A method for building an adaptive website architecture based on artificial intelligence

Abstract

The present invention relates to the field of artificial intelligence technology, and in particular, to a method for building an adaptive website architecture based on artificial intelligence driving, which comprises the following steps: clarifying the goals of a website, a target user group and the needs of the target user group, formulating a project plan, and then comprehensively considering multiple technology stacks according to multiple dimensions, and then selecting the optimal technology stack; based on the selected optimal technology stack, defining an adaptability function to measure the degree of matching of existing modules of the website to user goals, defining a performance function to measure the page loading speed, and then quantifying the loading performance of the website, and finally evaluating the website architecture characteristics according to the adaptability and loading performance; the method improves the accuracy, foresight and practicality of the technology stack selection process, and makes the adaptability evaluation more comprehensive and dynamic, which helps to provide a more personalized and instantly responsive service experience, and can also more accurately reflect the actual performance of page loading.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a method for building an adaptive website architecture driven by artificial intelligence. Background Art

[0002] In the context of the current digital transformation, artificial intelligence technology is profoundly changing the design and operation of websites. The construction of adaptive website architecture relies on the advancement of AI technology, especially the application of machine learning, deep learning and other methods. It can automatically adjust content presentation, interface layout and resource allocation according to users' real-time behavior, preferences and access environment, thereby providing a more personalized and efficient user experience. This architecture not only needs to consider the responsive design of the front end to adapt to different devices, but also needs to process and analyze data through back-end intelligent algorithms to ensure that the website can maintain optimal performance in a dynamically changing environment and can continuously optimize its own functions and service quality with user feedback. In addition, the adaptive website architecture also emphasizes the importance of data management strategies. Organizations must modernize their data management processes to better support the needs of AI models for high-speed data analysis and processing. Such architectural design represents the development direction of the new generation of network platforms, aiming to create a more stable, efficient and agile digital service ecosystem. Traditional methods often rely on static evaluation criteria and limited data when selecting a technology stack. Static criteria cannot dynamically adjust weights or consider new factors, resulting in the selection of technology stacks that may not be flexible enough and difficult to cope with the ever-changing technical environment and business needs. In addition, traditional adaptive evaluation methods often only focus on the basic similarity between modules and user needs, ignoring the user's actual behavior data and time factors. Performance evaluation methods may only focus on the total page size and network speed, ignoring the impact of resource optimization and changes in network conditions. Therefore, an adaptive website architecture construction method driven by artificial intelligence is designed. Summary of the invention

[0003] The purpose of the present invention is to provide an adaptive website architecture construction method driven by artificial intelligence to solve the problems raised in the above-mentioned background technology that the selection of technology stack often relies on static evaluation criteria and limited data, and the static criteria cannot dynamically adjust the weights or consider new factors, resulting in the selection of technology stack may not be flexible enough, and the adaptive evaluation method ignores the user's actual behavior data and time factors, and the performance evaluation method ignores the impact of resource optimization and changes in network conditions.

[0004] To achieve the above object, the present invention provides a method for building an adaptive website architecture based on artificial intelligence, comprising the following steps:

[0005] S1. Clarify the website's goals, target user groups, and their needs, develop a project plan, and then comprehensively consider multiple technology stacks based on multiple dimensions to select the optimal technology stack. In the process of comprehensive consideration based on multiple dimensions, introduce time constraints for optimization;

[0006] S2. Based on the selected optimal technology stack, define an adaptability function to measure the matching degree of the existing modules of the website to the user goals, and then quantify the adaptability. At the same time, define a performance function to measure the page loading speed, and then quantify the loading performance of the website. Finally, evaluate the website architecture characteristics based on adaptability and loading performance. In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user goals, introduce the time decay factor and the module participation index for optimization. In the process of defining the performance function to measure the page loading speed, and then quantify the loading performance of the website, introduce resource optimization and network condition changes for optimization;

[0007] S3. Based on the evaluation results of the website architecture characteristics, the optimization objective function that maximizes the user experience is calculated, and then an optimization plan is generated. A new architecture of the website is generated based on the optimization plan and dynamically presented to the user.

[0008] As a further improvement of the present technical solution, in S1, the project plan includes a schedule, a budget and resource allocation.

[0009] As a further improvement of the present technical solution, in S1, the technology stack includes a front-end framework, a back-end framework, a database system and a cloud computing platform.

[0010] As a further improvement of this technical solution, in S1, multiple technology stacks are comprehensively considered according to multiple dimensions to select the optimal technology stack, wherein the multiple dimensions include performance benchmark, community activity, developer familiarity, and scalability, as follows:

[0011] Performance Benchmarks:

[0012] ;

[0013] in, Score performance; is the response time; is the throughput; is the stability coefficient;

[0014] Website activity:

[0015] ;

[0016] in, Score activity; for the number of stars; is the number of branches; The number of issues closed; is the total number of questions;

[0017] Developer familiarity:

[0018] ;

[0019] in, Score points for skill proficiency; is the number of skills; For the skill level; For the The importance of a skill; ;

[0020] Scalability:

[0021] ;

[0022] in, Score scalability; Rate the code quality; Rate the quality of the document; Score new feature support;

[0023] Overall rating:

[0024] ;

[0025] in, For comprehensive rating; weights for scoring performance; is the weight of the activity score; The weight of the skill proficiency score; The weight with which to score scalability.

[0026] As a further improvement of the technical solution, in S1, in the process of comprehensive consideration according to multiple dimensions, a time limit is introduced for optimization, and the optimization is as follows:

[0027] Performance Benchmarks:

[0028] ;

[0029] in, Score the optimized performance; for resource utilization; is the performance factor that decays over time; It is a prediction model trained with historical performance data and used to estimate the performance;

[0030] Website activity:

[0031] ;

[0032] in, is the optimized activity score; for problem solving efficiency; scoring the diversity of contributors; is the activity factor that decays over time; According to the current time A dynamically adjusted weight matrix to balance the importance of each activity indicator;

[0033] Developer familiarity:

[0034] ;

[0035] in, The optimized skill proficiency score; is the teamwork index; Weighted scores for project-related experience; is the skill proficiency factor that grows over time;

[0036] Scalability:

[0037] ;

[0038] in, Score the optimized scalability; is the code reuse rate; Update frequency for the community; is the scalability factor that grows over time; It is a prediction model trained with historical data and used to estimate the future moment Scalability performance;

[0039] Overall rating:

[0040] ;

[0041] in, It is the comprehensive score after optimization.

[0042] As a further improvement of the technical solution, in S2, an adaptability function is defined to measure the matching degree of the existing modules of the website to the user's goal, and then quantify the adaptability, as follows:

[0043] ;

[0044] in, is the fitness function; is the number of modules in the website; For user goals; For the The weight of each module; For modules and user needs similarity; For the website modules;

[0045] In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user's goals, the time decay factor and the module participation index are introduced for optimization. After optimization, the specific results are as follows:

[0046] ;

[0047] in, is the optimized fitness function; is the adjustment coefficient, which is used to control the influence of participation on adaptability score; For users in the module Engagement metrics on It is the highest engagement value among all modules; is the adjustment coefficient, which is used to adjust the impact of the most recently used module on the adaptability score; is a time decay factor.

[0048] As a further improvement of this technical solution, the user goal The determination is as follows:

[0049] Collect behavioral data about users’ interactions with the website through website logs or user behavior monitoring tools , contextual data and historical data , extracting behavioral pattern features related to target prediction from the collected data , contextual features and historical preference characteristics , based on the features , build a classification model to predict the target:

[0050] ;

[0051] in, For a given feature When the target user The probability distribution of is the model weight; is the bias term;

[0052] Select the target with the highest probability as the current prediction:

[0053] ;

[0054] in, To select the target with the highest probability as the current predicted target.

[0055] As a further improvement of the technical solution, in S2, a performance function is defined to measure the page loading speed, thereby quantifying the loading performance of the website, as follows:

[0056] ;

[0057] in, is the performance function; is the current page architecture; The total size of the current page architecture; The network speed of the current user;

[0058] In the process of defining performance functions to measure page loading speed and then quantifying the loading performance of the website, we introduce resource optimization and changes in network conditions for optimization. After optimization, the specific results are:

[0059] ;

[0060] in, is the optimized performance function; The size of the resources that have been cached in the page; is the network acceleration factor; is the network delay time; It is an adjustment item based on resource priority, which is used to consider the impact of resource priority loading on the critical rendering path on the overall perceived performance.

[0061] As a further improvement of the technical solution, in S2, the website architecture characteristics are evaluated according to adaptability and loading performance, and the specific steps are as follows:

[0062] S21. Based on the adaptability function, calculate the adaptability score of each module, and obtain the overall adaptability score of the entire website by taking a weighted average of the adaptability scores of different modules;

[0063] S22. Use the optimized performance function to evaluate the loading performance of each page in detail, identify the key factors affecting the performance, and make improvement suggestions;

[0064] S23. Combine the adaptability score and the performance evaluation results to generate a comprehensive evaluation report, which includes the adaptability score of each module, detailed indicators of page loading performance, and the overall comprehensive score of the website.

[0065] As a further improvement of the technical solution, in S3, the optimization goals of maximizing user experience and minimizing loading time are calculated according to the architecture characteristics, as follows:

[0066] ;

[0067] in, To maximize user experience; The importance of adaptability in the overall goal; To understand the importance of loading performance in the overall goal; The importance weight of external factors in the overall goal; is the impact weight of the device type; is the equipment factor; is the impact weight of network conditions; is the network factor; is the impact weight of the time period; is the time factor.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. In this AI-driven adaptive website architecture construction method, time limits are introduced for optimization in the process of comprehensive consideration based on multiple dimensions. The optimized formula improves the accuracy, foresight and practicality of the technology stack selection process through more detailed consideration factors, dynamically adjusted weights and prediction models based on historical data. It not only helps identify the current optimal technical solution, but also predicts possible problems in the future, providing a solid foundation for continuous optimization and improvement;

[0070] 2. In the method for building an adaptive website architecture driven by artificial intelligence, the time decay factor and the module participation index are introduced for optimization in the process of defining the adaptability function to measure the degree of matching of the existing modules of the website with the user goals. Not only the basic similarity between the module and the user's needs is considered, but also the user's actual behavior data and time factors are combined, so that the adaptability evaluation is more comprehensive and dynamic, which helps to provide a more personalized and responsive service experience. In the process of defining the performance function to measure the page loading speed and then quantifying the loading performance of the website, resource optimization and network condition changes are introduced for optimization, which can not only more accurately reflect the actual performance of page loading, but also better guide developers on how to optimize the page to improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The figure is a flow chart of the overall method of the present invention. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] See also Figure 1 As shown, a method for building an adaptive website architecture based on artificial intelligence driving is provided, comprising the following steps:

[0074] S1. Clarify the website's goals, target user groups, and their needs, develop a project plan, and then comprehensively consider multiple technology stacks based on multiple dimensions to select the optimal technology stack. In the process of comprehensive consideration based on multiple dimensions, introduce time constraints for optimization;

[0075] In S1, the project plan includes schedule, budget, and resource allocation;

[0076] Clear time planning helps ensure that the project is completed on time. By setting milestones and deadlines, team members can better understand the priority of tasks at each stage and the expected completion time, so as to effectively arrange the workflow. This not only improves work efficiency, but also helps management monitor progress and adjust strategies in time to deal with possible delays.

[0077] Budgeting is a key tool for cost control. Properly estimating the funds needed for a project and allocating them to different activities can help avoid overspending. In addition, budget management also includes financial preparation for potential risks, such as unexpected expenses or cost increases caused by market changes, which can make the project run more robustly.

[0078] Reasonable resource allocation ensures the effective use of human resources, material resources, and technical resources. Allocating resources according to project requirements and team capabilities can maximize productivity while ensuring that each participant can play a role in their area of ​​expertise. Effective resource management can also promote team collaboration and improve overall work efficiency.

[0079] In S1, the technology stack includes the front-end framework, back-end framework, database system, and cloud computing platform;

[0080] In S1, multiple technology stacks are comprehensively considered based on multiple dimensions to select the optimal technology stack. The multiple dimensions include performance benchmarks, community activity, developer familiarity, and scalability. Considering performance benchmarks when selecting a technology stack means ensuring that the selected technology can meet the requirements of website operation efficiency, including key indicators such as response time and processing speed. This not only helps to improve user experience, but also ensures that the website can maintain stability and reliability under high traffic conditions. A high-performance technology stack can reduce the consumption of server resources, thereby reducing operating costs;

[0081] A technology stack with a strong and active developer community is essential for long-term maintenance and development. An active community means that the technology is widely used and supported, and it is easier to find solutions when problems arise. At the same time, an active community also promotes the continuous advancement and innovation of the technology itself, bringing continuous value growth to projects using the technology.

[0082] Considering the level of proficiency of team members in a specific technology stack, it is a wise practice to prioritize technologies that they are already proficient in or easy to use during the selection process. Doing so can shorten development cycles, reduce training costs, and improve code quality, because familiar tools can enable developers to work more efficiently and reduce the incidence of errors;

[0083] As the business grows, the website may need to add new features or expand its scale to cope with more traffic. Therefore, it is very important to choose a technology stack with good scalability in the early stage. Good scalability allows the website to easily add new services, integrate external systems, and handle growing amounts of data without large-scale reconstruction, which is extremely beneficial to maintaining the flexibility and competitiveness of the project. The details are as follows:

[0084] Performance Benchmarks:

[0085] ;

[0086] in, Score performance; The response time is the time required for the system to process a single request, usually in milliseconds. For example, the average response time is ,but ; Throughput refers to the number of requests that a system can handle per unit time. For example, in a web service, if 1,000 requests can be processed per second, then ; is the stability factor, which is an adjustment factor used to reflect the performance of the system under high load. If the system can remain stable under high load and the response time does not increase much, you can set If your system responds significantly faster or experiences errors under heavy load, you may want to lower this value, e.g. to ;

[0087] Website activity:

[0088] ;

[0089] in, Score activity; is the number of stars, the number of stars a project has on GitHub, which reflects the popularity of the project. For example, if a project has 5,000 stars, then ; is the number of branches, the number of forks of a project, indicating how many developers are willing to contribute. If a project has 2,000 forks, then ; The number of closed issues is the number of issues that have been resolved, which reflects the maintainer's ability to solve problems. If a project has 1,000 issues closed, then ; is the total number of issues, the number of all issues reported in the project, including closed and unclosed issues. If there are 1500 issues, then ;

[0090] Developer familiarity:

[0091] ;

[0092] in, Score points for skill proficiency; is the number of skills; For the Each team member's self-assessment of the technical points they have mastered is usually from 1 (completely ignorant) to 5 (very proficient). For example, for JavaScript, a developer's self-assessment is 4; For the The importance of a skill is the importance of that skill to the success of the project, also from 1 (not important) to 5 (very important). For example, if the project mainly uses JavaScript, then its Maybe 5; ;

[0093] Scalability:

[0094] ;

[0095] in, Score scalability; Rate the code quality (a value between 1 and 5, with 5 being very high code quality); Rate the quality of the documentation (a value between 1 and 5, with 5 being very good); Rate new feature support (a value between 1 and 5, with 5 indicating that the community is actively adding new features);

[0096] Overall rating:

[0097] ;

[0098] in, For comprehensive rating; weights for scoring performance; is the weight of the activity score; The weight of the skill proficiency score; The weighting for scoring scalability;

[0099] In S1, in the process of comprehensive consideration based on multiple dimensions, time constraints are introduced for optimization. After optimization, the specific results are:

[0100] Performance Benchmarks:

[0101] ;

[0102] in, Score the optimized performance; Resource utilization (e.g. CPU usage); is the performance factor that decays over time; It is a prediction model trained with historical performance data and used to estimate the performance; the optimized performance score can more accurately reflect the system's changing trend over time and can predict future performance, which helps to adjust the architecture in time to deal with possible performance degradation or bottlenecks.

[0103] Website activity:

[0104] ;

[0105] in, is the optimized activity score; for problem solving efficiency; scoring the diversity of contributors; is the activity factor that decays over time; According to the current time A dynamically adjusted weight matrix is ​​used to balance the importance of various activity indicators. The optimized activity score increases the problem-solving efficiency and contributor diversity scores, and uses a dynamic weight matrix to balance the importance of different indicators, making the assessment of community activity more comprehensive and accurate.

[0106] Developer familiarity:

[0107] ;

[0108] in, The optimized skill proficiency score; is the teamwork index, which reflects the degree of cooperation among team members; Weighted scores for project-related experience; is a skill proficiency factor that grows over time; the optimized skill proficiency score better reflects the true technical level of team members and their potential contribution to the project, helping to ensure that the selected technology stack is one that the team can use and maintain efficiently.

[0109] Scalability:

[0110] ;

[0111] in, Score the optimized scalability; is the code reuse rate; Update frequency for the community; is the scalability factor that grows over time; It is a prediction model trained with historical data and used to estimate the future moment The optimized scalability score can better predict the future development potential and adaptability of the technology stack. This is conducive to building an architecture that is both stable and easy to expand.

[0112] Overall rating:

[0113] ;

[0114] in, is the comprehensive score after optimization;

[0115] The optimized formula improves the accuracy, foresight and practicality of the technology stack selection process through more detailed considerations, dynamically adjusted weights and prediction models based on historical data. It not only helps identify the current optimal technical solution, but also predicts possible problems in the future, providing a solid foundation for continuous optimization and improvement. This method makes the construction of adaptive website architecture more scientific and reasonable, and enhances the stability, maintainability and long-term development capabilities of the system.

[0116] S2. Based on the selected optimal technology stack, define an adaptability function to measure the matching degree of the existing modules of the website to the user goals, and then quantify the adaptability. At the same time, define a performance function to measure the page loading speed, and then quantify the loading performance of the website. Finally, evaluate the website architecture characteristics based on adaptability and loading performance. In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user goals, introduce the time decay factor and the module participation index for optimization. In the process of defining the performance function to measure the page loading speed, and then quantify the loading performance of the website, introduce resource optimization and network condition changes for optimization;

[0117] In S2, the adaptability function is defined to measure the matching degree of the existing modules of the website to the user's goals, and then quantify the adaptability, as follows:

[0118] ;

[0119] in, is the fitness function; is the number of modules in the website; For user goals; For the The weight of each module; For modules and user needs similarity; For the website modules;

[0120] In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user's goals, the time decay factor and the module participation index are introduced for optimization. After optimization, the specific results are as follows:

[0121] ;

[0122] in, is the optimized fitness function; is the adjustment coefficient, which is used to control the influence of participation on adaptability score; For users in the module Engagement metrics on the site (e.g. dwell time, clicks, etc.); It is the highest engagement value among all modules; is the adjustment coefficient, which is used to adjust the impact of the most recently used module on the adaptability score; is a time decay factor, indicating the time from the last time the module was used The time interval to the current time can usually be defined as ,in is the time interval, is the decay rate parameter;

[0123] By introducing user engagement and normalizing it to In the range, this formula increases the influence of users’ actual interaction behavior on the adaptability evaluation. High engagement means that users are more interested in the module or use it more frequently, so they should be given a higher score; It reflects the "recent use" principle, that is, the modules that users have recently visited may be more in line with current needs. This item uses exponential decay to make the modules that have been visited more recently get higher scores. and As adjustable parameters, the influence of participation and time factors on the final adaptability score can be flexibly adjusted according to specific application scenarios; the optimized adaptability function not only considers the basic similarity between the module and user needs, but also combines the user's actual behavior data and time factors, making the adaptability evaluation more comprehensive and dynamic. This helps to provide a more personalized and responsive service experience.

[0124] User Goals The determination is as follows:

[0125] Collect behavioral data about users’ interactions with the website through website logs or user behavior monitoring tools , contextual data and historical data , extracting behavioral pattern features related to target prediction from the collected data , contextual features and historical preference characteristics , based on the features , build a classification model to predict the target:

[0126] ;

[0127] in, For a given feature When the target user The probability distribution of is the model weight, which is used to transform the feature vector Mapping to the probability space of different targets; is a bias term used to adjust the output of the model;

[0128] Select the target with the highest probability as the current prediction:

[0129] ;

[0130] in, To select the target with the highest probability as the current predicted target;

[0131] The ability to dynamically adjust prediction results based on user behavior patterns, contextual information, and historical preferences means that even if user behavior changes, the system can respond in a timely manner and provide the most relevant content and services. Classification models built on large amounts of data can capture subtle data patterns and trends, thereby providing strong support for business decisions. For example, advertising delivery strategies can be optimized based on user behavior patterns to improve conversion rates.

[0132] In S2, a performance function is defined to measure the page loading speed and thus quantify the loading performance of the website, as follows:

[0133] ;

[0134] in, is the performance function; is the current page architecture; The total size of the current page architecture; The network speed of the current user;

[0135] In the process of defining performance functions to measure page loading speed and then quantifying the loading performance of the website, we introduce resource optimization and changes in network conditions for optimization. After optimization, the specific results are:

[0136] ;

[0137] in, is the optimized performance function; The size of the resources that have been cached in the page. By subtracting this size, the actual amount of data that needs to be transmitted can be more accurately reflected; The network acceleration factor can be the bandwidth improvement brought by CDN (Content Distribution Network) or the transmission efficiency improvement brought by other technical means (such as HTTP / 2 multiplexing); Network latency, which reflects the time from when a request is sent to when the first byte is received, has a significant impact on user experience. It is an adjustment item based on resource priority, which is used to consider the impact of resource priority loading on the critical rendering path on the overall perceived performance;

[0138] The introduction of means that when evaluating page load performance, resources that are already in the user's browser cache are taken into account and do not need to be downloaded again, thus reducing the amount of data actually transferred; It can represent the improvement in network transmission efficiency brought about by using CDN, compression algorithms (such as Gzip or Brotli), preloading and other technical means. This helps to more accurately predict page loading time; Consider the impact of network latency on page loading speed. Even if the network speed is fast, if the server responds slowly or the DNS resolution takes a long time, it will significantly increase the page loading time; It reflects the importance of the order in which different resources are loaded. For example, loading the key parts of CSS and JavaScript first can make the page present the main content faster and improve the user experience. This can be quantified according to the importance and loading order of resources. Through such optimization, the performance function can not only more accurately reflect the actual performance of page loading, but also better guide developers on how to optimize the page to improve the user experience.

[0139] In S2, the website architecture characteristics are evaluated based on adaptability and loading performance. The specific steps are as follows:

[0140] S21. Based on the adaptability function, calculate the adaptability score of each module, and obtain the overall adaptability score of the entire website by taking a weighted average of the adaptability scores of different modules;

[0141] S22. Use the optimized performance function to evaluate the loading performance of each page in detail, identify the key factors affecting the performance, and make improvement suggestions;

[0142] S23. Combine the adaptability score and the performance evaluation results to generate a comprehensive evaluation report, which includes the adaptability score of each module, detailed indicators of page loading performance, and the overall comprehensive score of the website.

[0143] S3. Calculate the optimization objective function for maximizing user experience based on the evaluation results of the website architecture characteristics, generate an optimization plan, and generate a new architecture of the website based on the optimization plan, and dynamically present it to the user;

[0144] In S3, based on the architectural characteristics, the optimization goals of maximizing user experience and minimizing loading time are calculated as follows:

[0145] ;

[0146] in, To maximize user experience; The importance of adaptability in the overall goal; To understand the importance of loading performance in the overall goal; The importance weight of external factors in the overall goal; is the impact weight of the device type; Device factor is an adjustment factor defined based on the user device type (e.g. mobile device, desktop device); is the impact weight of network conditions; Network factor, which is an adjustment factor defined according to network conditions (such as 4G, 5G, Wi-Fi); is the impact weight of the time period; Time factor, which is an adjustment factor defined according to time periods (such as peak hours and off-peak hours);

[0147] By introducing external factors such as device type, network conditions, and time periods, the optimization objective function not only takes into account the degree of match between website modules and user needs and page loading speed, but also dynamically adapts to changes in different external environments. This comprehensive approach enables the website to provide the best user experience in various situations, for example, focusing more on performance optimization on mobile devices and enhancing caching strategies during peak hours to ensure that users can get a fast and relevant experience regardless of the device they use or the network conditions they are in. This approach improves the flexibility and responsiveness of the website, helping to improve overall user satisfaction and business performance.

[0148] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for building an adaptive website architecture based on artificial intelligence, characterized in that: The steps include: S1. Clarify the website's goals, target user groups, and their needs, develop a project plan, and then comprehensively consider multiple technology stacks based on multiple dimensions to select the optimal technology stack. In the process of comprehensive consideration based on multiple dimensions, introduce time constraints for optimization; S2. Based on the selected optimal technology stack, define an adaptability function to measure the matching degree of the existing modules of the website to the user goals, and then quantify the adaptability. At the same time, define a performance function to measure the page loading speed, and then quantify the loading performance of the website. Finally, evaluate the website architecture characteristics based on adaptability and loading performance. In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user goals, introduce the time decay factor and the module participation index for optimization. In the process of defining the performance function to measure the page loading speed, and then quantify the loading performance of the website, introduce resource optimization and network condition changes for optimization; S3. Calculate the optimization objective function for maximizing user experience based on the evaluation results of the website architecture characteristics, generate an optimization plan, and generate a new architecture of the website based on the optimization plan, and dynamically present it to the user; The multiple dimensions include performance benchmarks, community activity, developer familiarity, and scalability, as follows: Performance Benchmarks: ； in, Score performance; is the response time; is the throughput; is the stability coefficient; Website activity: ； in, Score activity; is the number of stars; is the number of branches; The number of issues closed; is the total number of questions; Developer familiarity: ； in, Score points for skill proficiency; is the number of skills; For the skill level; For the The importance of a skill; ; Scalability: ； in, Score scalability; Rate the code quality; Rate the quality of the document; Score new feature support; Overall rating: ； in, For comprehensive rating; weights for scoring performance; is the weight of the activity score; The weight of the skill proficiency score; The weighting for scoring scalability; Among them, the time limit is introduced for optimization, and the specific results after optimization are as follows: ； in, Score the optimized performance; for resource utilization; is the performance factor that decays over time; It is a prediction model trained with historical performance data and used to estimate the performance; ； in, is the optimized activity score; for problem solving efficiency; scoring the diversity of contributors; is the activity factor that decays over time; According to the current time A dynamically adjusted weight matrix to balance the importance of each activity indicator; ； in, The optimized skill proficiency score; is the teamwork index; Weighted scores for project-related experience; is the skill proficiency factor that grows over time; ； in, Score the optimized scalability; is the code reuse rate; Update frequency for the community; is the scalability factor that grows over time; It is a prediction model trained with historical data and used to estimate the future moment Scalability performance; ； in, It is the comprehensive score after optimization.

2. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 1, characterized in that: In S1, the project plan includes schedule, budget and resource allocation.

3. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 2, characterized in that: In S1, the technology stack includes a front-end framework, a back-end framework, a database system, and a cloud computing platform.

4. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 3 is characterized in that: In S2, the adaptability function is defined to measure the matching degree of the existing modules of the website to the user's goal, and then quantify the adaptability, as follows: ； in, is the fitness function; is the number of modules in the website; For user goals; For the The weight of each module; For modules and user needs similarity; For the website modules; In the process of defining the adaptability function to measure the matching degree of the existing modules of the website to the user's goals, the time decay factor and the module participation index are introduced for optimization. After optimization, the specific results are as follows: ； in, is the optimized fitness function; is the adjustment coefficient, which is used to control the influence of participation on adaptability score; For users in the module Engagement metrics on It is the highest engagement value among all modules; is the adjustment coefficient, which is used to adjust the impact of the most recently used module on the adaptability score; is a time decay factor.

5. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 4 is characterized in that: The user goal The determination is as follows: Collect behavioral data about users’ interactions with the website through website logs or user behavior monitoring tools , contextual data and historical data , extracting behavioral pattern features related to target prediction from the collected data , contextual features and historical preference characteristics , based on the features , build a classification model to predict the target: ； in, For a given feature When the target user The probability distribution of is the model weight; is the bias term; Select the target with the highest probability as the current prediction: ； in, To select the target with the highest probability as the current predicted target.

6. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 5, characterized in that: In S2, a performance function is defined to measure the page loading speed, thereby quantifying the loading performance of the website, as follows: ； in, is the performance function; is the current page architecture; The total size of the current page architecture; The network speed of the current user; In the process of defining performance functions to measure page loading speed and then quantifying the loading performance of the website, we introduce resource optimization and changes in network conditions for optimization. After optimization, the specific results are: ； in, is the optimized performance function; The size of the resources that have been cached in the page; is the network acceleration factor; is the network delay time; It is an adjustment item based on resource priority, which is used to consider the impact of resource priority loading on the critical rendering path on the overall perceived performance.

7. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 6, characterized in that: In S2, the website architecture characteristics are evaluated based on adaptability and loading performance, and the specific steps are as follows: S21. Based on the adaptability function, calculate the adaptability score of each module, and obtain the overall adaptability score of the entire website by taking a weighted average of the adaptability scores of different modules; S22. Use the optimized performance function to evaluate the loading performance of each page in detail, identify the key factors affecting the performance, and make improvement suggestions; S23. Combine the adaptability score and the performance evaluation results to generate a comprehensive evaluation report, which includes the adaptability score of each module, detailed indicators of page loading performance, and the overall comprehensive score of the website.

8. The method for building an adaptive website architecture based on artificial intelligence drive according to claim 7, characterized in that: In S3, the optimization goals of maximizing user experience and minimizing loading time are calculated based on the architectural characteristics, as follows: ； in, To maximize user experience; The importance of adaptability in the overall goal; To understand the importance of loading performance in the overall goal; The importance weight of external factors in the overall goal; is the impact weight of the device type; is the equipment factor; is the impact weight of network conditions; is the network factor; is the impact weight of the time period; is the time factor.

Images