Optimization method and system for intelligently adjusting user experience
Through layered architecture design and intelligent analysis, the technical challenges in personalized user interface adaptation were solved, enabling real-time analysis of user behavior and personalized interface recommendations, thereby improving user experience and system scalability while reducing operating costs.
Patent Information
- Application Number
- CN202510927650.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-07
- Publication Date
- 2025-10-17
AI Technical Summary
Existing technologies in user interface personalization adaptation have problems such as insufficient adaptation accuracy, high rule maintenance costs, high real-time inference latency, poor interpretability of algorithm decisions, single data collection dimension, prominent performance bottlenecks, and limited system scalability, resulting in fragmented user experience and high development costs.
By intelligently adjusting user experience optimization methods, including user behavior data collection, tag generation, and dynamic interface configuration, and employing a layered architecture design, it utilizes a lightweight SDK, real-time event pipeline, real-time stream processing and offline batch processing, clustering algorithms, and classification models to achieve real-time analysis of user behavior and personalized interface recommendations.
It enables precise and personalized adjustments to the user interface, reduces development and maintenance costs, improves the immediacy and accuracy of the user experience, enhances system scalability and user retention, and reduces reliance on manual operations.
Smart Images
Figure CN120803447A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of modern software engineering processing, in particular to an intelligent adjustment user experience optimization method and system. BACKGROUND
[0002] In the field of modern software engineering, the personalization of user interfaces has become a core requirement for improving user experience. With the popularization of multi-terminal and multi-scene applications, enterprises are generally faced with the contradiction between user interface customization and development efficiency.
[0003] The existing technical solutions mainly include three types: static template solution, rule engine solution and machine learning solution. The static template solution realizes basic adaptation by predefining a limited number of interface templates, but has the problems of insufficient adaptation accuracy and lack of dynamic adjustment capability; the rule engine solution performs conditional rendering based on predefined if-then rules, which can achieve a certain degree of flexibility, but requires modification of 15 associated rules for each new user tag, resulting in high rule maintenance cost and poor cold start effect for new users; the machine learning solution predicts user preferences through algorithms such as collaborative filtering, but has the defects of high real-time reasoning delay and poor algorithm decision interpretability, making it difficult to meet the demand for rapid response.
[0004] The current technical solution has the following problems in data collection, adaptation strategy and performance: the data collection dimension is single, most systems only rely on clickstream data, ignoring composite indicators such as dwell time and scroll depth, and not combining with environmental context information such as device performance and network conditions, resulting in incomplete user behavior analysis; secondly, the adaptation strategy is coarse-grained, with the smallest adjustment unit being a complete page, which cannot achieve fine-grained layout optimization, and lacks quantitative feedback loops between user satisfaction and interface adjustment, resulting in low strategy iteration efficiency; finally, the performance bottleneck is prominent, dynamic style injection leads to a decline in client rendering performance, and traditional relational databases cannot support real-time analysis of TB-level user behavior data. The above problems collectively result in fragmented user experience, high development labor costs, and limited system scalability, making it difficult to meet the rapidly growing demand for personalization. SUMMARY
[0005] To solve the above problems, the present application provides the following technical solutions: An intelligent adjustment user experience optimization method, characterized by comprising the following steps: S1, a user logs into the system; S2, the system identifies user behavior data and tags the user with behavior tags based on the user's behavior data; S3, the backend adjusts the generation logic and application scenarios of the behavior tags in real time based on the user behavior data, user feedback data and interaction result data, and generates adjustment parameters; S4, the front end dynamically configures the layout and design of the front end interface according to the adjustment returned by the back end, and automatically displays the interface and interaction preferred by the user.
[0006] Based on the above technical solutions, the application can also be improved as follows.
[0007] Further, the step S2 further comprises the following steps: S2-1, the system judges whether the user has a user behavior label; S2-2, when the user does not have a behavior label, the system starts recording user behavior points; S2-3, record the behavior and preference of the user, and match the behavior label for the key words of the behavior and preference whose data volume exceeds the preset value; S2-4, label matching is completed for the content body of the label; S2-5, make behavior label rules; S2-6, configure the corresponding behavior label for the user's behavior according to the formulated label rules.
[0008] Further, in the step S2-2, when the user has a behavior label, directly jump to step S2-6.
[0009] The intelligent adjustment user experience optimization system is used to realize the above method, and is characterized by comprising: A data collection layer for collecting user behavior data; A label calculation layer for preliminary judgment of the behavior label of the user's instant behavior, and dynamic adjustment of the label rules; An intelligent analysis layer for deep analysis and dynamic evaluation of user behavior data, and real-time generation of adjustment parameters capable of adjusting the behavior label and its application scenario; An application layer for displaying personalized content recommendation matched with user preferences; A user interaction behavior preference database for storing user behavior data labeled by the user;
[0010] Further, the behavior data includes but is not limited to user clicks, browsing path, and dwell time data.
[0011] Further, the data collection layer can integrate lightweight SDK in the front end (Web / App), automatically capture user clicks, browsing path, dwell time and other interaction data, the data collection layer uses Kafka or RabbitMQ to build a real-time event pipeline as an event bus architecture, and the data collection layer uses Apache NiFi and / or custom ETL process to clean and standardize the original behavior data.
[0012] Further, the label calculation layer real-time processing engine is Apache Flink, and the label calculation layer batch processing system is Spark, which can build an offline calculation pipeline and periodically process historical data to optimize labels.
[0013] Further, the label calculation layer uses Drools as a rule engine, which can realize configurable label rule management and support business personnel to dynamically adjust rules.
[0014] Further, the intelligent analysis layer builds a user feature vector based on PySpark / SciKit-Learn, and uses a clustering algorithm model for automatically identifying different feature user groups, a classification model for predicting and labeling potential preference labels of users, and a time series model for capturing patterns of evolution of user behavior over time.
[0015] Further, the application layer uses a label service API and an application real-time recommendation engine. Advantages
[0016] Compared with the prior art, the application can accurately capture user operation habits and potential needs by collecting user behavior data in real time and generating behavior labels, thereby realizing dynamic adjustment of interface layout and design during user interaction and significantly improving the immediacy and accuracy of personalized experience. Compared with traditional static template solutions or rule engines, the application does not need to preset fixed interface templates or rely on complex rule maintenance, but directly associates user behavior features with interface configuration through intelligent data processing and analysis, thereby greatly reducing the human cost of development and maintenance.
[0017] In terms of technical implementation, the system realizes efficient data processing and strategy iteration through hierarchical architecture design. The data collection layer uses lightweight SDK and real-time event pipeline to ensure comprehensive collection and low-latency transmission of user behavior data. The label calculation layer combines real-time stream processing and offline batch processing to support dynamic adjustment and optimization of label rules, which meets the real-time needs of high-frequency interaction scenarios and guarantees the deep mining ability of historical data. The intelligent analysis layer cooperatively applies clustering algorithms, classification models and time series models to automatically identify user group features, predict preference labels and capture behavior evolution trends, thereby providing scientific basis for interface optimization. This multi-dimensional and multi-level analysis capability enables the system to adapt to dynamic changes in user needs and avoid experience fragmentation caused by rigid strategies.
[0018] In addition, the application further enhances the accurate matching ability of personalized content through the integration of the tag service API and the real-time recommendation engine. The system can not only quickly generate basic tags when the user first interacts, but also continuously optimize the tag system in subsequent use, realizing seamless connection from cold start to fine operation. This dynamic evolution mechanism significantly improves user retention rate and product stickiness. At the same time, through automatic rule configuration and model iteration, the dependence on manual operation is reduced, and the long-term maintenance cost is reduced. BRIEF DESCRIPTION OF DRAWINGS
[0019] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor.
[0020] Figure 1 Flowchart of the application; DETAILED DESCRIPTION
[0021] In order to facilitate the understanding of the present application, the following will be described in more detail with reference to the related drawings. The drawings show several embodiments of the present application. However, the present application can be realized in many different forms, and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.
[0022] Please refer to Figure 1 The intelligent adjustment user experience optimization method of the present application comprises the following steps: S1, the user logs in the system; S2, the system identifies user behavior data, and labels the user according to the user's behavior data; S2-1, the system judges whether the user has user behavior label; S2-2, when the user does not have behavior label, the system starts recording user behavior burying point; S2-3, record the user's behavior and preference, and match the behavior label for the key words of the behavior and preference whose data volume exceeds the preset value; In step S2-2, when the user has behavior label, go to step S2-6 directly; S2-4, label matching is completed, and the label is labeled with content body; S2-5, make behavior label rules; S2-6, configure the corresponding behavior label for the user's behavior according to the formulated label rules.
[0023] S3, the backend adjusts the generation logic and application scenarios of the behavior label in real time according to the user behavior data, user feedback data and interaction result data, and generates adjustment parameters; S4, the front end dynamically configures the layout and design of the front-end interface according to the adjustment returned by the back end, and automatically displays the interface and interaction preferred by the user.
[0024] Embodiment 1 An intelligent adjustment user experience optimization system for implementing an intelligent adjustment user experience optimization method, comprising a data collection layer for collecting user behavior data, a label calculation layer, an intelligent analysis layer, an application layer, and a user interaction behavior preference database for storing user behavior data labeled by the user.
[0025] Specifically, the label calculation layer is used for preliminary judgment of the behavior label of the user's instant behavior and dynamic adjustment of the label rule. The label calculation layer combines real-time stream processing and offline batch processing to support dynamic adjustment and optimization of the label rule, which not only meets the real-time demand of high-frequency interaction scenarios, but also guarantees the deep mining ability of historical data. The intelligent analysis layer is used for deep analysis and dynamic evaluation of user behavior data. The intelligent analysis layer can automatically identify user group characteristics, predict preference labels and capture behavior evolution trends through the collaborative application of clustering algorithms, classification models and time series models, providing scientific basis for interface optimization, and generating adjustment parameters for adjusting behavior labels and their application scenarios in real time. The application layer is used to display personalized content recommendations matched with user preferences.
[0026] Through hierarchical architecture design, the system realizes full-process automation of data collection, label generation, intelligent analysis and content recommendation. The data collection layer ensures comprehensive collection and low-latency transmission of user behavior data through lightweight SDK and real-time event pipeline. This multi-dimensional and multi-level analysis capability enables the system to adapt to dynamic changes in user needs and avoid experience fragmentation caused by rigid strategies. Embodiment
[0027] Unlike Embodiment 1, the behavior data includes but is not limited to user clicks, browsing paths, and dwell time data. The data collection layer can integrate lightweight SDK in the front end (Web / App) to automatically capture user interaction data such as clicks, browsing paths and dwell time. The data collection layer uses Kafka or RabbitMQ to build a real-time event pipeline as an event bus architecture, and uses Apache NiFi and / or custom ETL processes to clean and standardize raw behavior data.
[0028] Through real-time event pipeline and standardized processing, the system can efficiently handle the parallel processing demand of 100,000+ behavior events per second, ensuring the completeness and real-time nature of data collection. The cleaned data provides high-quality input for subsequent label calculation and intelligent analysis, while supporting cross-terminal and cross-scene user behavior tracking, laying a data foundation for personalized recommendation and interface adaptation. Embodiment
[0029] Unlike Embodiment 2, the label calculation layer real-time processing engine is Apache Flink, and the batch processing system of the label calculation layer is Spark, which can build an offline computing pipeline to periodically process historical data to optimize labels. The label calculation layer uses Drools as the rule engine, which can realize configurable label rule management and support business personnel to dynamically adjust rules.
[0030] Through the collaborative application of Apache Flink and Spark, the system realizes seamless connection between real-time and offline processing. Flink is responsible for real-time stream processing, ensuring that label generation delay is <500ms, meeting the immediacy needs of high-frequency interaction scenarios; Spark optimizes label logic through periodic offline computation, improving label coverage to 90%+ automation. The introduction of the Drools rule engine allows business personnel to dynamically adjust label rules without relying on the development team, shortening the online period of new labels from weekly to hourly, significantly reducing operational costs and improving system flexibility. Embodiment
[0031] Unlike Embodiment 3, on the basis of Embodiment 3, the intelligent analysis layer constructs a user feature vector based on PySpark / SciKit-Learn, the intelligent analysis layer uses a clustering algorithm model for automatically identifying different characteristic user groups, a classification model for predicting and tagging user's potential preference labels, and a time series model for capturing the evolution of user behavior over time, the application layer uses a label service API, and the application uses a real-time recommendation engine.
[0032] Through the clustering algorithm model, the system can automatically identify the characteristics of 15%+ emerging user groups, providing precise label matching for cold-start users; the classification model predicts potential preferences based on user historical behavior, improving recommendation CTR by 40-60%; the time series model dynamically adjusts label generation logic by capturing the time evolution of user behavior, ensuring long-term consistency between recommended content and user needs. The label service API of the application layer combined with the real-time recommendation engine further enhances the precision matching ability of personalized content, increasing user retention rate by 25%, while automated rule configuration and model iteration reduce the dependence on manual operation and reduce long-term maintenance costs.
[0033] The method implemented when the intelligent adjustment user experience optimization system is executed can refer to the embodiments of the intelligent adjustment user experience optimization system of the present application, which will not be described herein.
[0034] It should be noted that in this document, the terms "comprise", "comprising", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not only include those elements, but can also include other elements not expressly listed or inherent to such process, method, article, or system. Without more limitations, the element defined by the statement "comprises a" does not exclude the presence of additional identical elements in the process, method, article, or system that includes the element.
[0035] The above-mentioned embodiment numbers of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.
[0036] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.
[0037] From the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be realized by means of software and necessary general hardware platforms, of course, they can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the methods described in the various embodiments of the present application.
[0038] The above merely describes preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which are made by using the content of the present application specification and drawings, are also included in the patent protection scope of the present application The above merely describes preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which are made by using the content of the present application specification and drawings, are also included in the patent protection scope of the present application
Claims
1. Intelligent adjustment of user experience optimization method, characterized by: The steps include: S1. User logs into the system; S2. The system identifies user behavior data and assigns behavior tags to users based on their behavior data. S3: The backend adjusts the generation logic and application scenarios of behavior tags in real time based on user behavior data, user feedback data, and interaction result data, and generates adjustment parameters; S4. The front-end dynamically configures the layout and design of the front-end interface based on the adjustments returned by the back-end, and automatically displays the interface and interaction preferred by the user.
2. The method for intelligently adjusting user experience optimization according to claim 1, characterized in that: The step S2 further comprises the following steps: S2-1, the system determines whether the user has a user behavior tag; S2-2: When the user does not have a behavior tag, the system starts recording the user's behavior tracking points; S2-3. Record the user's behavior and preferences, and match behavior tags to keywords for behaviors and preferences whose data volume exceeds a preset value; S2-4, after tag matching is completed, the content body is marked on the tag; S2-5. Establish behavioral labeling rules; S2-6. Configure corresponding behavior labels for user behaviors according to the established labeling rules.
3. The method for intelligently adjusting user experience optimization according to claim 1, characterized in that: In step S2-2, when the user has a behavior tag, the process directly jumps to step S2-6.
4. An intelligent adjustment user experience optimization system, which is used to implement any one of the methods described in claims 1 to 3, characterized in that: include: Data collection layer, which is used to collect user behavior data; The tag calculation layer is used to make preliminary judgments on the user's immediate behavior tags and dynamically adjust the tag rules; The intelligent analysis layer is used to conduct in-depth analysis and dynamic evaluation of user behavior data, and to generate adjustment parameters in real time to adjust behavior tags and their application scenarios; The application layer is used to display personalized content recommendations that match user preferences; The user interaction behavior preference database is used to store user behavior data that has been labeled by users.
5. The intelligent adjustment user experience optimization system according to claim 4, characterized in that: The behavioral data includes but is not limited to user clicks, browsing paths, and stay time data.
6. The intelligent adjustment and user experience optimization system according to claim 4, characterized in that: The data collection layer can integrate a lightweight SDK on the front end (Web / App) to automatically capture interaction data such as user clicks, browsing paths, and dwell time. The data collection layer uses Kafka or RabbitMQ to build a real-time event pipeline as the event bus architecture, and the data collection layer uses Apache NiFi and / or custom ETL processes to clean and standardize the raw behavioral data.
7. The intelligent adjustment and user experience optimization system according to claim 4, characterized in that: The real-time processing engine of the label calculation layer is Apache Flink, and the batch processing system of the label calculation layer is Spark, which can build an offline computing pipeline and periodically process historical data to optimize labels.
8. The intelligent adjustment and user experience optimization system according to claim 4, characterized in that: The tag calculation layer uses Drools as the rule engine, which can implement configurable tag rule management and support business personnel to dynamically adjust rules.
9. The intelligent adjustment and user experience optimization system according to claim 4, characterized in that: The intelligent analysis layer constructs user feature vectors based on PySpark / SciKit-Learn. The intelligent analysis layer adopts a clustering algorithm model for automatically identifying user groups with different characteristics, a classification model for predicting and labeling users' potential preferences, and a time series model for capturing the patterns of user behavior evolving over time.
10. The intelligent adjustment user experience optimization system according to claim 4, characterized in that: The application layer adopts the tag service API and applies a real-time recommendation engine.