Product design iteration method and device, electronic equipment and storage medium

Abstract

The invention provides a product design iteration method and device, electronic equipment and a storage medium, and relates to the technical field of artificial intelligence. The product design iteration method is applied to a product design iteration system, the product design iteration system comprises a data collection agent, an industry expert agent and a user simulation agent, and the product design iteration method comprises the steps that a competing product analysis report is generated through the data collection agent; optimizing the competitive product analysis report through the industry expert agent to generate a product essence report; and optimizing the product essence report through the user simulation agent to generate personalized optimization suggestions. Task transmission and feedback sharing between agents are achieved through a unified coordination protocol, high automation and intelligence of the product design process are achieved, the limitation of poor real-time performance, information isolated island and manual analysis lag in a traditional method is effectively broken through, and the overall product research and development efficiency and response capacity are improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to a product design iteration method, apparatus, electronic device, and storage medium. Background Technology

[0002] Against the backdrop of the rapid development of artificial intelligence and natural language processing technologies, large language models (LLMs) have achieved remarkable results in text generation, information extraction and semantic understanding, and are widely used in content creation, intelligent customer service, code generation, product analysis and other scenarios.

[0003] However, with the increasing diversification and complexity of product design and user needs, the traditional approach of relying on a single LLM or human team for market research and competitor analysis has gradually revealed the following shortcomings: (1) knowledge updates are lagging behind and lack real-time updates; (2) tasks are highly complex and lack collaboration mechanisms; and (3) personalized experiences are difficult to achieve on a large scale. Summary of the Invention

[0004] The problem addressed by this invention is how to achieve intelligent and efficient product design iteration.

[0005] To address the above problems, the present invention provides a product design iteration method, apparatus, electronic device, and storage medium.

[0006] In a first aspect, the present invention provides a product design iteration method applied to a product design iteration system, the product design iteration system including a data collection agent, an industry expert agent, and a user simulation agent, and the product design iteration method comprising: The data collection agent generates a competitor analysis report; The competitor analysis report is optimized by the industry expert agent to generate a product summary report; The product summary report is optimized using the user simulation agent to generate personalized optimization suggestions.

[0007] Optionally, generating the competitor analysis report through the data collection agent includes: Determine survey needs based on the survey instructions input by the user; Based on the aforementioned research requirements, external tools were invoked through a multi-tool invocation protocol to obtain multi-source data. The competitive analysis report is generated by fusing the multi-source data using a weighted scoring algorithm.

[0008] Optionally, the step of invoking external tools through a multi-tool invocation protocol to obtain multi-source data includes: Calling web search engine interfaces and / or web crawler tools to obtain publicly available industry information; The retrieval enhancement generation module is invoked to retrieve and supplement missing information from the preset knowledge base.

[0009] Optionally, the fusion of the multi-source data based on the weighted scoring algorithm includes: Data from different data sources are scored separately, and the overall score of the multi-source data is determined based on the scores and weights of each data source. The data sources include the web search engine interface, the crawler tool, and the search enhancement generation module.

[0010] Optionally, optimizing the competitor analysis report through the industry expert agent includes: The competitor analysis report is parsed based on a pre-defined large language model; Based on the analysis results, assess the characteristics of competitors and propose suggestions for iteration directions, generating a product summary report that includes a summary of competitor characteristics and suggestions for iteration optimization.

[0011] Optionally, optimizing the product summary report through the user simulation agent includes: Multiple user roles are generated, and these user roles are constructed based on user profiles. For each user role, a simulated usage scenario and task instructions are constructed. The user simulation agent executes the interaction between the user role and the product and generates simulated feedback. Based on the simulated feedback, user needs are determined to generate the personalized optimization suggestions.

[0012] Optionally, determining user needs based on the simulated feedback includes: Sentiment analysis is performed on the simulated feedback to identify the emotional bias of user evaluations; Keyword extraction is performed based on simulated feedback text to obtain the core product-related concerns in the simulated feedback. Using the emotional tendencies and core concerns as input, cluster analysis is performed to identify common pain points reported by multiple user roles, and product improvement suggestions are generated based on these common pain points.

[0013] Secondly, the present invention provides a product design iteration device applied to a product design iteration system, the product design iteration system including a data collection agent, an industry expert agent, and a user simulation agent, the product design iteration device comprising: The first module is used to generate a competitor analysis report through the data collection agent; The second module is used to optimize the competitor analysis report through the industry expert agent to generate a product summary report; The third module is used to optimize the product essence report through the user simulation agent to generate personalized optimization suggestions.

[0014] Thirdly, the present invention provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the product design iteration method as described in the first aspect when executing the computer program.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the product design iteration method as described in the first aspect.

[0016] The beneficial effects of the product design iteration method of the present invention are as follows: By introducing a multi-LLM-Agent collaborative system consisting of data collection agents, industry expert agents, and user simulation agents, a closed-loop product design iteration process is established that runs through "data collection - professional analysis - user simulation - optimization suggestions". Task transfer and feedback sharing between agents can be realized through a unified coordination protocol, achieving a high degree of automation and intelligence in the product design process. This effectively breaks through the limitations of poor real-time performance, information silos, and lag in manual analysis in traditional methods, and improves the overall product development efficiency and responsiveness. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the product design iteration method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the product design iteration system according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the process for generating a competitor analysis report according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the process for acquiring multi-source data according to an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the optimized competitive analysis report process according to an embodiment of the present invention. Figure 6 A schematic diagram illustrating the optimized process of the product summary report in this embodiment of the invention; Figure 7 This is a schematic diagram of the process for determining user needs according to an embodiment of the present invention; Figure 8 This is a system architecture diagram of the product design iteration device according to an embodiment of the present invention; Figure 9 This is a system architecture diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0019] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0020] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0021] It should be noted that the terms "one" and "more" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0022] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0023] like Figure 1 As shown in the illustration, an embodiment of the present invention provides a product design iteration method applied to a product design iteration system. The product design iteration system includes a data collection agent, an industry expert agent, and a user simulation agent. The product design iteration method includes: S100: Generate a competitor analysis report through the data collection agent.

[0024] Specifically, the data collection agent (Agent A) is responsible for the real-time collection and preliminary analysis of market data. It can integrate web search tools, data crawling tools, and search enhancement generation (RAG) tools through the Multi-Tool Call Protocol (MCP). It can collect industry dynamic information and competitor-related data from the Internet and specified databases in real time, and generate a structured preliminary competitor analysis report based on the collected data. The data collection agent (Agent A) combines the MCP protocol with the ability to call various external tools to ensure the dynamic acquisition of competitor information and market trends, effectively solving the problems of lagging knowledge updates and insufficient real-time performance in existing technologies.

[0025] Among them, combined Figure 2 As shown, the product design iteration system includes data collection agents, industry expert agents, and user simulation agents. These agents can interact and collaborate in real time through the A2A protocol (inter-agent collaboration protocol), avoiding information silos, improving task decomposition and execution efficiency, and ensuring the high efficiency of task decomposition and information sharing.

[0026] S200: The competitor analysis report is optimized by the industry expert agent to generate a product summary report.

[0027] Specifically, industry expert agents (Agent B) are used to conduct professional evaluations of competitive analysis reports and provide optimization suggestions. Through a pre-set professional prompt-driven LLM model, they can conduct in-depth evaluations of competitive analysis reports, identify the strengths and weaknesses of competitors, and generate a feature iteration report that includes the extraction of core product features and suggestions for iteration directions, i.e., a product essence report.

[0028] S300: Optimize the product summary report using the user simulation agent to generate personalized optimization suggestions.

[0029] Specifically, the user simulation agent (Agent C) is used to simulate different user roles, experience the product, and generate personalized feedback. By simulating different user roles and usage scenarios, it can evaluate the effectiveness of optimized product design and generate personalized optimization suggestions for different user needs. By building diverse user profiles and performing large-scale simulated interactions, combined with natural language processing algorithms for feedback analysis and personalized suggestion generation, the user simulation agent (Agent C) realizes automated modeling and expansion of personalized experiences, breaking through the technical bottlenecks of high cost and low efficiency in personalized testing and adaptation of traditional methods.

[0030] In this embodiment, a multi-LLM-Agent collaborative system consisting of data collection agents, industry expert agents, and user simulation agents is introduced to establish a closed-loop product design iteration process that runs through "data collection - professional analysis - user simulation - optimization suggestions". Through a unified coordination protocol, task transfer and feedback sharing between agents are realized, achieving a high degree of automation and intelligence in the product design process. This effectively breaks through the limitations of poor real-time performance, information silos, and lag in manual analysis in traditional methods, and improves the overall product development efficiency and responsiveness.

[0031] Optionally, generating the competitor analysis report through the data collection agent includes: S110: Determine the survey requirements based on the survey instructions entered by the user.

[0032] Specifically, in combination Figure 3 As shown, the data collection agent (Agent A) receives the survey instructions input by the user, performs semantic parsing of the instructions through natural language understanding technology, and clarifies the survey requirements, such as competitor brands, product models, and functions of interest.

[0033] S120: Based on the aforementioned survey requirements, external tools are invoked through a multi-tool invocation protocol to obtain multi-source data.

[0034] Specifically, in combination Figure 3 As shown, the Multi-Tool Call Protocol (MCP) is enabled to call external tools in parallel to obtain multi-source data. By calling external tools, the latest market data and industry trends can be dynamically obtained, ensuring that competitor analysis and product iteration decisions are always based on real-time information, rather than static, outdated time-based data that a single LLM system relies on (a single LLM system is limited by a pre-trained knowledge base and cannot reflect market changes in a timely manner). This not only improves the accuracy of analysis, but also provides enterprises with the ability to respond quickly to market dynamics, making product iteration more timely and competitive.

[0035] Among them, multi-source data can be preprocessed, including: (1) deduplication and cleaning (removing invalid fields and advertising information); (2) extracting key fields, such as functional parameters, user reviews, price ranges, etc.; (3) structuring, including uniformly formatted fields, timestamps, competitor dimensions, etc.

[0036] S130: The multi-source data is fused based on a weighted scoring algorithm to generate the competitor analysis report.

[0037] Specifically, in combination Figure 3As shown, a competitive analysis report is generated by integrating multi-source data based on a weighted scoring algorithm. This results in the generation of a competitive analysis report, which includes the extraction of core features of each competitor (such as unique selling points and technical parameters), a multi-dimensional comparison table (compared with other competitors / previous generation products), a summary of user feedback trends (such as the distribution of subjective experience keywords), and risk warnings or information items to be supplemented.

[0038] In this optional embodiment, by introducing a research objective parsing, multi-tool invocation, and weighted fusion mechanism, the data collection agent can automatically and accurately obtain multi-dimensional competitor data. Compared with the traditional method of relying on manual search or static model invocation, it supports dynamic research and real-time context adjustment, effectively improving the timeliness and coverage of market data, and providing a reliable and high-quality data foundation for subsequent analysis.

[0039] Optionally, the step of invoking external tools through a multi-tool invocation protocol to obtain multi-source data includes: S121: Use web search engine interfaces and / or web crawler tools to obtain publicly available industry information.

[0040] Specifically, in combination Figure 4 As shown, by calling the web search engine interface and using automated query requests, real-time information such as web pages, news, reviews, and forum discussions of target competitors can be obtained to understand industry trends and competitor dynamics and other publicly available industry information. Alternatively, web crawling tools can be used to crawl structured data from competitor websites, e-commerce pages, forums, video platforms, etc., to obtain publicly available data from competitor websites, such as product specification tables, pricing, user reviews, full-function inspection videos, test drive videos, etc.

[0041] S122: Call the retrieval enhancement generation module to retrieve and supplement missing information from the preset knowledge base.

[0042] Specifically, in combination Figure 4 As shown, the retrieval enhancement generation module is invoked to retrieve relevant industry knowledge, product standards, user preferences, and other information from the preset knowledge base, supplementing missing information and ensuring the comprehensiveness and accuracy of the data.

[0043] In this optional embodiment, by integrating a web search engine, crawler tools, and RAG module, the system can flexibly collect structured and unstructured information, and supplement the semantic context with the help of a knowledge base, realizing the complementarity and fusion of multi-source information. Compared with the approach of relying on a single channel or static database, it significantly improves the breadth, depth and semantic integrity of data acquisition, and enhances the sensitivity to industry changes.

[0044] Optionally, the fusion of the multi-source data based on the weighted scoring algorithm includes: Data from different data sources are scored separately, and the overall score of the multi-source data is determined based on the scores and weights of each data source. The data sources include the web search engine interface, the crawler tool, and the search enhancement generation module.

[0045] Specifically, taking data sources including web search engine interfaces, crawler tools, and retrieval enhancement generation modules as an example, the corresponding weights can be: web search engine interface (0.4), crawler tools (0.3), and retrieval enhancement generation modules (0.3). The scoring dimensions include user satisfaction, comprehensiveness of functions, technological advancement, market popularity, etc. The overall score of multi-source data is determined based on the scores and weights of each data source. For example, the data returned by different data sources are scored separately, and the scores are multiplied by the corresponding weights and summed to obtain the overall score of multi-source data. The overall score of multi-source data is the quantitative basis for the competitive analysis report. For example, the scoring results are used to generate core indicators in the competitive analysis report, including quantitative comparison tables, advantages and disadvantages trend analysis, and recommendations, thereby ensuring that the generated analysis report has data-supported objectivity and reliability. This not only improves the credibility and consistency of the report, but also provides hints, references, and structural frameworks for the content generated by the large language model.

[0046] An example overall scoring formula is as follows: ; Where S represents the overall score of the multi-source data, w i This represents the weight of the data source. The weight of different data sources can be determined based on the reliability of the corresponding data source. i The score represents the data source rating (e.g., values ​​from 0 to 1), and n represents the number of data sources (e.g., three).

[0047] In this optional embodiment, a weighted scoring algorithm is introduced to normalize data from different sources and fuse multi-dimensional scores, which helps to improve the objectivity of information processing.

[0048] Optionally, optimizing the competitor analysis report through the industry expert agent includes: S210: Parse the competitor analysis report according to the preset large language model.

[0049] Specifically, in combination Figure 5As shown, the industry expert agent (Agent B) receives a structured competitive analysis report generated by the data collection agent (Agent A). Using natural language understanding technology, it analyzes the core fields in the report (such as product features, user feedback summaries, technical parameters, and market performance ratings). Using a professional prompt template, it guides the LLM (Local Management Analyzer) to perform semantic analysis on each competitive item. This includes comparing the differences between the competitor's features and the product's features, extracting potential strengths, and revealing areas for improvement in user pain points or negative feedback. Through professional prompts and knowledge-driven approaches, the industry expert agent (Agent B) provides in-depth analytical capabilities far exceeding those of a general LLM, accurately analyzing the strengths and weaknesses of competitors and offering scientific and forward-looking suggestions for product optimization. This ensures the quality of competitive analysis and the relevance of product iteration, providing strong support for enterprises to create products with greater technological advantages and user value.

[0050] Among them, professional prompts can be pre-organized and configured by product managers, data analysts, and industry experts based on product design processes, industry terminology, and competitor comparison logic. Alternatively, they can be automatically extracted using existing corpora such as analysis reports, Q&A pairs, and survey texts through text summarization, keyword extraction, and template summarization.

[0051] S220: Based on the analysis results, evaluate the characteristics of competitors and propose suggestions for iteration directions, generating a product summary report that includes a summary of competitor characteristics and suggestions for iteration optimization.

[0052] Specifically, in combination Figure 5 As shown, based on the analysis results, a SWOT analysis model is executed to evaluate the strengths, weaknesses, opportunities, and threats of competitors. This involves identifying competitors' leading characteristics (strengths), their core competencies, advantageous resources, and leading features, such as technological leadership, brand influence, and positive user reputation; marking the weaknesses (weaknesses) of oneself or competitors, including internal weaknesses and disadvantages such as high costs, missing features, poor user experience, and maintenance difficulties; analyzing potential breakthroughs (opportunities) in conjunction with industry trends, identifying potential positive factors in the external environment, such as market trends, new technology applications, increased user demand, and policy support; and identifying factors that may lead to user churn (threats), such as external adverse factors like increased competition, the threat of substitutes, regulatory changes, and shifts in user preferences. Iterative direction suggestions are then proposed, resulting in a product summary report that includes a summary of competitor characteristics (e.g., a summary of competitor strengths and weaknesses) and iterative optimization suggestions.

[0053] In this optional embodiment, competitive semantic analysis is performed using LLM, and the characteristics of competitors are structurally mined by combining SWOT analysis model. This can effectively improve the professionalism and strategic nature of the analysis. Compared with the traditional method that relies on industry experience or static templates, it has stronger knowledge generalization ability and logical reasoning ability, and can output more forward-looking and guiding product optimization suggestions.

[0054] Optionally, optimizing the product summary report through the user simulation agent includes: S310: Generate multiple user roles, which are constructed based on user profiles.

[0055] Specifically, in combination Figure 6 As shown, the user simulation agent (Agent C) receives the product essence report generated by the industry expert agent (Agent B), loads the preset user persona database (User Persona DB) and professional personality database (Professional Persona DB), and models the user persona, including the following attributes: age, gender, occupation, income level, lifestyle habits, product preferences, and technology sensitivity. Through the personality synthesis engine, the user persona and personality are combined to generate prompt seeds and generate multiple user roles, such as "35-year-old IT man who prefers performance and dislikes redundant functions" and "60-year-old retired user who pursues ease of use and reasonable price".

[0056] S320: Construct simulated usage scenarios and task instructions for each user role, execute the interaction between the user role and the product through the user simulation agent and generate simulated feedback, determine user needs based on the simulated feedback, and generate the personalized optimization suggestions.

[0057] Specifically, in combination Figure 6As shown, for each simulated user role, specific usage scenarios and product tasks (interaction goals constructed for the user role) are generated, such as voice navigation, seat heating control, and OTA upgrade experience. The generated prompt seeds are input into the Scene-Task Mapper, which generates complete simulated prompt words, such as "As a 30-year-old female driver, you use the voice-controlled navigation system during rush hour every day. Please simulate the experience and provide feedback." The simulated prompt words are input into the Product API under test. The user simulation agent simulates the user's interaction process (i.e., the process of executing task instructions, where the user role operates the product, experiences functions, and generates feedback in a virtual environment) and feedback text, such as "The navigation voice response is slow and there are repeated announcements" or "The buttons are too small to click while driving." The interaction results are input into the Feedback Parser, and the output of the Feedback Parser enters the MetricEvaluator to calculate metrics such as satisfaction and task success rate. The generated structured feedback is then input into the Clustering & Topic modeling module. The model then uses clustering or topic modeling algorithms to identify frequently occurring feedback topics across multiple user roles. It also uses a common pain point discovery module (Issue Miner) to summarize common product issues or defects. For example, multiple roles expressing dissatisfaction with OTA upgrade speed can be identified as a core improvement direction. Furthermore, it can generate differentiated suggestions based on feedback from different roles. The results from the common pain point discovery module are input into the improvement recommender, which generates simulation reports and iteration suggestions. For example, it might add advanced settings for technical users or simplify the interaction process for elderly users. The improvement recommender updates the product version, and the updated version flows back to the scenario task mapper to enter the next round of simulation, forming an iterative closed loop.

[0058] In this optional embodiment, a virtual user profile generation and usage scenario construction mechanism is introduced, enabling the system to simulate product usage feedback from different user perspectives. This achieves a large-scale, low-cost, and widely covered user testing process. This mechanism effectively replaces the traditional user interview and survey process, significantly reducing costs and improving the feasibility and simulation quality of personalized testing.

[0059] Optionally, determining user needs based on the simulated feedback includes: S321: Perform sentiment analysis on the simulated feedback to identify the emotional tendency of the user's evaluation.

[0060] Specifically, in combination Figure 7 As shown, the emotional tendencies in simulated feedback can be identified and user satisfaction trends can be quantified. For example, natural language sentiment analysis can be performed on each simulated feedback text to classify emotions into positive emotions (such as "very easy to use" and "smooth experience"), neutral emotions (such as "the function basically meets expectations"), and negative emotions (such as "loading is too slow" and "complex operation").

[0061] S322: Extract keywords based on simulated feedback text to obtain the core concerns related to the product in the simulated feedback.

[0062] Specifically, in combination Figure 7 As shown, the simulated feedback text (such as opinions expressed after user experience) is segmented and subjected to dependency parsing to extract the core product features and opinions that users care about. The keyword extraction algorithm is used to extract high-frequency words and important phrases, such as "complex operation" → interaction experience problem; "slow voice recognition" → response speed problem; "beautiful color" → positive feedback on appearance design.

[0063] S323: Using the emotional tendency and the core concern as input, perform cluster analysis to identify common pain points reported by multiple user roles, and generate product improvement suggestions based on the common pain points.

[0064] Specifically, in combination Figure 7 As shown, the extracted keyword set or feedback text is vectorized, and a clustering algorithm is applied to aggregate multiple feedbacks into topic clusters. Each topic cluster is named, such as "voice interaction problem", "OTA update lag", and "complex interface". The coverage and average sentiment value of each topic among all users (based on sentiment analysis) are statistically analyzed to identify consistent problems or trending topics among multiple simulated users, forming a list of product pain points (including weight, severity, and suggested improvement directions) to drive product iteration.

[0065] In this optional embodiment, by performing sentiment analysis, keyword extraction, and topic clustering, automated attribution analysis is performed on massive simulated feedback to identify common pain points and form structured product optimization suggestions. This achieves intelligent parsing and closed-loop processing of feedback data. Compared with the inefficient method of manually sorting feedback, this process greatly improves the product iteration response speed and feedback processing depth.

[0066] like Figure 8 As shown in the figure, an embodiment of the present invention provides a product design iteration device 800, which is applied to a product design iteration system. The product design iteration system includes a data collection agent, an industry expert agent, and a user simulation agent. The product design iteration device 800 includes: The first module 810 is used to generate a competitor analysis report through the data collection agent; The second module 820 is used to optimize the competitor analysis report through the industry expert agent to generate a product summary report; The third module 830 is used to optimize the product essence report through the user simulation agent to generate personalized optimization suggestions.

[0067] like Figure 9 As shown, an electronic device 900 provided in this embodiment of the invention includes a memory 920 and a processor 910; the memory 920 is used to store a computer program; the processor 910 is used to implement the product design iteration method as described above when the computer program is executed.

[0068] Alternatively, an electronic device 900 includes a memory 920 and a processor 910 coupled to the memory 920; the memory 920 is configured to store a computer program; and the processor 910 is configured to perform the following operations when the computer program is executed: The data collection agent generates a competitor analysis report; The competitor analysis report is optimized by the industry expert agent to generate a product summary report; The product summary report is optimized using the user simulation agent to generate personalized optimization suggestions.

[0069] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the product design iteration method described above.

[0070] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: The data collection agent generates a competitor analysis report; The competitor analysis report is optimized by the industry expert agent to generate a product summary report; The product summary report is optimized using the user simulation agent to generate personalized optimization suggestions.

[0071] The present invention will now be described an electronic device 900 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 900 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 900 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0072] Electronic device 900 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0073] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0074] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A product design iteration method, characterized in that, An application is made in a product design iteration system, which includes a data collection agent, an industry expert agent, and a user simulation agent. The product design iteration method includes: The data collection agent generates a competitor analysis report; The competitor analysis report is optimized by the industry expert agent to generate a product summary report; The product summary report is optimized using the user simulation agent to generate personalized optimization suggestions.

2. The product design iteration method according to claim 1, characterized in that, The generation of the competitive analysis report through the data collection agent includes: Determine survey needs based on the survey instructions input by the user; Based on the aforementioned research requirements, external tools were invoked through a multi-tool invocation protocol to obtain multi-source data. The competitive analysis report is generated by fusing the multi-source data using a weighted scoring algorithm.

3. The product design iteration method according to claim 2, characterized in that, The method of invoking external tools through a multi-tool invocation protocol to obtain multi-source data includes: Calling web search engine interfaces and / or web crawler tools to obtain publicly available industry information; The retrieval enhancement generation module is invoked to retrieve and supplement missing information from the preset knowledge base.

4. The product design iteration method according to claim 3, characterized in that, The fusion of the multi-source data based on the weighted scoring algorithm includes: Data from different data sources are scored separately, and the overall score of the multi-source data is determined based on the scores and weights of each data source. The data sources include the web search engine interface, the crawler tool, and the search enhancement generation module.

5. The product design iteration method according to claim 1, characterized in that, The optimization of the competitor analysis report through the industry expert agent includes: The competitor analysis report is parsed based on a pre-defined large language model; Based on the analysis results, assess the characteristics of competitors and propose suggestions for iteration directions, generating a product summary report that includes a summary of competitor characteristics and suggestions for iteration optimization.

6. The product design iteration method according to claim 1, characterized in that, The optimization of the product summary report through the user simulation agent includes: Multiple user roles are generated, and these user roles are constructed based on user profiles. For each user role, a simulated usage scenario and task instructions are constructed. The user simulation agent executes the interaction between the user role and the product and generates simulated feedback. Based on the simulated feedback, user needs are determined to generate the personalized optimization suggestions.

7. The product design iteration method according to claim 6, characterized in that, Determining user needs based on the simulated feedback includes: Sentiment analysis is performed on the simulated feedback to identify the emotional bias of user evaluations; Keyword extraction is performed based on simulated feedback text to obtain the core product-related concerns in the simulated feedback. Using the emotional tendencies and core concerns as input, cluster analysis is performed to identify common pain points reported by multiple user roles, and product improvement suggestions are generated based on these common pain points.

8. A product design iteration device, characterized in that, An application is made in a product design iteration system, which includes a data collection agent, an industry expert agent, and a user simulation agent. The product design iteration device includes: The first module is used to generate a competitor analysis report through the data collection agent; The second module is used to optimize the competitor analysis report through the industry expert agent to generate a product summary report; The third module is used to optimize the product essence report through the user simulation agent to generate personalized optimization suggestions.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the product design iteration method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program that, when executed by a processor, implements the product design iteration method as described in any one of claims 1 to 7.

Images