0043] The "evaluation axis ID" is information that uniquely identifies each of the evaluation axis sets containing one or more evaluation axes, and is composed of a string or similar. Instead of the evaluation axis ID, a set name consisting of a string that allows a human to uniquely perceive the set of evaluation axes may be used.

[0044] The "type" is information indicating which of several types the set of evaluation axes belongs to. The type may be set in correspondence with predetermined conditions that the referenced content satisfies. This allows for the setting of evaluation axes, etc., that are appropriate for the target content. The type may be set to correspond to one or more of the predetermined conditions mentioned above; for example, "restaurants," "restaurants in Azabu-Juban," etc., may be set.

[0045] An "evaluation axis" is an index used to evaluate reference content and may include information that identifies at least one evaluation axis. In this embodiment, each evaluation axis may be set as a string in natural language in order to perform the evaluation of each content using a large-scale language model. In this embodiment, the evaluation axes corresponding to "restaurants" may be set as "1. Price, 2. Atmosphere, 3. Trendiness, 4. Popularity, 5. Service".

[0046] The "Explanation of Evaluation Criteria" section contains detailed explanations and evaluation criteria for each evaluation criterion. If these are clear from the name of the evaluation criterion, the explanation section does not need to be included.

[0047] "Weighting" is information that specifies which evaluation axis is given more weight when determining the overall evaluation of referenced content using each evaluation axis. Weighting can be specified numerically or as information indicating a level such as "high" or "low".

[0048] Next, the artificial intelligence unit 237 generates a prompt (Activity A103) based on content information for each referenced content and information on a provisional evaluation axis. Then, the artificial intelligence unit 237 inputs the generated prompt into the large-scale language model (Activity A104). The prompt includes instructions to output a provisional evaluation result showing the result of evaluating each referenced content from the perspective of each evaluation axis. The large-scale language model takes the content information describing the content of the referenced content and evaluations to the user, as well as the content of each evaluation axis, explanations of the evaluation axes, and weightings as input and outputs the evaluation result for each referenced content.

[0049] The acquisition unit 231 acquires the provisional evaluation results output by the large-scale language model. The identification unit 233 then identifies the provisional evaluation results that show the evaluation results for each of the multiple reference contents based on the provisional evaluation axis (Activity A105).

[0050] Figure 6(B) shows the shop summary table T2, which manages information related to the evaluation of each store (referenced content). The shop summary table T2 records information used to evaluate each store, as well as information related to the evaluation results. In the shop summary table T2, basic information, open data, evaluation axis ID, output instruction (prompt), score, evaluation reason, and characteristics of the referenced content are recorded in association with the referenced content ID (store name).

[0051] A "reference content ID" is information that uniquely identifies each store as reference content, and is composed of a string of characters or the like. The reference content ID may be the store name, which is composed of a string of characters that allows a human to uniquely perceive the store. The stores as reference content may be, for example, stores located in a specified area that meet certain conditions, and a predetermined number of stores may be extracted in order of their high customer review ratings on a specified website.

[0052] "Basic information" refers to information that identifies the attributes of the referenced content, and includes store name, category, price range, location, business hours, etc. Basic information can be identified based on content published on official websites, etc., which is included in open data.

[0053] "Open data" refers to information collected from the internet in order to evaluate the referenced content. Open data may include, for example, content published on the official websites of individual stores, review scores and content posted on review sites, and user posts on social media.

[0054] The evaluation axis ID is information that identifies a provisional evaluation axis used to evaluate the referenced content. The provisional evaluation axis can be identified by associating it with an evaluation axis ID recorded in the evaluation axis table T1. The evaluation axis may be specified by the user, or it may be identified based on predetermined conditions that the referenced content satisfies and the type of set of evaluation axes in the evaluation axis table T1.

[0055] An "output instruction (prompt)" is a statement of instruction to cause a large-scale language model to output the evaluation results of the referenced content. A prompt is generated by referring to at least "open data" and "evaluation axis IDs," and may be composed of natural language such as, for example, "For stores related to 'referenced content ID,' refer to the contents of 'open data,' and output the 'score' (0-5 points), 'evaluation reason,' and overall 'characteristics of the referenced content' for each evaluation axis included in 'evaluation axis ID' as evaluation results."

[0056] "Score," "Evaluation Reason," and "Characteristics of the Referenced Content" are pieces of information that indicate a provisional evaluation result of the referenced content and can be output by a large-scale language model. The "Score" is a numerical representation of the evaluation result for each evaluation axis of the referenced content. The score may include a total score that is an aggregate of the scores for each evaluation axis, and the total score may be the sum of the scores for each evaluation axis, the average value, or a weighted average value that takes weights into account. "Evaluation Reason" is information that describes the reason for calculating each score in relation to open data. The evaluation reason may show the reasons for calculating multiple evaluation axes together, and may include natural language descriptions such as, for example, "In reviews, it is rated as high-quality and satisfying despite being on the expensive side." "Characteristics of the Referenced Content" is information that describes the overall characteristics of the referenced content and may include natural language descriptions such as, "A modern and calm cafe. Although it is on the expensive side, its Instagrammable atmosphere is highly rated."

[0057] Returning to Figure 4, the identification unit 233 identifies predetermined conditions for identifying evaluation axes, etc. (Activity A106). In this embodiment, the identification unit 233 identifies the store category "restaurant" and the region "Azabu-Juban" as predetermined conditions.

[0058] Next, if the processor 23 determines that a sufficient number of preliminary evaluation results (shop summaries) for reference content that meets the predetermined conditions have not been obtained, the process returns to activity A101 and continues. If it determines that a sufficient number of shop summaries have been obtained, the process proceeds to identify evaluation axes, etc. (area summaries).

[0059] The artificial intelligence unit 237 generates a prompt (Activity A107) based on information regarding a provisional evaluation axis and provisional evaluation results for each reference content. The artificial intelligence unit 237 then inputs the generated prompt into a large-scale language model (Activity A108). The prompt includes instructions to output the result of a comprehensive evaluation of multiple reference content items that meet predetermined conditions from the perspective of each evaluation axis included in the provisional evaluation axis. The large-scale language model takes the content of the provisional evaluation results for each reference content as input and outputs the result of a comprehensive evaluation of multiple reference content items that meet the predetermined conditions.

[0060] The acquisition unit 231 acquires the results output by the large-scale language model. The evaluation result is the result of a comprehensive evaluation of "restaurants in Azabu-Juban" as reference content that meets predetermined conditions, and is information that indicates the regional characteristics of "restaurants in a certain area." In the following description, the evaluation result may be referred to as an area summary. In this embodiment, evaluation axes and the like for evaluating the target content are provided based on the area summary.

[0061] In other words, the identification unit 233 identifies at least one of the evaluation axis and the weighting based on open data and a provisional evaluation axis. The identification unit 233 also identifies at least one of the evaluation axis and the weighting based on the provisional evaluation axis and provisional evaluation results for multiple reference contents. Preferably, the identification unit 233 identifies at least one of the evaluation axis and the weighting by inputting a prompt to the large-scale language model that includes a provisional evaluation axis, predetermined conditions, and provisional evaluation results for multiple reference contents, thereby causing the large-scale language model to output at least one of the evaluation axis and the weighting.

[0062] Figure 6(C) shows an area summary table T3 that manages information for comprehensively evaluating reference content that meets predetermined conditions. Area summary table T3 records information used, for example, to comprehensively evaluate stores that meet predetermined conditions, as well as information related to the evaluation results. In area summary table T3, the content of the conditions, the shop summary, the evaluation axis ID, the output instruction (prompt), the overall explanation, the explanation of the evaluation axis, and the weighting of the evaluation axis are recorded in association with the condition ID (area name).

[0063] The "Condition ID" is information that uniquely identifies a predetermined condition related to the referenced content, and is composed of a string or similar element. The "Condition Content" is information that describes the predetermined condition in natural language, and may include one or more category names, region names, etc.

[0064] The "Shop Summary" is information used to identify an Area Summary and may include a provisional evaluation result (score, evaluation reason, characteristics of the referenced content) in the Shop Summary Table T2. To identify an Area Summary, a predetermined number (e.g., 100) or more Shop Summaries related to referenced content may be referenced.

[0065] The "Evaluation Axis ID" is information that uniquely identifies each of the evaluation axes in a set of evaluation axes that includes one or more evaluation axes, and is composed of a string or similar element. The Evaluation Axis ID may be the same as the evaluation ID used to identify shop summaries that meet certain conditions in the shop summary table T2, or a different evaluation ID may be specified.

[0066] An "output instruction (prompt)" is a statement of instruction to cause a large-scale language model to output an area summary (at least one of the evaluation axes and their weights). A prompt is generated by referring to at least a "shop summary" and may consist of natural language such as, for example, "Refer to the "shop summary" and evaluate the overall characteristics of "condition ID" that satisfy "condition content" based on the evaluation axis of "evaluation axis ID". Output the evaluation results so that they include "overall explanation", "explanation for each evaluation axis", and "weighting". In other words, the "overall explanation", "explanation of evaluation axes", and "weighting of evaluation axes" are information (area summary) that shows the results of a comprehensive evaluation of reference content that satisfies the given conditions, output by the large-scale language model.

[0067] A "comprehensive explanation" is information that describes the common characteristics of reference content that meets certain conditions. For example, it can be described in natural language as follows:

[0068] <1> Explanation of the analysis results for restaurants The Azabu-Juban area is a sophisticated neighborhood where tradition and innovation coexist, and it's a popular gourmet spot with a wide variety of restaurants.

[0069] The "Explanation of Evaluation Criteria" is information that describes the common characteristics of reference content that meets the specified conditions, for each evaluation axis. For example, it may be described in natural language as shown below. The "Weighting" is set to be consistent with the "Explanation of Evaluation Criteria." The weighting may also be described in natural language in the "Explanation of Evaluation Criteria."

[0070] (1) Price Users in the Azabu-Juban area tend to place a strong emphasis on "cost performance," valuing the balance between quality and price at high-end establishments, and the reasonable prices relative to the location at more casual restaurants. (2) Atmosphere ... (3) Topic ... (4) Popularity ... (5) Services ...

[0071] Then, instead of the "explanation of evaluation axis" and "weighting" recorded in the evaluation axis table T1, the target content is evaluated based on the "explanation of evaluation axis" and "weighting" recorded in the area summary table T3. For example, "price" will be evaluated from the perspective of "cost performance" based on the description in the "explanation of evaluation axis," and the relative importance (weighting) of the evaluation of "price" alone will decrease. In other words, the identification unit 233 identifies at least one of the evaluation axis and weighting by modifying the evaluation axis (provisional evaluation axis) in the evaluation axis table T1 to show characteristics related to predetermined conditions based on the shop summary (provisional evaluation result).

[0072] In other words, the specific unit 233 identifies at least one of the following: an evaluation axis (area summary) for evaluating a predetermined target content based on open data (shop summary) and predetermined conditions (region and store category), and weights for each evaluation axis.

[0073] Next, referring to Figure 5, we will explain the process flow for performing an evaluation of the target content based on the identified evaluation axes, etc. First, the acquisition unit 231 acquires information about the target content (Activity A201).

[0074] Information regarding the target content may include at least one of open data and local data that is not publicly available on the internet. For example, in response to the acquisition unit 231 acquiring at least some of the basic information of a store as the target content (such as the store name and address), the acquisition unit 231 may be configured to collect open data about the store from the internet. Alternatively, the reception unit 232 may be configured to accept input of local data about the store from the user, for example, it may accept information that identifies the store's performance, employee composition, business processes, etc.

[0075] Next, the identification unit 233 identifies predetermined conditions that the target content satisfies (Activity A202). Furthermore, the identification unit 233 identifies an area summary that satisfies the identified predetermined conditions by referring to the area summary table T3 (Activity A203). This identifies the evaluation axes used to evaluate the target content. Note that the predetermined conditions and area summaries may be specified by the user. In other words, the identification unit 233 identifies evaluation axes according to the type of target content. The type of target content is identified based on predetermined content that the target content satisfies. Then, the output unit 234 outputs the evaluation results based on the identified evaluation axes.

[0076] Next, the artificial intelligence unit 237 generates a prompt based on information about the target content and information about the evaluation axes, etc. (Activity A204). Then, the artificial intelligence unit 237 inputs the generated prompt into the large-scale language model (Activity A205). The prompt includes instructions for outputting the results of evaluating the target content based on the evaluation axes, etc. The large-scale language model takes the information about the target content and information about the evaluation axes, etc. as input and outputs the evaluation results of the target content evaluated based on the evaluation axes, etc.

[0077] The acquisition unit 231 acquires the evaluation results output by the large-scale language model. These evaluation results represent the evaluation of the target content based on evaluation axes and other factors that indicate characteristics related to predetermined conditions. This allows the target content to be evaluated more appropriately.

[0078] Figure 6(D) shows the target content table T4, which manages information about the target content. The target content table T4 records information for identifying the target content, information for managing the evaluation criteria, and information about the evaluation results. In the area summary table T3, basic information, open data, local data, condition ID (area name), score, evaluation reason, and characteristics of the target content are recorded in association with the target content ID (store name).

[0079] The "Target Content ID" is information that uniquely identifies each store as target content, and is composed of a string of characters or similar. The Target Content ID may be the store name, which is composed of a string of characters that allows a human to uniquely perceive the store.

[0080] Basic information, open data, and local data are information that identifies the content of the target content to be evaluated. "Basic information" is the same as the basic information in the shop summary table T2, and the target content may be specified by the user entering basic information. "Open data" and "local data" include content that records information collected for the purpose of evaluating the target content. Open data may be collected when basic information is specified. Local data may be entered by the user along with the basic information.

[0081] The "Condition ID" is information that identifies the evaluation criteria, etc., of the target content. By referencing the area summary table T3 based on the Condition ID, an area summary that matches the "predetermined conditions" that the target content satisfies can be identified. This allows for the identification of information regarding evaluation criteria, etc., that are suitable for the predetermined conditions that the target content satisfies.

[0082] The score, evaluation reason, and characteristics of the target content are pieces of information that show the evaluation results of the target content. The "score" is a numerical representation of the evaluation result for each evaluation axis. The score may include a total score which is an aggregate of the scores for each evaluation axis, and the total score may include the sum of the scores for each evaluation axis, the average value, or a weighted average value that takes into account predetermined weights. The "evaluation reason" is information that describes the reason for calculating each score in relation to the area summary. For example, it may include a natural language description such as, "Although the price is high, considering the high quality of the service, it matches the characteristic of Azabu-Juban where cost performance is highly valued." The "characteristics of the target content" is information that describes the overall characteristics of the target content, and may include a natural language description such as, "A cafe where you can enjoy a high-quality time typical of Azabu-Juban. Although the price is high, the luxurious feel and cost performance that outweighs the price are highly rated."

[0083] Next, the output unit 234 outputs an evaluation result (Activity A207) showing the result of evaluating the target content based on at least one of the identified evaluation axis and weighting. Then, the display control unit 235 displays the evaluation result on the user terminal 3 so that it can be recognized by the user who is the provider of the target content.

[0084] The evaluation results presented to the user may include explanations of the evaluation criteria. This allows the user to consider the reasons behind the evaluation results for the target content. In particular, it enables the user to understand the strengths and weaknesses of the target content, taking into account the characteristics of the predetermined conditions that the target content satisfies. In other words, the output unit 234 outputs evaluation results that include the evaluation criteria and / or explanations for each evaluation criterion.

[0085] 3.2 Specific Examples of Information Processing This section further describes specific examples of processes performed in Information Processing System 1. For each specific example, the related processes will be explained with reference to the activity diagrams shown in Figures 4 and 5 as appropriate. In other words, each specific example of a process may be included within the scope defined in the overview and details of the information processing described above.

[0086] Figure 7 shows an example of a screen whose display is controlled by the display control unit 235 in the information processing system 1. Figure 7 shows an example of an evaluation screen 900 displayed on the display unit 34 of the user terminal 3.

[0087] The evaluation screen 900 is a screen displayed to the user when they operate user terminal 3 to perform an evaluation of the target content. For example, the evaluation screen 900 is a screen for evaluating a store (a cafe in Azabu-Juban) as the target content. User terminal 3 is associated with the store, and the basic information of the store has been specified in advance by the user.

[0088] The evaluation screen 900 includes a specification area 910 and an output area 940. The specification area 910 is an area for receiving information from the user to be used to evaluate the target content. The output area 940 is an area where the evaluation results of the target content are output.

[0089] The designated area 910 includes a data designated area 920, an evaluation content designated area 930, and an execution button 935. The data designated area 920 is an area that accepts instructions for obtaining information about the target content. The evaluation content designated area 930 is an area that accepts instructions for specifying the evaluation content of the target content. The execution button 935 is an object that accepts instructions for performing an evaluation of the target content.

[0090] In the data specification area 920, when button 921 is operated by the user, the acquisition unit 231 acquires open data related to the target content. For example, the acquisition unit 231 collects open data about the store based on the basic store information entered by the user. Also, when button 922 is operated by the user, the reception unit 232 displays a screen on the display unit 34 that accepts input of local data about the store from the user.

[0091] Each of the buttons 931-934 is associated with an evaluation of the target content. When the user has selected at least one of the buttons 931-934, and the execute button 935 is operated, an evaluation of the target content is performed based on the selected evaluation.

[0092] Button 931 is set to "Area Evaluation" as the evaluation content. Area evaluation is an evaluation performed on the store in question, with predetermined conditions set for the area in which the store is located. In this case, the area summary table T3 is referenced to identify the area summary corresponding to the area in which the store is located, and the evaluation axes, etc. are identified based on the identified area summary. The identified evaluation axes, etc. are recorded in the target content table T4, and the evaluation results are also recorded in the target content table T4.

[0093] Output area 940 includes evaluation output area 941, explanation output area 942, and feature output area 943. Evaluation output area 941 is the area where the score recorded in the target content table T4 is displayed. Explanation output area 942 is the area where the explanation for each evaluation axis recorded in the area summary table T3 is displayed. Feature output area 943 is the area where the features of the store recorded in the target content table T4 are displayed. In this way, by displaying the store's features in natural language in addition to the store's score in output area 940, users can concretely grasp the store's evaluation in that area. Furthermore, by displaying explanations for each evaluation axis, users can gain a deeper understanding of the evaluation. In addition, by outputting the text of an article describing the store's features in feature output area 943, the evaluation results can be used for store PR.

[0094] Furthermore, button 932 is set to "Language-Specific Evaluation" as the evaluation content. Language-Specific Evaluation is a user's evaluation of the store in a specific language. In this case, preferably, the shop summary shown in shop summary table T2 is identified based on open data written in a predetermined language. Then, predetermined conditions in area summary table T3 may include information specifying the language, and the area summary may be identified based on the shop summary identified by open data written in the predetermined language. This makes it possible to identify evaluation axes for stores that reflect the characteristics of users who use a specific language. In other words, the predetermined condition is the language used by the user to whom the evaluation results are presented. Furthermore, the acquisition unit 231 adjusts the range of open data acquisition according to the predetermined conditions and acquires the data.

[0095] Furthermore, the output area 940 displays the store evaluation results in each language. This makes it possible to automatically generate multilingual content. For example, store information and reviews can be optimized and disseminated to consumers in each country. It can also be linked with SNS platforms to disseminate the generated store information and other content to multinational users. For example, the language used on the linked SNS platform may be set based on information specifying "language" as a predetermined condition. With this configuration, the information processing system 1 can support the global expansion of stores and reach a wider range of users. Moreover, it can automatically generate content with expressions and designs suitable for the target market, taking into account the cultural background and trends of each region, rather than simply providing direct translations.

[0096] Furthermore, button 933 has "Recommended Influencers" set as one of its evaluation criteria. When button 933 is operated, for example, a store evaluation is performed, and the user is presented with information about potential sources that disseminate information about that store. This makes it possible to identify influential users (influencers) on social media and promote the store by collaborating with them.

[0097] In this case, the acquisition unit 231 acquires information about potential senders who disseminate information about the store (target content). For example, the acquisition unit 231 acquires online posts of candidate senders as open data. Based on the acquired open data, the identification unit 233 identifies the trends in the candidate's posts and target users, and evaluates their consistency with the store's area evaluation results. The output unit 234 outputs candidates with high consistency with the store's evaluation results and presents them to the user. The candidate evaluation results may also be output by inputting the store's evaluation results and sender information into a large-scale language model. In other words, the output unit 234 outputs evaluation results showing the evaluation of the sender based on at least one of the identified evaluation axes and weightings, and the sender information. This allows the user to be presented with candidate senders who are expected to disseminate information about the store that will resonate with consumers.

[0098] Furthermore, button 934 is set to "Operational Improvement Suggestions" as an evaluation option. By operating button 934, suggestions for improvements related to store operations may be submitted.

[0099] In this case, preferably, the acquisition unit 231 acquires local data that is not publicly available on the internet. The local data includes provider information about the provider (user in this embodiment) that provides a predetermined product. The provider information may include information such as specifications showing the specific composition of the product, raw materials, composition, etc., sales figures showing business conditions, various expenses, information such as the price and profit margin of each product, and information such as manuals and procedures related to store operations. The identification unit 233 then refers to information based on open data such as the area evaluation results of the store in question and outputs an evaluation of at least one of the product and the operations for providing the product. This allows for the analysis of the collected open data and supports the optimization of operations such as sales management and procurement ordering for each store. For example, by comparing the sales data of each store with market trends, the AI ​​makes a demand forecast and recommends an appropriate procurement quantity. In other words, the output unit 234 outputs an evaluation result showing the result of evaluating at least one of the product and the operations for providing the product based on at least one of the identified evaluation axes and weightings, and the provider information.

[0100] 4. Others

[0101] In one embodiment, a functional unit implemented by the processor 23 of the server device 2 is described, but at least a part of this may be implemented as a functional unit implemented by another server. Alternatively, it may be implemented as a functional unit implemented by the processor 33 of the user terminal 3. Furthermore, the various types of information described in the above example may be stored not only in the storage unit 22 of the server device 2, but also distributedly on other external devices using blockchain technology or the like.

[0102] Furthermore, they may be provided in the following embodiments.

[0103] (1) An information processing system comprising at least one processor, wherein the processor is configured to perform the following steps by reading a program, the acquisition step of acquiring open data published on the Internet, wherein the open data includes content information relating to one or more reference contents, the identification step of identifying at least one of an evaluation axis for evaluating a predetermined target content and a weight for each of the evaluation axes based on the open data and predetermined conditions, and the output step of outputting an evaluation result showing the result of evaluating the target content based on at least one of the identified evaluation axis and the weight.

[0104] This approach allows for the evaluation of appropriate target content that meets predetermined conditions while utilizing open data, and enables the immediate provision of results.

[0105] (2) An information processing system as described in (1) above, wherein the evaluation result includes an explanation of the evaluation axis.

[0106] This approach facilitates understanding of the evaluation criteria and allows users to more intuitively judge the value of the target content.

[0107] (3) An information processing system according to (1) or (2) above, wherein in the acquisition step, information relating to a provisional evaluation axis is acquired, and in the identification step, at least one of the evaluation axis and the weighting is identified based on the open data and the provisional evaluation axis.

[0108] This approach allows for the identification of more appropriate evaluation criteria for the target content.

[0109] (4) An information processing system as described in (3) above, wherein in the specific step, a provisional evaluation result is identified that shows the result of evaluating each of the multiple reference contents based on the provisional evaluation axis, and at least one of the evaluation axis and the weighting is identified based on the provisional evaluation axis and the provisional evaluation results for the multiple reference contents.

[0110] This approach allows for the identification of more appropriate evaluation criteria for the target content.

[0111] (5) An information processing system as described in (4) above, wherein the open data includes content information relating to the reference content that satisfies the predetermined conditions, and in the identification step, the information processing system identifies at least one of the evaluation axis and the weighting by modifying the provisional evaluation axis to show characteristics relating to the predetermined conditions based on the provisional evaluation result.

[0112] This approach allows for the identification of more appropriate evaluation criteria for the target content.

[0113] (6) An information processing system as described in (5) above, wherein in the specific step, the system identifies at least one of the evaluation axis and the weights by inputting a prompt to the large language model that includes the provisional evaluation axis, the predetermined conditions, and the provisional evaluation results for a plurality of reference contents, thereby causing the large language model to output at least one of the evaluation axis and the weights.

[0114] This approach allows for the identification of more appropriate evaluation criteria for the target content.

[0115] (7) An information processing system according to any one of (1) to (6) above, wherein in the identification step, the evaluation axis according to the type of target content is identified, and in the output step, the evaluation result is output based on the identified evaluation axis.

[0116] This approach allows for evaluation that is appropriate to the characteristics of the target content, and enables obtaining more appropriate evaluation results.

[0117] (8) An information processing system as described in any one of (1) to (7) above, wherein the reference content and the target content are stores that provide a predetermined product, and the predetermined condition is the region or location where the store provides the product.

[0118] This approach enables region-specific evaluations and provides more practical information.

[0119] (9) An information processing method comprising each step of the information processing system described in any one of (1) to (8) above.

[0120] According to this embodiment, one method can be provided.

[0121] (10) A program that causes a computer to perform each step of the information processing system described in any one of (1) to (8) above.

[0122] According to this embodiment, one program can be provided. Of course, this is not always the case.

[0123] Finally, while various embodiments relating to this disclosure have been described, these are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of symbols]

[0124] 1: Information Processing System 2: Server device 23: Processor 231: Acquisition Department 232: Reception Department 233: Specific part 234: Output section 235: Display Control Unit 236: Arithmetic section 237: Artificial Intelligence Department 3: User terminal

Claims

1. An information processing system, The system comprises at least one processor, the processor configured to perform the following steps by reading a program: In the acquisition step, open data that is publicly available on the internet is acquired, and here, The aforementioned open data includes content information relating to one or more referenced contents, In a specific step, at least one of the following is identified based on the open data and predetermined conditions: an evaluation axis for evaluating a predetermined target content, and weights for each of the evaluation axes. An information processing system that, in the output step, outputs an evaluation result showing the result of evaluating the target content based on at least one of the identified evaluation axis and the weighting.

2. In the information processing system described in claim 1, The aforementioned evaluation results include an information processing system with an explanation of the evaluation axes.

3. In the information processing system described in claim 1, In the acquisition step described above, information regarding a provisional evaluation axis is acquired, An information processing system that, in the specified step, identifies at least one of the evaluation axis and the weighting based on the open data and the provisional evaluation axis.

4. In the information processing system described in claim 3, In the aforementioned specific step, Based on the aforementioned provisional evaluation axis, a provisional evaluation result is identified that shows the evaluation result for each of the multiple reference contents, An information processing system that identifies at least one of the evaluation axis and the weighting based on the aforementioned provisional evaluation axis and the provisional evaluation results for a plurality of the aforementioned reference contents.

5. In the information processing system described in claim 4, The open data includes the content information relating to the reference content that satisfies the predetermined conditions, An information processing system that, in the specified step, identifies at least one of the evaluation axis and the weighting by modifying the provisional evaluation axis to show characteristics relating to the predetermined conditions based on the provisional evaluation results.

6. In the information processing system described in claim 5, An information processing system that, in the specified step, identifies at least one of the evaluation axis and the weights by inputting a prompt to a large language model that includes the provisional evaluation axis, predetermined conditions, and provisional evaluation results for a plurality of reference contents, thereby causing the large language model to output at least one of the evaluation axis and the weights.

7. In the information processing system described in claim 1, In the aforementioned specific step, the evaluation axis corresponding to the type of target content is identified, The output step involves outputting the evaluation results based on the identified evaluation axis.

8. In the information processing system described in claim 1, The aforementioned reference content and the aforementioned target content are stores that provide the specified products, The aforementioned predetermined condition is a system in which the store provides the aforementioned goods in a region or location.

9. Information processing method, A method comprising each step of the information processing system described in any one of claims 1 to 8.

10. It is a program, A program that causes a computer to perform each step of the information processing system described in any one of claims 1 to 8.

Images