system
A system with AI-driven consistency and similarity checks automates the creation of novel and competitive benefit proposals, addressing inefficiencies in service design by accurately comparing in-house and third-party benefits, thereby improving quality and reducing workload.
Patent Information
- Application Number
- JP2025069962
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-04-21
- Publication Date
- 2026-11-02
AI Technical Summary
The process of verifying consistency with existing in-house benefits and comparing with other companies' benefits in service design is cumbersome, inefficient, and subjective, leading to delayed creation of novel and competitive benefit proposals, increased workload, and reduced work efficiency and quality.
A system utilizing an existing benefit information storage unit, a third-party benefit information storage unit, an input receiving unit, and a generation AI that automatically performs consistency checks and similarity evaluations, generating new benefit proposals with high accuracy and efficiency.
Automates complex tasks, reduces workload, ensures efficient and accurate creation of unique benefit proposals, enhances service quality and competitiveness, and provides transparent and reproducible design processes.
Smart Images

Figure 2026175172000001_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The technology of the present disclosure relates to a system.
BACKGROUND ART
[0002] Patent Document 1 discloses a persona chatbot control method executed by at least one processor, the method comprising the steps of: receiving a user utterance; adding the user utterance to a prompt including an instruction sentence associated with a description relating to a character of the chatbot; encoding the prompt; and inputting the encoded prompt into a language model to generate a chatbot utterance responding to the user utterance.
PRIOR ART DOCUMENTS
PATENT DOCUMENTS
[0003]
PATENT DOCUMENT 1
SUMMARY OF THE INVENTION
PROBLEMS TO BE SOLVED BY THE INVENTION
[0004] The problems that this invention aims to solve are that when planning and designing new plans and campaign benefits in the service design department, the process of verifying consistency with existing in-house benefits is cumbersome and inefficient, and furthermore, comparing and examining the benefits of other companies requires a great deal of time and effort. As a result, the creation of novel and highly competitive benefit proposals is delayed, and the improvement of service quality and market competitiveness is hindered. Specifically, it is necessary to individually examine the specifications of existing benefits and other companies' benefits and manually check for duplication, contradictions, and similarities, making the entire design process dependent on individuals and unsystematic, and making it difficult to differentiate benefit proposals and ensure their uniqueness. In addition, there is a high risk of overlooking substantial duplication or competition due to differences in wording and expression of benefit content. Furthermore, the evaluation of consistency and competitiveness becomes subjective, making objective and comprehensive judgment difficult. Moreover, this workload is concentrated on the design staff, leading to a decline in work efficiency and quality. [Means for solving the problem]
[0005] The means for solving the problems according to the present invention includes an existing benefit information storage unit that stores specification information of existing plans and campaign benefits, a third-party benefit information storage unit that stores specification information of third-party benefits, an input receiving unit that receives information on new benefit proposals entered by a user, and a generation system AI that compares the existing benefit specification information stored in the existing benefit information storage unit and the third-party benefit specification information stored in the third-party benefit information storage unit with the new benefit proposal information entered via the input receiving unit, automatically performs consistency checks with existing benefits and similarity evaluations with third-party benefits, and based on the content of the new benefit proposal The system includes an evaluation and proposal unit that generates new benefit proposals evaluated from the perspectives of whether there is any overlap or contradiction with existing benefits, the degree of differentiation from other companies' benefits, novelty, and competitiveness, and outputs the new benefit proposals and evaluation results. The evaluation and proposal unit uses a generative AI to calculate the semantic similarity of benefit specification information through natural language processing, extracts existing benefits or other companies' benefits that have a similarity above a predetermined threshold, and determines the consistency and competitiveness of the new benefit proposals based on the extraction results. This automates the conventionally complicated consistency checks and comparisons with other companies, enabling the efficient and highly accurate creation of novel and competitive benefit proposals. [Brief explanation of the drawing]
[0006] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]
[0007] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.
[0008] First, let's explain the terminology used in the following explanation.
[0009] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), and APU (Accelerated Processing Unit).
[0010] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.
[0011] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.
[0012] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).
[0013] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be A alone, B alone, or a combination of A and B. Furthermore, in the present specification, when three or more matters are expressed in a connected manner using "and / or", the same concept as that for "A and / or B" applies.
[0014] [First Embodiment]
[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.
[0016] As shown in FIG. 1, the data processing system 10 includes a data processing apparatus 12 and a smart device 14. An example of the data processing apparatus 12 is a server.
[0017] The data processing apparatus 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of the "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network), and the like.
[0018] The smart device 14 includes a computer 36, a reception apparatus 38, an output apparatus 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception apparatus 38, the output apparatus 40, the camera 42, and the communication I / F 44 are also connected to the bus 52.
[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user inputs. The touch panel 38A receives a user input from contact of an indicator by detecting the contact of the indicator (for example, a pen, a finger, or the like). The microphone 38B receives a user input by voice by detecting the user's voice. The control unit 46A in the processor 46 transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.
[0020] The output device 40 includes a display 40A, a speaker 40B, and the like, and presents data to the user 20 by outputting the data in a representation form (for example, voice and / or text) perceptible to the user 20. The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs voice in accordance with instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.
[0021] The communication I / F 44 is connected to the network 54. The communication I / Fs 44 and 26 manage the exchange of various types of information between the processor 46 and the processor 28 via the network 54.
[0022] Figure 2 illustrates an example of the main functions of the data processing device 12 and the smart device 14.
[0023] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.
[0024] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.
[0025] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.
[0026] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.
[0027] (Example 1)
[0028] The flow of a specific process in Example 1 will now be described. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."
[0029] (Embodiments for carrying out the present invention)
[0030] Embodiments of the present invention will be described in more detail. The system of the present invention consists of a server and terminals, and each functional unit is implemented on the server or terminal according to its role. The following will provide a very detailed explanation, including the configuration of each unit, the processing flow, the method of utilizing the generative AI, specific usage examples, operation and expansion examples, and aspects of security and operational management.
[0031] First, the existing benefits information storage unit will be implemented as a database on the server. This database will store the specifications for all plans and campaign benefits that the service design department has designed and operated to date. For example, detailed information such as timing, target, conditions, benefit content, provision period, and exclusion conditions for benefits such as "2023 Spring New Life Support Campaign," "Continuance Benefits for Long-Term Users," "Student-Only Discount Plan," and "Data Sharing Benefits for Family Contracts" will be stored as structured data (e.g., JSON, CSV, RDBMS) or unstructured data (e.g., PDF, Word, text files). Examples of benefit specifications include "100 points awarded monthly to monthly users," "First month free benefit for new subscribers," "Coupon distribution in the birth month," and "Discount for long-term users to prevent cancellation." This data will contain multifaceted attribute information such as benefit name, target audience, provision conditions, benefit content, award method, award timing, exclusion conditions, past implementation results, and usage data.
[0032] The third-party benefits information storage unit will also be implemented as a database on the server. This will store information on the benefits offered by competitors. For example, "cashback for new subscribers from competitor A," "friend referral campaign from competitor B," "limited-time data bonus from competitor C," and "birthday month benefits from competitor D," collected from industry research departments and external research companies, will be stored as structured and unstructured data. This information may be automatically collected or manually registered from competitors' service websites, advertising materials, public documents, review sites, industry news, etc. For example, it is also conceivable to automatically collect information using web crawlers or scraping tools, and then store it in the database after automatic summarization and classification by AI.
[0033] The input reception section is implemented on the terminal side, and new benefit proposals are entered through the user interface (UI) of personal computers, tablet devices, smartphones, etc., used by service design personnel. The UI offers a variety of input methods, including text input fields, voice input, template selection, reuse of past benefit proposals, and conditional selection forms. For example, benefit proposals can be entered through free-form text or semi-automatic generation, such as "grant 500 points to new subscribers with their first billing," "distribute special coupons to existing users during their birth month," "add extra data capacity to all family members every month," or "offer unlimited data usage to students during the summer vacation period." The input content is transmitted from the terminal to the server via secure communication (e.g., SSL / TLS).
[0034] The evaluation and proposal section is implemented on the server side and is based on generative AI. Generative AI is composed of a combination of elements such as large-scale language models, natural language understanding models, semantic vector computation models, knowledge graphs, and similarity calculation engines. First, the input new benefit proposal is analyzed using natural language processing to extract elements such as benefit content, target audience, provision conditions, grant method, timing, and exclusion conditions. For example, from the input "Offer a free first month benefit to new subscribers," the main elements "new subscribers," "first month," and "free" are extracted.
[0035] Next, the generative AI semantically compares all the data in the existing benefits information storage unit with the proposed new benefits. It compares not only simple keyword matching, but also semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching timing of provision, and essential overlap of benefits. For example, while "monthly point accrual" and "monthly coupon distribution" are different expressions, the AI automatically determines that their incentive effects on users are similar. Furthermore, even if the expressions and conditions differ, such as "first month free for new subscribers" and "discount on the first billing amount for new subscribers," the AI can extract essential overlaps and contradictions in the benefits. Additionally, even if the definition of the target audience differs, such as "student-only discount" and "discount for those under 25," the AI recognizes overlaps in age groups and attributes, and extracts overlap risks and points of differentiation.
[0036] The same process is used for matching with the information storage unit for other companies' benefits. The generation AI accurately determines the similarity of the benefit structure and user benefits, even if the wording and conditions differ, such as "Company A's friend referral campaign" and "a benefit that awards points to both the referrer and the referred person that our company is considering." Furthermore, even if the timing and method of awarding differ, such as "a cashback offer by Company B for a limited time" and "a limited-time point increase offer that our company is considering," the AI evaluates the similarity from the perspective of the appeal of the benefits and market competitiveness. For example, if "Company C distributes coupons during the birth month, but our company is considering a benefit that increases data capacity during the birth month," the AI automatically determines the differences in incentives, the risk of overlap, and the potential for differentiation between the two.
[0037] Based on these matching results, the generation AI calculates a consistency score with existing benefits, a similarity score with competitors' benefits, and a novelty / competitiveness score. For example, if the similarity with existing benefits is high, it evaluates it as "there is a risk of overlap with existing benefits," if the similarity with competitors' benefits is high, it evaluates it as "similar to competitors' benefits, differentiation needs to be considered," and if both are low, it evaluates it as "high novelty and competitiveness." Furthermore, the AI breaks down the content of the proposed benefits and automatically generates specific improvement suggestions, such as "the benefit content partially overlaps with existing benefits, but differentiation is possible by changing the target audience or conditions of provision," or "it is similar to competitors' benefits, but uniqueness can be ensured by devising the method and timing of provision."
[0038] The evaluation and proposal department automatically generates evaluation comments and proposals using natural language generation capabilities via AI and sends them back to the terminal. For example, it outputs specific and easy-to-understand evaluation comments such as, "The content partially overlaps with the existing 'long-term contract discount,' so the terms of provision need to be differentiated," "It is similar to other companies' 'friend referral bonuses,' but differentiation is possible by providing incentives not only to the referrer but also to the referred person," and "There are no similar bonuses existing or offered by other companies, so it is highly novel and competitive." Furthermore, the AI can also present the reasons for the evaluation, excerpts of existing and competitor bonuses that serve as the basis, a similarity score, and details of improvement proposals.
[0039] On the terminal side, the person in charge can instantly check the evaluation comments and suggestions from the AI. For example, they can use the evaluation comments to revise the content of the benefit proposal, emphasize the differentiating points, or adopt the benefit proposal that is judged to be highly innovative, thereby improving the efficiency and quality of the design work. Furthermore, the person in charge can also input feedback on the AI's evaluation results. For example, they can input comments such as "This benefit is not actually duplicated" or "This differentiating point is important," allowing for continuous improvement of the AI's evaluation algorithm and suggestions.
[0040] Furthermore, the databases for existing and competitor benefit information are regularly updated. For example, when a new campaign is implemented or when new benefits from competitors are discovered through market research, this information is added and updated as needed, enabling evaluations and suggestions to always be based on the latest information. The generative AI can maintain advanced semantic understanding and generation capabilities that reflect trends in benefit specifications and industry developments through updates to its learning data and fine-tuning.
[0041] Furthermore, it is possible to implement a function that saves the history of AI-generated evaluations and suggestions in a database, allowing users to refer to past reward proposals and evaluation results. This enables quality control and knowledge sharing throughout the entire design process, including trend analysis of previously rejected and accepted reward proposals, tracking of improvement history, and understanding the design tendencies of each person in charge.
[0042] In terms of security, to protect confidential information such as reward specifications, evaluation results, and user input, communication between servers and terminals is encrypted (SSL / TLS), and measures are taken to manage database access rights, record audit logs, and prevent misuse of AI models. For example, it is possible to restrict the reward information that each user can access, or to introduce digital signatures and hash value management to prevent tampering with AI evaluation results.
[0043] In terms of operation, for example, it is possible to perform periodic model retraining and parameter tuning to improve the accuracy of AI evaluations, and to optimize the AI's decision criteria by utilizing user feedback. Furthermore, the evaluation and proposal process for incentives can be linked with a workflow management system, automating business flows such as approval, rejection, revision, and re-evaluation.
[0044] As described above, the embodiment of the present invention has a configuration in which the existing benefit information storage unit and the third-party benefit information storage unit are implemented on the server side, the input reception unit is implemented on the terminal side, and the evaluation and proposal unit is implemented on the server side. The generation AI is incorporated as an essential element in the server-side evaluation and proposal unit and plays a central role in all major processes, including calculating the semantic similarity of benefit specifications, extracting overlaps, contradictions, and differentiating points, automatically generating evaluation comments, history management, feedback reflection, security management, and operational optimization. This enables the efficient and highly accurate automation of complex tasks that previously relied on manual processes, such as consistency checks, comparisons with other companies, evaluation comment creation, knowledge sharing, and quality control. This simultaneously reduces the workload of the service design department, strengthens competitiveness, improves service quality, and ensures transparency and reproducibility of the design process.
[0045] (System Configuration)
[0046] The system according to this embodiment comprises an existing benefits information storage unit, a third-party benefits information storage unit, an input reception unit, an evaluation / proposal unit, a communication control unit, a history management unit, a user interface unit, and a security management unit. The existing benefits information storage unit is a database on a server that stores specifications for all plans and campaign benefits that the service design department has designed and operated in the past. For example, it stores data such as the timing, target, conditions, benefit content, provision period, exclusion conditions, past implementation results, and usage results as structured or unstructured data, and manages them as PDFs, text files, spreadsheet files, and database records. For example, if a proposed new benefit is "a free first month benefit for new subscribers," the existing benefits information storage unit can extract similar benefits that have been implemented in the past, such as "a discount on the first billing amount for new subscribers" or "points awarded to new subscribers," and provide them as comparison targets. Furthermore, the existing benefits information storage unit manages the history of changes to benefits and discontinued benefits, and supports the decision-making of design personnel by allowing them to refer to past failure and success stories.
[0047] The Competitor Benefits Information Storage Unit is a database on a server that stores information on the benefits offered by competitors. For example, it stores information collected from industry research departments and external research companies as structured and unstructured data, such as "Competitor A's cashback for new subscribers," "Competitor B's referral campaign," "Competitor C's limited-time data bonus," and "Competitor D's birthday month benefit." The Competitor Benefits Information Storage Unit can automatically collect information from competitors' service sites and advertising materials using, for example, web crawlers and scraping tools, and store it in the database after automatic summarization and classification by AI. Furthermore, the Competitor Benefits Information Storage Unit provides comparative data to evaluate similarity from the perspective of the appeal and market competitiveness of benefits, even when the timing and method of awarding differ, such as "Competitor B's limited-time cashback" and "our company's planned limited-time point bonus." In addition, the Competitor Benefits Information Storage Unit manages the change history of competitors' benefit content and completed campaigns, providing basic data for understanding industry-wide benefit trends.
[0048] The input reception unit has the function of inputting new benefit proposals through user interfaces such as personal computers, tablet terminals, and smartphones used by service design personnel. It offers a variety of input methods, such as text input fields, voice input, template selection, conditional selection forms, and a function to reuse past benefit proposals. The input reception unit allows users to input benefit proposals through free-form text or semi-automatic generation, such as "granting 500 points to new subscribers on their first billing cycle," "distributing special coupons to existing users during their birth month," "adding data capacity to all family members every month," or "offering unlimited data usage to students during the summer vacation period." The input reception unit securely transmits the input content from the terminal to the server and also has functions such as automatic saving of input content, saving as a draft, automatic completion of input content, and automatic detection of input errors.
[0049] The evaluation and proposal department is implemented on the server side, with a generating AI at its core. It semantically matches all data from the input new benefit proposals, existing benefit information storage, and competitor benefit information storage using natural language processing technology. It extracts elements such as benefit content, target audience, provision conditions, grant method, timing, and exclusion conditions, and compares them from multiple perspectives, including not only simple keyword matching but also semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching provision timing, and essential overlap of benefit content. The evaluation and proposal department automatically determines whether the incentive effect on users is similar, even beyond differences in expression such as "monthly point grants" and "monthly coupon distribution." It also extracts essential overlaps and contradictions in benefit content even with different expressions and conditions such as "first month free for new subscribers" and "discount on the first billing amount for new subscribers." Furthermore, even when the definition of target audience differs, such as "student-only discount" and "discount for those under 25," it recognizes overlaps in age groups and attributes and extracts overlap risks and differentiation points. Furthermore, the Evaluation and Proposal Department accurately determines the similarity of the reward structure and user benefits when comparing it with the Competitor Reward Information Storage Department, such as comparing "Competitor A's friend referral campaign" with "a reward that awards points to both the referrer and the referred person that our company is considering." It also evaluates the similarity from the perspective of the appeal and market competitiveness of the reward, even when the timing and method of awarding differ, such as comparing "Competitor B's limited-time cashback" with "a limited-time point increase that our company is considering." Based on these comparison results, the Evaluation and Proposal Department calculates a consistency score with existing rewards, a similarity score with other companies' rewards, and a novelty / competitiveness score. It makes evaluations such as "there is a risk of overlap with existing rewards," "it is similar to other companies' rewards, differentiation needs to be considered," and "it has high novelty and competitiveness," and breaks down the content of the reward proposal to automatically generate specific improvement suggestions, such as "the reward content partially overlaps with existing rewards, but differentiation is possible by changing the target audience and conditions of provision" and "it is similar to other companies' rewards, but uniqueness can be ensured by devising the method and timing of awarding." The Evaluation and Proposal Department automatically generates evaluation comments and proposals using natural language generation capabilities via AI, sends them back to the user's device, and also presents the reasons for the evaluation, excerpts of existing and competitor benefits that serve as the basis, similarity scores, and details of improvement proposals.
[0050] The communication control unit controls the transmission and reception of data between terminals and servers, and exchanges new reward proposal information from the input reception unit and evaluation results and proposal content from the evaluation and proposal unit over a secure communication channel. The communication control unit prevents information leakage using encryption technologies such as SSL and TLS, and supports a variety of communication scenarios, such as real-time transmission of input content, immediate return of evaluation results, automatic synchronization when the database is updated, and simultaneous access control from multiple terminals. The communication control unit ensures the consistency and real-time nature of communication even when, for example, a design engineer inputs a reward proposal from a tablet terminal while on the go and receives evaluation results on the spot, or when multiple personnel evaluate different reward proposals simultaneously.
[0051] The History Management Department centrally manages past benefit proposal input history, evaluation and proposal history, acceptance and rejection history, and operation history for each person in charge on a server. The History Management Department enables, for example, trend analysis of previously rejected and accepted benefit proposals, tracking of improvement history, understanding of design trends for each person in charge, and analysis of changes in evaluation comments, thereby realizing quality control and knowledge sharing for the entire design process. The History Management Department manages specific cases in chronological order, such as "The student campaign for summer 2022 was rejected because it overlapped with existing benefits" and "The family contract benefit for spring 2023 was accepted because it was evaluated as differentiating itself from competitors," allowing people in charge to design new benefit proposals while referring to past cases.
[0052] The user interface unit operates on the terminal and provides screens for designers to input new benefit proposals, view evaluation comments and proposals from the evaluation and proposal unit, and refer to information from the history management unit. The user interface unit has a variety of screen configurations, such as input forms, evaluation result display screens, past history reference screens, feedback input screens, and improvement suggestion display screens, to improve the work efficiency of the person in charge. The user interface unit allows users to, for example, revise the content of benefit proposals based on evaluation comments, emphasize differentiating points, and input feedback on AI evaluation results.
[0053] The Security Management Department implements measures to protect confidential information such as reward specifications, evaluation results, and user input, including encrypting communications between servers and terminals, managing database access rights, recording audit logs, and preventing the misuse of AI models. For example, the Security Management Department restricts the reward information that each user can access, introduces electronic signatures and hash value management to prevent tampering with AI evaluation results, and operates firewalls and intrusion detection systems to prevent unauthorized access and information leaks from external sources.
[0054] Specific examples of prompt statements to be fed into the generating AI when implementing the present invention include, for example, "For the following proposed new benefits, semantically compare it with all benefit specification information stored in the existing benefits information storage unit and the competitor's benefits information storage unit, evaluate it from the perspectives of duplication, contradictions, similarities, points of differentiation, novelty, and competitiveness, and generate an easy-to-understand comment along with the reasons for the evaluation. Proposed new benefits: 'Grant 500 points to new subscribers upon their first billing.'" "Compare the following proposed benefits with existing benefits and competitor benefits, analyze in detail the duplication, similarities, uniqueness, and improvement suggestions, and output an evaluation comment. Proposed benefits: 'Add data capacity to all family subscribers every month.'" "Refer to past 'birthday month benefits' stored in the existing benefits information storage unit and 'birthday month coupon distribution' stored in the competitor's benefits information storage unit, evaluate the points of differentiation and novelty of the following proposed new benefits 'Increase data capacity in the birth month,' and present specific improvement suggestions." These prompt statements are used as instructions for the generating AI to perform evaluation and proposal of benefit proposals with high accuracy and from multiple perspectives.
[0055] Thus, the system according to this embodiment comprises an existing benefits information storage unit, a third-party benefits information storage unit, an input reception unit, an evaluation and proposal unit, a communication control unit, a history management unit, a user interface unit, and a security management unit. Each unit is designed to handle specific and diverse cases, and with the utilization of generation AI at its core, it streamlines and enhances the benefits proposal design work of the service design department, thereby strongly supporting the development of competitive services.
[0056] (Implementation Steps)
[0057] Step 1: Collect and register existing and third-party benefit information. First, the service design department and industry research department collect specifications for all plans and campaign benefits previously implemented by the company, as well as information on benefit specifications from competitors, and register them in the existing benefit information storage unit and the competitor benefit information storage unit. For example, information on the company's own benefits, such as past "retention benefits for long-term users," "student-only discount plans," and "data sharing benefits for family contracts," as well as information on competitor benefits, such as "cashback for new customers from competitor A" and "friend referral campaign from competitor B," are stored in the database as structured and unstructured data. The information is registered using various methods, including manual input, file import, and automated collection by web crawlers, and has multifaceted attribute information such as benefit name, target audience, terms of provision, benefit content, grant method, timing, exclusion conditions, and past implementation record.
[0058] Step 2: Enter new benefit proposal Service designers input proposed new benefits through the user interface of their terminals. Input methods are diverse, including text input, voice input, template selection, conditional forms, and reuse of past benefit proposals. Examples include "granting 500 points to new subscribers upon their first billing," "adding extra data capacity to all family members each month," and "offering unlimited data during summer vacation exclusively for students." The input content is securely transmitted from the terminal to the server.
[0059] Step 3: Semantic matching and analysis of the proposed benefits using a generating AI. The Evaluation and Proposal Department uses a generating AI to semantically match and analyze all benefit specification information stored in the input new benefit proposal, existing benefit information storage unit, and third-party benefit information storage unit using natural language processing technology. The generating AI extracts elements such as benefit content, target audience, provision conditions, grant method, timing, and exclusion conditions, and compares them from multiple perspectives, including not just keyword matching, but also semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching provision timing, and essential overlap of benefit content. For example, it automatically determines if the incentive effect on users is similar, even if the expression is different, such as "monthly point grants" and "monthly coupon distribution," and extracts essential overlaps and contradictions in benefit content even with different expressions and conditions, such as "first month free for new subscribers" and "discount on the first billing amount for new subscribers." Specific examples of prompt text to be fed into the generating AI include: "For the following proposed new benefit, semantically compare it with all benefit specification information stored in the existing benefit information storage unit and the other company's benefit information storage unit, evaluate it from the perspectives of duplication, contradictions, similarities, points of differentiation, novelty, and competitiveness, and generate an easy-to-understand comment along with the reasons for the evaluation. Proposed new benefit: '500 points will be awarded to new subscribers upon their first billing.'"
[0060] Step 4: Automatic generation of evaluation comments and proposals by AI. The generating AI calculates consistency scores with existing benefits, similarity scores with competitors' benefits, and novelty / competitiveness scores based on the matching and analysis results. It then breaks down the content of the proposed benefit and automatically generates evaluation comments and improvement suggestions using natural language generation. For example, it outputs specific and easy-to-understand evaluation comments such as, "The content partially overlaps with the existing 'discount for long-term contract holders,' so the terms of provision need to be differentiated," "It is similar to a competitor's 'referral program,' but differentiation is possible by providing incentives not only to the referrer but also to the referred," and "There are no similar benefits existing or offered by competitors, so it is highly novel and competitive." A concrete example of a prompt to be fed into the generating AI is, "Compare the following proposed benefit with existing benefits and competitors' benefits, analyze in detail the overlap, similarity, uniqueness, and improvement suggestions, and output evaluation comments. Proposed benefit: 'Add data capacity to all family members every month.'"
[0061] Step 5: Output of evaluation results and proposed content, and confirmation and revision by the person in charge. The evaluation and proposal department outputs evaluation comments and proposal content, automatically generated by AI, to terminals via the user interface department, allowing staff to review them immediately. Staff can use the evaluation comments to revise the content of the reward proposal, emphasize differentiating points, and provide feedback on the AI's evaluation results. For example, they can change the terms of the reward proposal based on the evaluation comments, add differentiating points from competitors, or redesign the reward content to avoid duplication with existing rewards, thereby improving the efficiency and quality of design work.
[0062] Step 6: History management and knowledge sharing The History Management Department centrally manages past benefit proposal input history, evaluation / proposal history, acceptance / rejection history, and operation history for each person in charge on a server, enabling people in charge to design new benefit proposals while referring to past examples. For example, it manages specific examples in chronological order, such as "The student campaign for summer 2022 was rejected because it overlapped with existing benefits" and "The family contract benefit for spring 2023 was accepted because it was evaluated as differentiating itself from competitors," enabling quality control and knowledge sharing throughout the entire design process.
[0063] (Specific use cases)
[0064] For example, let's assume that when the service design department plans a new campaign benefit, the design engineer inputs a benefit proposal such as "grant 500 points to new subscribers upon their first billing" using the terminal's user interface. The input reception unit receives this benefit proposal and transmits it to the server via the communication control unit. On the server side, the existing benefit information storage unit stores past benefit specifications such as "first month free benefit for new subscribers," "discount on the first billing amount for new subscribers," and "points awarded to new subscribers," while the competitor benefit information storage unit stores information on competing benefits such as "cashback for new subscribers from competitor A," "points awarded to new subscribers from competitor B," and "first month free for new subscribers from competitor C." The evaluation and proposal unit uses a generation AI to semantically compare the input new benefit proposal with all the benefit specification information in the existing benefit information storage unit and the competitor benefit information storage unit. The generating AI extracts elements such as the content of the benefit, the target audience, the terms of provision, the method of granting, the timing, and the exclusion conditions, and accurately determines the similarity of the incentive effect and user benefits between, for example, "points awarded on the first billing cycle" and "discount on the first billing cycle" or "first month free." Furthermore, it comprehensively compares the differences in granting methods, timing, and content between "cashback for new subscribers" and "points awarded for new subscribers" stored in the third-party benefit information storage unit, and extracts points of differentiation and risks of duplication.
[0065] The generating AI executes processing by reading a prompt message such as the following: "For the following proposed new benefit, semantically compare it with all benefit specification information stored in the existing benefit information storage unit and the competitor's benefit information storage unit, evaluate it from the perspectives of duplication, contradiction, similarities, points of differentiation, novelty, and competitiveness, and generate an easy-to-understand comment along with the reasons for the evaluation. Proposed new benefit: 'Award 500 points to new subscribers upon their first billing.'" Based on this prompt, the generating AI comprehensively evaluates the degree of duplication and similarity with existing benefits, the degree of differentiation from competitors' benefits, novelty, and competitiveness, and automatically generates evaluation comments such as, "The incentive effect is similar to the existing benefit 'First month free for new subscribers,' but it can be differentiated because the method of awarding points is different," and "It provides the same economic benefit as competitors' 'Cashback for new subscribers,' but the point awarding format is expected to have an effect in promoting usage."
[0066] Furthermore, if, for example, a designer inputs a benefit proposal such as "adding extra data capacity to all family members every month," the generating AI will analyze the differences in the scope of eligible recipients, benefit content, and granting method between this proposal and existing benefit information stored in the benefit information memory, such as "data sharing benefit for family members," "discount for family members," and "point grant for family members." It will also evaluate the uniqueness of the terms and incentives compared to "data increase for family members from company D" and "discount for family members from company E" stored in the other company's benefit information memory. A concrete example of a prompt message to be fed into the generating AI is: "Compare the following benefit proposal with existing benefits and other companies' benefits, analyze in detail the overlap, similarity, uniqueness, and improvement suggestions, and output evaluation comments. Benefit proposal: 'Adding extra data capacity to all family members every month'."
[0067] Furthermore, if, for example, a proposed benefit such as "unlimited data usage for students during the summer vacation period" is entered, the generating AI semantically compares existing benefit information storage units such as "student-only discount plan," "student-only data increase," and "summer-only campaign" with other companies' benefit information storage units such as "Company F's summer campaign for students" and "Company G's unlimited data usage for students," and extracts points of overlap and differentiation from the perspectives of target audience, period, benefit content, and provision conditions. The generating AI automatically generates evaluation comments such as, "This benefit has a different provision period from the existing 'student-only discount plan,' and different benefit content from other companies' 'summer campaigns for students,' making it highly innovative."
[0068] Thus, the system of the present invention comprehensively compares and evaluates the input new benefit proposal with various benefit specification information stored in the existing benefit information storage unit and the competitor benefit information storage unit using a generating AI. It comprehensively determines overlaps, contradictions, similarities, differentiating points, novelty, and competitiveness, and automatically generates easy-to-understand comments and improvement suggestions along with the reasons for the evaluation. As a result, the service design department can efficiently and accurately perform consistency checks and competitor comparisons, which were previously cumbersome and reliant on individual expertise, and quickly realize the creation of highly competitive new benefit proposals.
[0069] (Application Example 1) The flow of a specific process in Application Example 1 will be explained below. Each part of the system described below is implemented by a data processing device 12 and a smart device 14. The data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."
[0070] (Embodiments for carrying out the present invention)
[0071] Embodiments of the present invention will be described in more detail and in great detail, covering a wide range of topics, using a promotion strategy design support system for e-commerce sites as an example. The system of this embodiment comprises an existing company strategy information storage unit, a competitor strategy information storage unit, an input reception unit, an evaluation and proposal unit, a user interface unit, a communication control unit, a history management unit, and a security management unit, with each unit working in cooperation with the others.
[0072] The existing in-house promotion information storage unit stores specification information for all promotional measures implemented by the e-commerce site operation department in the past and present as a database on the server. For example, it stores diverse attribute information such as the name of the measure, target audience, terms of provision, benefit details, method of provision, usage conditions, timing, exclusion conditions, implementation results, usage data, start and end dates of the measure, person in charge of the measure, purpose of the measure, KPIs of the measure, results of the measure's effectiveness measurement, improvement history of the measure, and reasons for discontinuing the measure, such as "10% discount coupon for first-time buyers," "Double points when member rank is upgraded," "Free shipping campaign for a limited time," "Birthday points gift for new member registration," "Special sale for gold members only," "Coupon gift for both parties when referring a friend," "5x points campaign for all products," "Free shipping on first purchase for new member registration," "Discount limited to specific brands," "Point rewards limited to outlet products," "App-exclusive coupon distribution," "Special discount for repeat buyers," "Benefits given via birthday email," and "Weekend-only time sale," as structured data (e.g., CSV, RDBMS, NoSQL) or unstructured data (e.g., text file, PDF, Word), including the measure name, target audience, terms of provision, benefit details, method of provision, usage conditions, timing, exclusion conditions, implementation results, usage data, start and end dates of the measure, person in charge of the measure, purpose of the measure, KPIs of the measure, results of the measure's effectiveness measurement, improvement history of the measure, and reasons for discontinuing the measure. For example, the "2023 Spring New Life Support Campaign" has detailed information such as "20% discount coupon usable only on the first purchase for new member registrants", "Period: March 1, 2023 to April 30, 2023", "Target: New member registrants", "Usage conditions: Only for the first purchase", "Method of issuance: Coupon code sent via email upon completion of member registration", "Implementation results: 5,000 registrants, coupon usage rate 30%", "KPI: Number of new member registrations, first purchase rate", and "Effect measurement: Number of new member registrations increased by 150% compared to the same period last year". Furthermore, the existing in-house policy information storage unit manages policies that have been discontinued or temporarily suspended in the past, and by making it possible to refer to past failures, successes, and improvement histories, such as "2022 Summer Outlet Limited Point Redemption Campaign (Reason for discontinuation: Low usage rate)" and "2021 Winter App Limited Coupon Distribution (Improvement history: Usage rate improved due to increased coupon amount)", it provides multifaceted support for decision-making by those in charge.
[0073] The Competitor Strategy Information Storage Unit stores information on competitors' promotional strategies, collected using industry research departments, external research companies, web crawlers, scraping tools, etc., as a database on the server. For example, it stores strategy specifications with similar attribute information, such as "Competitor A: 500 points awarded for new member registration," "Competitor B: Weekend-only time sale," "Competitor C: Friend referral campaign," "Competitor D: Coupon awarded for posting reviews," "Competitor E: Free shipping during your birthday month," "Competitor F: Point rewards based on member rank," "Competitor G: Double points for app users only," "Competitor H: Discounts on outlet products only," "Competitor I: Free shipping for new members only," "Competitor J: Special benefits via birthday email," and "Competitor K: Special discounts for repeat customers." The Competitor Strategy Information Storage Unit utilizes diverse information sources, such as automated collection by web crawlers, manual registration of industry news and advertising materials, extraction of information from review sites and social media, regular monitoring of competitors' campaign pages, and information gathering at industry events and exhibitions, to comprehensively manage information such as strategy content, target audience, provision conditions, distribution methods, usage conditions, timing, exclusion conditions, implementation results, strategy start and end dates, strategy objectives, strategy KPIs, strategy effectiveness measurement results, strategy improvement history, and reasons for strategy discontinuation. Furthermore, the Competitor Strategy Information Storage Unit also manages the change history of competitors' strategies and completed campaigns, providing basic data for understanding industry-wide promotion trends and the competitive environment, such as "Competitor B's weekend-only time sale (April 2022 - March 2023, reason for discontinuation: increased costs)" and "Competitor C's friend referral campaign (increased referrer incentives in summer 2021)."
[0074] The input reception unit has the function of inputting new promotional measures through user interfaces such as personal computers, tablet terminals, and smartphones used by service design personnel. There are various input methods, including text input fields, voice input, template selection, conditional selection forms, a function to reuse past measures, a drag-and-drop combination of measure elements, AI-powered auto-completion, and a measure simulation function. For example, measures such as "Give new members a 20% discount coupon usable only on their first purchase," "Offer free shipping during the birth month exclusively for gold members," "Give points usable on the next purchase to all reviewers," "Give coupons to both parties for friend referrals," "5x points on all products for a limited time," "Free shipping on first purchases exclusively through the app," "Additional discounts exclusively for outlet products," "Provide benefits via birthday email," and "Special discounts for repeat customers" can be entered through free-form text or semi-automatic generation. The input content is transmitted from the terminal to the server via secure communication, and the system also includes functions such as automatic saving of input content, draft saving, automatic completion of input content, automatic detection of input errors, real-time preview of input content, and prediction of the effectiveness of measures based on the input content.
[0075] The evaluation and proposal department is implemented on the server side, with a generative AI at its core. It semantically matches and analyzes all data from the input new policy proposals, existing company policy information storage, and competitor policy information storage using natural language processing technology. The generative AI extracts elements such as policy content, target audience, provision conditions, granting method, usage conditions, timing, exclusion conditions, implementation results, policy objectives, policy KPIs, policy effectiveness measurement results, policy improvement history, and reasons for policy discontinuation. It performs comparisons from multiple perspectives, not just simple keyword matching, but also semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching provision timing, essential overlap of benefit content, economic value of incentives, impact on user experience, policy cost structure, policy risk factors, and policy legal and regulatory compliance. For example, it can accurately determine the overlap or similarity of essential incentives and user benefits even with different expressions and conditions, such as "discount for first-time buyers" and "coupon for new member registrants," or "birthday month point gift" and "free shipping during birthday month." Furthermore, the system analyzes the differences in reward methods and incentives from multiple perspectives, such as "granting points usable on future purchases to all reviewers" and "receiving coupons for submitting reviews," to extract points of differentiation and risks of overlap. For example, if a new policy proposal such as "offering free shipping during the birth month exclusively for gold members" is entered, it will be compared with existing company policy information storage items such as "special sale for gold members only," "birth month point gift," and "free shipping during birth month (previously implemented)" to comprehensively evaluate overlap in target audiences, differences in benefit content, differences in provision conditions, past implementation results, and effectiveness measurement results. It will also be compared with competitor policy information storage items such as "competitor E's free shipping during birth month" and "competitor J's benefit provision via birthday email" to evaluate points of differentiation, novelty, and competitiveness compared to competitors.
[0076] The Evaluation and Proposal Department comprehensively analyzes the consistency score with existing measures, similarity score with competing measures, novelty and competitiveness score, economic impact of the measure, impact on user experience, risk assessment of the measure, and legal and regulatory compliance assessment calculated by the generative AI, and automatically generates evaluation comments and proposals for the final new measure using natural language generation functionality. For example, it outputs specific and easy-to-understand evaluation comments such as: "There is some overlap with the existing measure '10% discount coupon for first-time buyers,' but it can be differentiated by the different discount rate and usage conditions," "The incentive effect is similar to competitor A's point reward measure, but it has uniqueness in the form of a discount coupon," "There are no similar measures among existing or competing companies, so it has high novelty and competitiveness," "The past 'free shipping during birthday month' measure had a low utilization rate, but by limiting it to gold members, we can reduce costs while expecting to improve the satisfaction of loyal customers," and "Giving points to all reviewers is more cost-effective than competitor D's coupon distribution measure, and is expected to improve the repurchase rate." Furthermore, the AI can also present the reasons for the evaluation, excerpts of existing and competing measures that serve as the basis, similarity scores, details of improvement proposals, predictions of the economic impact of the measures, risk assessments, and results of legal and regulatory compliance checks.
[0077] The user interface unit operates on the terminal and provides screens for design engineers to input new policy proposals, view evaluation comments and proposals from the evaluation and proposal unit, and refer to information from the history management unit. For example, it has a variety of screen configurations, such as input forms, evaluation result display screens, past history reference screens, feedback input screens, improvement proposal display screens, policy effect simulation screens, policy cost analysis screens, policy risk evaluation screens, and legal and regulatory compliance check screens, thereby increasing the work efficiency of the person in charge. The person in charge can revise the content of policy proposals based on evaluation comments, emphasize differentiating points, input feedback on AI evaluation results, run policy effect simulations, and analyze policy costs and risks.
[0078] The communication control unit controls the transmission and reception of data between terminals and servers, and exchanges new policy proposal information from the input reception unit and evaluation results and proposal content from the evaluation and proposal unit over a secure communication channel. It prevents information leakage using encryption technologies such as SSL and TLS, and supports a variety of communication scenarios, including real-time transmission of input content, immediate return of evaluation results, automatic synchronization when the database is updated, simultaneous access control from multiple terminals, automatic retransmission in case of communication failure, and recording of communication logs.
[0079] The History Management Department centrally manages past policy proposal input history, evaluation / proposal history, adoption / rejection history, operation history for each person in charge, policy improvement history, policy effectiveness measurement results, reasons for policy discontinuation, and feedback history on a server, enabling staff to design new policy proposals while referring to past examples. For example, it manages specific examples in chronological order, such as "The campaign for new members in the summer of 2022 was rejected because it overlapped with an existing policy," "The benefits for gold members in the spring of 2023 were adopted because they were evaluated as differentiating themselves from competitors," "The app-exclusive coupon distribution in the winter of 2021 saw increased usage due to an increase in coupon value," and "The outlet-exclusive point reward campaign in the fall of 2020 was discontinued due to low usage," enabling quality control of the entire design process and knowledge sharing.
[0080] The Security Management Department takes measures to protect confidential information such as policy specifications, evaluation results, and input from personnel, including encrypting communications between servers and terminals, managing access rights to databases, recording audit logs, preventing the misuse of AI models, implementing firewalls and intrusion detection systems to prevent unauthorized access and information leaks from external sources, using digital signatures and hash value management to prevent tampering with AI evaluation results, and restricting the policy information that each person can access.
[0081] Furthermore, it is possible to implement a function that saves the history of AI-generated evaluations and suggestions in a database, allowing users to refer to past policy proposals and evaluation results. This enables quality control and knowledge sharing throughout the entire design process, including trend analysis of rejected and adopted policy proposals, tracking of improvement history, understanding design tendencies of each person in charge, analysis of policy effectiveness measurement results, analysis of policy cost structure and risk factors, and management of policy legal and regulatory compliance history. It is also possible to perform periodic model retraining and parameter tuning to improve the accuracy of AI evaluations, and to optimize the AI's judgment criteria using user feedback.
[0082] As described above, the embodiment of the present invention comprises an existing in-house policy information storage unit, a competitor policy information storage unit, an input reception unit, an evaluation / proposal unit, a user interface unit, a communication control unit, a history management unit, and a security management unit. With a generative AI at its core, it plays a central role in all major processes, including calculating the semantic similarity of policy specifications, extracting overlaps, contradictions, and differentiating points, automatically generating evaluation comments, history management, feedback reflection, security management, operational optimization, policy effect simulation, policy cost analysis, policy risk assessment, and legal / regulatory compliance checks. This enables the efficient and highly accurate automation of complex tasks that previously relied on manual processes, such as consistency checks, competitor comparisons, evaluation comment creation, knowledge sharing, quality control, policy effect analysis, risk assessment, and legal compliance checks. This simultaneously reduces the workload of the e-commerce site operation department, strengthens competitiveness, improves service quality, and ensures transparency and reproducibility of the design process.
[0083] (System Configuration)
[0084] The system according to this embodiment comprises an existing company policy information storage unit, a competitor policy information storage unit, an input reception unit, an evaluation / proposal unit, a user interface unit, a communication control unit, a history management unit, and a security management unit. The existing company policy information storage unit is a unit that stores specification information of all promotional policies implemented by the e-commerce site operation department in the past and present as a database on the server. For example, it stores policy names, target audiences, terms of provision, benefit details, method of provision, usage conditions, timing, exclusion conditions, implementation results, usage data, policy start and end dates, policy person in charge, policy objectives, policy KPIs, policy effectiveness measurement results, policy improvement history, policy discontinuation, policy abolition, etc. Diverse attribute information, such as reasons, is stored as structured and unstructured data. For example, the "2023 Spring New Life Support Campaign" has detailed information such as "20% discount coupon usable only on the first purchase for new member registrants," "Period: March 1, 2023 - April 30, 2023," "Target: New member registrants," "Usage conditions: Only for the first purchase," "Method of issuance: Coupon code sent via email upon completion of member registration," "Implementation results: 5,000 registrants, coupon usage rate 30%," "KPI: Number of new member registrations, first purchase rate," and "Effect measurement: Number of new member registrations increased by 150% compared to the same period last year." In addition, it manages past measures that have been discontinued or temporarily suspended, and allows reference to past failures, successes, and improvement histories, such as "2022 Summer Outlet Limited Point Redemption Campaign (Reason for discontinuation: Low usage rate)" and "2021 Winter App Limited Coupon Distribution (Improvement history: Usage rate improved due to increased coupon amount)," thereby providing multifaceted support for decision-making by those in charge.
[0085] The Competitor Strategy Information Storage Unit stores information on competitors' promotional strategies collected using industry research departments, external research companies, web crawlers, scraping tools, etc., as a database on the server. For example, it stores strategy specifications with similar attribute information, such as "Competitor A: 500 points awarded for new member registration," "Competitor B: Weekend-only time sale," "Competitor C: Friend referral campaign," "Competitor D: Coupon awarded for posting reviews," "Competitor E: Free shipping during your birthday month," "Competitor F: Point rewards based on member rank," "Competitor G: Double points for app users only," "Competitor H: Discounts on outlet products only," "Competitor I: Free shipping for new members only," "Competitor J: Special benefits via birthday email," and "Competitor K: Special discounts for repeat customers." This information is collected automatically by web crawlers, industry news, etc. By utilizing diverse information sources such as manual registration of campaigns and advertising materials, extraction of information from review sites and social media, regular monitoring of competitors' campaign pages, and information gathering at industry events and exhibitions, the system comprehensively manages campaign details, target audience, terms of provision, method of distribution, usage conditions, timing, exclusion conditions, implementation results, start and end dates, objectives, KPIs, effectiveness measurement results, improvement history, and reasons for discontinuation. It also manages the change history of competitors' campaign details and completed campaigns, providing basic data for understanding industry-wide promotion trends and the competitive environment, such as "Competitor B's weekend-only time sale (April 2022 - March 2023, reason for discontinuation: increased costs)" and "Competitor C's friend referral campaign (increased referrer incentives in summer 2021)."
[0086] The input reception unit has the function of inputting new promotional measures through user interfaces such as personal computers, tablet terminals, and smartphones used by service design personnel. It is equipped with a variety of input methods, including text input fields, voice input, template selection, conditional selection forms, a function to reuse past measures, a drag-and-drop function to combine measures elements, AI-powered auto-completion, and a measure simulation function. Measures such as "Give new members a 20% discount coupon usable only on their first purchase," "Offer free shipping during the birth month exclusively for gold members," "Give points usable on the next purchase to all reviewers," "Give coupons to both parties for friend referrals," "5x points on all products for a limited time," "Free shipping on first purchases exclusively through the app," "Additional discounts exclusively for outlet products," "Birthday email benefits," and "Special discounts for repeat customers" can be entered through free-form text or semi-automatic generation. Input content is securely transmitted from the terminal to the server, and the unit also includes functions such as automatic saving of input content, draft saving, automatic completion of input content, automatic detection of input errors, real-time preview of input content, and prediction of measure effectiveness based on input content.
[0087] The evaluation and proposal department is implemented on the server side, with a generative AI at its core. It semantically matches and analyzes all data from the input new policy proposals, existing company policy information storage, and competitor policy information storage using natural language processing technology. It extracts elements such as policy content, target audience, provision conditions, granting method, usage conditions, timing, exclusion conditions, implementation results, policy objectives, policy KPIs, policy effectiveness measurement results, policy improvement history, and reasons for policy discontinuation. It does not simply match keywords, but also considers semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching provision timing, essential overlap of benefit content, and more. We compare Sentiment from multiple perspectives, including its economic value, impact on user experience, cost structure, risk factors, and legal / regulatory compliance. We accurately determine the overlap and similarity of essential incentives and user benefits even with different expressions and conditions, such as "first-time purchase discount" and "coupon for new member registration," or "birthday month points gift" and "free shipping during your birthday month." We also analyze the differences in granting methods and incentives from multiple perspectives, such as "granting points usable on your next purchase to all reviewers" and "receiving a coupon for submitting a review," to identify differentiating points. When a new policy proposal such as "Offer free shipping during the birth month exclusively for gold members" is entered, the system extracts overlapping risks and compares it with existing company policy information memory such as "Special sale exclusively for gold members," "Birth month point gift," and "Birth month free shipping (past implementation)" to comprehensively evaluate overlapping target audiences, differences in benefit content, differences in provision conditions, past implementation results and effectiveness measurement results. It also compares it with competitor policy information memory such as "Competitor E's birthday month free shipping" and "Competitor J's benefit provision via birthday email" to evaluate points of differentiation from competitors, novelty, and competitiveness. The system comprehensively analyzes consistency scores with existing policies, similarity scores with competitor policies, novelty and competitiveness scores, the economic impact of the policy, impact on user experience, risk assessment of the policy, and legal and regulatory compliance assessment. Finally, the system automatically generates evaluation comments and proposals for the new policy proposal using natural language generation, such as "There is some overlap with the existing policy '10% discount coupon for first-time buyers,' but it can be differentiated because the discount rate and usage conditions are different," "The incentive effect is similar to competitor A's point provision policy, but it has uniqueness in the form of a discount coupon," and "There are no similar policies among existing or competitors,The system outputs specific and easy-to-understand evaluation comments such as, "High novelty and competitiveness," "While past 'free shipping during birthday month' campaigns saw low utilization rates, limiting it to gold members is expected to reduce costs while improving the satisfaction of loyal customers," and "Awarding points to all reviewers is more cost-effective than competitor D's coupon giveaway campaign and is expected to increase repeat purchase rates." It also presents the reasons for the evaluation, excerpts of existing and competing campaigns that support the evaluation, similarity scores, detailed improvement proposals, projected economic impact of the campaign, risk assessment, and results of legal and regulatory compliance checks.
[0088] The user interface unit operates on the terminal and provides screens for designers to input new policy proposals, view evaluation comments and proposals from the evaluation and proposal unit, and refer to information from the history management unit. It has a variety of screen configurations, including input forms, evaluation result display screens, past history reference screens, feedback input screens, improvement proposal display screens, policy effect simulation screens, policy cost analysis screens, policy risk evaluation screens, and legal and regulatory compliance check screens, thereby increasing the work efficiency of the person in charge. It allows them to modify the content of policy proposals based on evaluation comments, emphasize differentiating points, input feedback on AI evaluation results, run policy effect simulations, and analyze policy costs and risks.
[0089] The communication control unit controls the transmission and reception of data between terminals and servers, securely exchanging new policy proposal information from the input reception unit and evaluation results and proposal content from the evaluation and proposal unit via a secure communication channel. It prevents information leakage using encryption technologies such as SSL and TLS, and supports a variety of communication scenarios, including real-time transmission of input content, immediate return of evaluation results, automatic synchronization when the database is updated, simultaneous access control from multiple terminals, automatic retransmission in case of communication failure, and recording of communication logs.
[0090] The History Management Department centrally manages on a server the input history of past policy proposals, evaluation and proposal history, adoption and rejection history, operation history for each person in charge, policy improvement history, policy effectiveness measurement results, reasons for policy discontinuation, and feedback history. This allows people in charge to design new policy proposals while referring to past examples. Specific examples are managed chronologically, such as "The new member campaign in summer 2022 was rejected because it overlapped with existing policies," "The gold member benefits in spring 2023 were adopted because they were differentiated from competitors," "The app-exclusive coupon distribution in winter 2021 saw increased usage due to increased coupon amounts," and "The outlet-exclusive point reward campaign in autumn 2020 was discontinued due to low usage," enabling quality control and knowledge sharing throughout the entire design process.
[0091] The Security Management Department takes measures to protect confidential information such as policy specifications, evaluation results, and input from personnel, including encrypting communications between servers and terminals, managing access rights to databases, recording audit logs, preventing the misuse of AI models, implementing firewalls and intrusion detection systems to prevent unauthorized access and information leaks from external sources, using digital signatures and hash value management to prevent tampering with AI evaluation results, and restricting the policy information that each person can access.
[0092] Specific examples of prompt statements to be input to the generating AI when implementing the present invention include: "For the following proposed new promotional measures, semantically compare them with all the measure specification information stored in the existing company's measure information storage unit and the competitor's measure information storage unit, evaluate them from the perspectives of duplication, contradictions, similarities, points of differentiation, novelty, competitiveness, economic impact, risks, and legal / regulatory compliance, and generate easy-to-understand comments along with the reasons for the evaluation. Proposed new measures: 'Provide new members with a 20% discount coupon that can be used only on their first purchase.'" In this way, the system according to this embodiment has each part working closely together, with the generating AI at its core, to achieve efficiency, sophistication, quality improvement, and enhanced competitiveness in measure design work.
[0093] (Implementation Steps)
[0094] Step 1: Collect and register information on existing company strategies and competitor strategies. First, the e-commerce site operations department and industry research department register the specifications of all past and present in-house promotional measures into the existing in-house measures information storage unit, and simultaneously register information on competitors' promotional measures into the competitor measures information storage unit. For example, in-house measures such as "10% discount coupon for first-time buyers," "double points when member rank is upgraded," "limited-time free shipping campaign," and "birthday month point gift," as well as competitor measures such as "500 points awarded for new member registration at competitor A," "weekend-only time sale at competitor B," and "friend referral campaign at competitor C," are stored in the database along with various attribute information such as measure name, target audience, terms of service, benefit details, method of awarding, usage conditions, timing, exclusion conditions, implementation results, KPIs, effectiveness measurement results, improvement history, and reasons for discontinuation. Information is registered using various methods such as manual input, file import, automatic collection by web crawlers, and extraction from social media and review sites.
[0095] Step 2: Enter your new promotional strategy proposal Service designers use the terminal's user interface to input new promotional strategy proposals. Input methods are diverse, including text input, voice input, template selection, conditional forms, reuse of past proposals, AI-powered auto-completion, and strategy simulation functions. Examples include "Give new members a 20% discount coupon usable only on their first purchase," "Offer free shipping during the birth month for gold members," "Give points to all reviewers for their next purchase," "Give coupons to both parties for friend referrals," and "Offer 5x points on all products for a limited time." Proposals can be entered through free-form text or semi-automatic generation. The input content is transmitted securely from the terminal to the server.
[0096] Step 3: Semantic matching and analysis of policy proposals using generative AI The Evaluation and Proposal Department uses a generating AI to semantically match and analyze all policy specification information stored in the input new policy proposals, existing company policy information storage units, and competitor policy information storage units using natural language processing technology. The generating AI extracts elements such as policy content, target audience, provision conditions, granting methods, usage conditions, timing, exclusion conditions, implementation results, KPIs, effectiveness measurement results, improvement history, and reasons for discontinuation. It performs comparisons from multiple perspectives, including not just keyword matching, but also semantic vectors, contextual understanding, similarity of conditions, overlapping target audiences, matching provision timing, essential overlap of benefit content, economic value of incentives, impact on user experience, cost structure of policies, risk factors, and legal and regulatory compliance. Specific examples of prompt text to be fed into the generating AI include: "For the following proposed new promotional measure, semantically compare it with all measure specification information stored in the existing company measure information storage unit and competitor measure information storage unit, evaluate it from the perspectives of duplication, contradictions, similarities, points of differentiation, novelty, competitiveness, economic impact, risk, and legal / regulatory compliance, and generate an easy-to-understand comment along with the reasons for the evaluation. Proposed new measure: 'Provide new members with a 20% discount coupon that can be used only on their first purchase.'"
[0097] Step 4: Automatic generation of evaluation comments and proposals by AI. The generating AI comprehensively analyzes the consistency score with existing measures, the similarity score with competing measures, the novelty and competitiveness score, the economic impact of the measure, the impact on user experience, the risk assessment, and the legal and regulatory compliance assessment based on the matching and analysis results, and automatically generates evaluation comments and proposals for the final new measure using natural language generation. For example, it outputs specific and easy-to-understand evaluation comments such as, "There is some overlap with the existing measure, the '10% discount coupon for first-time buyers,' but it can be differentiated because the discount rate and usage conditions are different," "The incentive effect is similar to competitor A's point reward measure, but it has uniqueness in the form of a discount coupon," "There are no similar measures in place with existing or competing companies, so it has high novelty and competitiveness," and "The past 'free shipping during birthday month' measure had a low utilization rate, but by limiting it to gold members, we can expect to improve the satisfaction of loyal customers while keeping costs down." Examples of prompts to be fed into the generating AI include: "Compare the following proposed policy with existing policies and competing policies, analyze in detail the overlap and similarity of content, originality, proposed improvements, economic impact, risks, and legal compliance, and output evaluation comments. Proposed policy: 'Award points that can be used next time to all reviewers.'"
[0098] Step 5: Output of evaluation results and proposed content, and confirmation and revision by the person in charge. The evaluation and proposal department outputs evaluation comments and proposal content, automatically generated by AI, to the terminal via the user interface department, allowing the person in charge to review them immediately. The person in charge can then use the evaluation comments to revise the content of the policy proposal, emphasize differentiating points, input feedback on the AI's evaluation results, and perform additional tasks such as policy effect simulations, cost analysis, risk assessment, and legal compliance checks.
[0099] Step 6: History management and knowledge sharing The History Management Department centrally manages past policy proposal input history, evaluation / proposal history, adoption / rejection history, operation history for each person in charge, policy improvement history, policy effectiveness measurement results, reasons for policy discontinuation, and feedback history on a server, enabling people in charge to design new policy proposals while referring to past examples. For example, "The campaign for new members in the summer of 2022 was rejected because it overlapped with an existing policy," "The benefits for gold members in the spring of 2023 were adopted because they were evaluated as differentiating themselves from competitors," "The app-exclusive coupon distribution in the winter of 2021 saw increased usage due to an increase in coupon value," and "The outlet-exclusive point reward campaign in the fall of 2020 was discontinued due to low usage." By managing specific examples in chronological order, quality control of the entire design process and knowledge sharing are achieved.
[0100] (Specific use cases)
[0101] For example, if the e-commerce site operations department is considering a new promotional measure such as "giving new members a 20% discount coupon usable only on their first purchase," the person in charge would enter the proposed measure into the text input field on the user interface of the input reception unit. The input content is sent to the server via the communication control unit, and the evaluation and proposal unit starts the analysis process using the generation AI. The existing in-house measure information storage unit stores various in-house measure information such as "10% discount coupon for first-time buyers," "free shipping on first purchase for new member registrations," "double points when member rank is upgraded," "limited-time free shipping campaign," and "birthday month point gift." The competitor measure information storage unit stores competitor measure information such as "500 points awarded for new member registrations at competitor A," "weekend-only time sale at competitor B," "friend referral campaign at competitor C," "coupon awarded for posting reviews at competitor D," and "free shipping during the birthday month at competitor E." The generating AI semantically compares and analyzes all specification information of the input policy proposals, existing company policy information, and competitor policy information from multiple perspectives, including policy content, target audience, provision conditions, granting method, usage conditions, timing, exclusion conditions, implementation results, KPIs, effectiveness measurement results, improvement history, and reasons for discontinuation. For example, it compares in detail differences such as discount rate and usage conditions with "10% discount coupon for first-time buyers," differences in incentive type and user experience with "free shipping on first purchase for new member registration," differences in economic value and timing of use with "competitor A's 500 points awarded for new member registration," differences in the appeal and duration of the policy with "competitor B's weekend-only time sale," and differences in the scope of the target audience and the method of distributing incentives with "competitor C's friend referral campaign," and comprehensively evaluates overlaps, similarities, differentiation points, novelty, competitiveness, economic impact, risk, and legal and regulatory compliance.The generating AI automatically creates evaluation comments that include reasons for evaluation, evidence, improvement suggestions, risks, and legal compliance check results, such as: "There is some overlap with the existing '10% discount coupon for first-time buyers' campaign, but it can be differentiated by the different discount rate and usage conditions," "The incentive effect is similar to competitor A's point reward campaign, but it has uniqueness in the form of a discount coupon," "There are no similar campaigns among existing companies or competitors, so it is highly novel and competitive," and "The previous 'free shipping during your birthday month' campaign had a low utilization rate, but this campaign, by limiting it to new member registrants, can be expected to acquire new members while keeping costs down." These comments are output to the person in charge via the user interface. The person in charge can use these evaluation comments as a reference to revise the content of the campaign proposal, emphasize the points of differentiation, input feedback on the AI's evaluation results, and perform additional campaign effect simulations, cost analyses, risk assessments, and legal compliance checks.
[0102] A concrete example of a prompt message to be fed into the generating AI is: "For the following proposed new promotional measure, semantically compare it with all the measure specification information stored in the existing company measure information storage unit and the competitor measure information storage unit, evaluate it from the perspectives of duplication, contradictions, similarities, points of differentiation, novelty, competitiveness, economic impact, risks, and legal / regulatory compliance, and generate an easy-to-understand comment along with the reasons for the evaluation. Proposed new measure: 'Provide new members with a 20% discount coupon that can be used only on their first purchase.'"
[0103] Furthermore, if, for example, a proposed policy such as "award points to all reviewers for use on their next purchase" is entered, the generating AI will compare the differences in incentive types, awarding methods, and usage conditions with existing company policy information storage items such as "exclusive coupons for reviewers," "special discounts for repeat purchasers," and "double points upon member rank upgrades." It will also analyze the differences in economic value and user experience with competitor policy information storage items such as "competitor D offers coupons for review submissions" and "competitor F offers point rewards based on member rank," and automatically generate evaluation comments such as "awarding points to all reviewers is more cost-effective than competitor D's coupon awarding policy and is expected to improve the repurchase rate" and "there is some overlap with the existing policy 'exclusive coupons for reviewers,' but the point awarding format is expected to have a promotional effect." In this way, the system of the present invention can comprehensively compare and evaluate the input new policy proposal with existing company policy information and competitor policy information using the generating AI, and can determine overlaps, contradictions, similarities, differentiating points, novelty, competitiveness, economic impact, risks, and legal and regulatory compliance, and automatically generate easy-to-understand comments and improvement suggestions along with the reasons for the evaluation.
[0104] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.
[0105] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.
[0106] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.
[0107] For example, the data collection unit is implemented by the control unit 46A of the smart device 14 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart device 14 and processes it with the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the output device 40 of the smart device 14 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the devices and control units is not limited to the examples described above and can be modified in various ways.
[0108] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.
[0109] [Second Embodiment]
[0110] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.
[0111] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.
[0112] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).
[0113] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and communication interface 44 are also connected to the bus 52.
[0114] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.
[0115] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).
[0116] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.
[0117] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.
[0118] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.
[0119] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.
[0120] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48. The smart glasses 214 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the identification processing unit 290 using these models.
[0121] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is realized by the data processing device 12 and the smart glasses 214. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".
[0122] (Example 1)
[0123] Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.
[0124] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.
[0125] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.
[0126] Furthermore, the processing by the data processing system 10 described above is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.
[0127] For example, the data collection unit is implemented by the control unit 46A of the smart glasses 214 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the smart glasses 214 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 of the smart glasses 214 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.
[0128] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.
[0129] [Third Embodiment]
[0130] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.
[0131] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.
[0132] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).
[0133] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, display 343, and communication interface 44 are also connected to the bus 52.
[0134] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.
[0135] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).
[0136] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.
[0137] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.
[0138] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.
[0139] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.
[0140] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.
[0141] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the headset terminal 314. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".
[0142] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.
[0143] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.
[0144] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.
[0145] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.
[0146] For example, the data collection unit is implemented by the control unit 46A of the headset terminal 314 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the headset terminal 314 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using generation AI. For example, the serving unit is implemented by the speaker 240 and display 343 of the headset terminal 314 or the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.
[0147] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.
[0148] [Fourth Embodiment]
[0149] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.
[0150] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.
[0151] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).
[0152] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, controlled object 443, and communication interface 44 are also connected to the bus 52.
[0153] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.
[0154] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).
[0155] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.
[0156] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.
[0157] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.
[0158] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.
[0159] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.
[0160] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.
[0161] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. Each part of the system described below is implemented by the data processing device 12 and the robot 414. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".
[0162] (Example 1) Since the process flow is the same as in Example 1 described in the First Embodiment above, the explanation will be omitted.
[0163] (Application Example 1) Since the process flow is the same as in Application Example 1 described in the first embodiment above, the explanation will be omitted.
[0164] The specific processing unit 290 transmits the result of the specific processing obtained in the embodiment 1 or application example 1 described above to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.
[0165] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or external devices, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or external devices.
[0166] For example, the data collection unit is implemented by the control unit 46A of the robot 414 or the specific processing unit 290 of the data processing device 12. For example, the acquisition unit acquires step count data using the camera 42 and communication I / F 44 of the robot 414 and processes it using the specific processing unit 290 of the data processing device 12. For example, the analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the data from the data collection unit and the acquisition unit. For example, the generation unit is implemented by the specific processing unit 290 of the data processing device 12 and generates a cooking menu using a generation AI. For example, the serving unit is implemented by the speaker 240 and controlled object 443 of the robot 414 and the specific processing unit 290 of the data processing device 12 and provides the generated cooking menu to the user. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.
[0167] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.
[0168] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.
[0169] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. In the upper and lower directions of the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. Also, the upper side of the concentric circles is where "pleasant" emotions are located, and the lower side is where "unpleasant" emotions are located. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.
[0170] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.
[0171] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.
[0172] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.
[0173] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."
[0174] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.
[0175] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.
[0176] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.
[0177] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.
[0178] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.
[0179] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.
[0180] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.
[0181] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.
[0182] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.
[0183] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.
[0184] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.
[0185] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.
[0186] Furthermore, the following additional information is disclosed regarding the above explanation.
[0187] (Claim 1)
[0188] This is a promotion strategy design support system for e-commerce sites, comprising an existing company strategy information storage unit, a competitor strategy information storage unit, an input reception unit, and an evaluation / proposal unit. The existing company strategy information storage unit stores specification information of past and currently implemented company promotion strategies as a database, along with various attribute information such as strategy name, target audience, terms of service, benefits, granting method, usage conditions, timing, exclusion conditions, and implementation results. The competitor strategy information storage unit stores information on competitors' promotion strategies collected through industry research and automated collection tools as a database, along with similar attribute information. The input reception unit receives information from the person in charge regarding new promotion strategies they wish to consider. The system provides a user interface that allows users to input policy proposals in various ways, such as text input, template selection, and conditional selection forms. The evaluation and proposal unit semantically compares the new policy proposals entered via the input reception unit with all policy specification information stored in the existing company policy information storage unit and the competitor policy information storage unit using generative AI and natural language processing technology. It extracts and compares elements such as policy content, target audience, provision conditions, granting methods, usage conditions, timing, and exclusion conditions. The system automatically evaluates consistency and overlap with existing policies, similarity and differentiation points with competitor policies, novelty and competitiveness, and automatically generates and outputs new policy proposals including evaluation reasons and improvement suggestions. (Claim 2)
[0189] The system according to claim 1, wherein the evaluation and proposal department, using natural language processing by generative AI, compares new policy proposals with existing company policy information and competitor policy information, not merely by keyword matching, but by comprehensively comparing them including detailed elements such as the contextual meaning of the policy content, differences in expression, provision conditions and scope, method of assignment, usage conditions, timing, exclusion conditions, and past implementation results, thereby determining with high accuracy whether the substantive content of the policies overlaps or is similar to those of the policies. (Claim 3)
[0190] The system according to claim 1, characterized in that the evaluation and proposal department, using a natural language generation function by a generative AI, automatically generates evaluation results for new policy proposals, including consistency with existing policies, similarity with competing policies, comprehensive evaluation comments regarding novelty and competitiveness, and specific proposals such as recommendations for adoption of the new policy proposal or consideration of differentiation, and outputs them to the person in charge in an easily understandable manner. [Explanation of Symbols]
[0191] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots
Claims
1. The system comprises: an existing benefit information storage unit that stores specification information of existing plans and campaign benefits; a third-party benefit information storage unit that stores specification information of third-party benefits; an input receiving unit that receives information on new benefit proposals entered by the user; and an evaluation / proposal unit that uses a generation AI to compare the existing benefit specification information stored in the existing benefit information storage unit and the third-party benefit specification information stored in the third-party benefit information storage unit with the new benefit proposal information entered via the input receiving unit, automatically performs consistency checks with existing benefits and similarity evaluations with third-party benefits, generates a new benefit proposal evaluated from the perspectives of whether there are any overlaps or contradictions with existing benefits, the degree of differentiation from third-party benefits, novelty and competitiveness based on the content of the new benefit proposal, and outputs the new benefit proposal and evaluation results. The evaluation / proposal unit uses a generation AI to calculate the semantic similarity of the benefit specification information by natural language processing, extracts existing benefits or third-party benefits that have a similarity of a predetermined threshold or higher, and determines the consistency and competitiveness of the new benefit proposal based on the extraction results.
2. The system according to claim 1, characterized in that when the evaluation and proposal unit compares the proposed new benefits information with the existing benefits specification information and the other company's benefits specification information, it uses natural language processing by a generative AI to make a multifaceted comparison including the contextual meaning of the benefits, differences in expression, and detailed elements such as the conditions for providing the benefits and the scope of eligibility, and determines with high accuracy whether the substantive content of the benefits is similar or not, rather than simply matching strings.
3. The system according to claim 1, characterized in that the evaluation and proposal unit automatically generates, using a natural language generation function of a generative AI, the evaluation results of the proposed new benefits, including the results of consistency with existing benefits, the results of similarity with other companies' benefits, a comprehensive evaluation comment regarding novelty and competitiveness, and specific proposals such as recommendations for adoption of the proposed new benefits or consideration of differentiation, and outputs them to the user in an easy-to-understand manner.
Citation Information
Patent Citations
Persona chatbot control method and system
JP2022180282A
