Automatic page point location configuration method and device, storage medium and computer equipment

Abstract

The invention relates to the technical field of computers, and discloses a page point location automatic configuration method and device, a storage medium and computer equipment, and the method comprises the steps: collecting a page picture of a display page, carrying out the feature recognition of the page picture to locate a target point location, carrying out the recognition and analysis of the display page, obtaining a point location keyword, and carrying out the matching with the target point location. Wherein the point position keyword is associated with a keyword editing area, then collecting a touch screen operation of a user for a target point position, obtaining and analyzing response data of the target point position, and finally, performing iterative optimization on the target point position and the point position keyword based on a response analysis result to update a display page. According to the method, the point locations are automatically set, the labor cost is reduced, the point location setting efficiency is improved, the point locations and the medical service scene are highly fused, the system fault tolerance is improved due to the fact that the point location keywords can be edited, an operator can optimize the point locations and the page according to the data analysis result of point location clicking, and the applicability of the display page in the medical field is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and more particularly to a method, apparatus, storage medium, and computer device for automatically configuring page positions. Background Art

[0002] CMS (Content Management System) is a software system used to create, edit, and organize digital content and publish it to the network. Specifically applied to the field of medical and health, the CMS system can specifically manage medical service information, drug information, health consultations, etc.

[0003] Dotting operation, that is, data collection is performed on the exposure and click operations of users in the H5 page module, and then reported uniformly for operation to perform data analysis on the placed pages. Specifically in the CMS system, when the operation builds the CMS page, the positions to be collected are predefined. For example, in the scenario of selling drugs in an online pharmacy, the predefined position of the drug picture module is: banner, the predefined position of the button module for the medication guide is: confirm, and the predefined positions of the online consultation and immediate purchase modules are: button. After the page is published, the dotting takes effect, and the data of these preset positions triggered by user operations are collected. After being reported to the system, they are uniformly summarized and analyzed, and then the operation personnel are notified immediately to adjust the page or update the content.

[0004] Currently, the CMS system performs position preset mainly through the following two methods. One is to manually identify the areas to be collected by the user and perform position preset manually. The disadvantage of this method is that the labor cost is relatively high, and the setting of all positions depends on manual definition, and there are scenarios of omission. The other is that the CMS system automatically performs position setting, but the finally preset positions have no business logic, can only recognize the module type and random number form, are not easy to understand, do not carry business attributes, and are easy to be confused. Summary of the Invention

[0005] In view of this, the present application provides a method, apparatus, storage medium, and computer device for automatically configuring page positions, mainly aiming to solve the technical problems in the prior art such as low efficiency of the position setting method, high labor cost, and chaotic logic that cannot match business attributes.

[0006] According to the first aspect of the present invention, there is provided a method for automatically configuring page positions, the method comprising:

[0007] Collect a page image of the display page, perform feature recognition on the page image to determine target features, and locate target positions based on the target features;

[0008] Identify and analyze the display page to obtain point keywords, and match the point keywords with the target point, where the point keywords are associated with a keyword editing area;

[0009] Collect the touch screen operations of the user for the target point to obtain the response data of the target point, and analyze the response data to obtain the response analysis result of the target point;

[0010] Iteratively optimize the target point and the point keyword corresponding to the target point based on the response analysis result to update the display page.

[0011] Optionally, the feature recognition of the page image to determine the target feature and the positioning of the target point based on the target feature include: preprocessing the page image, and performing feature recognition on the preprocessed page image to extract the target feature, where the target feature includes text feature, shape feature and color feature; generating a feature data set according to the target feature, inputting the feature data set into a preset point recognition model to obtain point area data; positioning the target point based on the point area data.

[0012] Optionally, the identifying and analyzing the display page to obtain point keywords includes: obtaining the page image of the display page, preprocessing the page image, and performing text extraction on the preprocessed page image to obtain a first keyword segment; performing feature recognition on the preprocessed page image and extracting effective features, where the effective features include color feature, texture feature and shape feature; generating a feature training set based on the effective features, training a preset keyword generation model using the feature training set, and obtaining a second keyword segment output by the keyword generation model; splicing the first keyword segment and the second keyword segment based on the first weight coefficient corresponding to the first keyword segment and the second weight coefficient corresponding to the second keyword to obtain the point keyword.

[0013] Optionally, the identifying and analyzing the display page to obtain point keywords includes: determining the page components in the display page, obtaining the source code corresponding to the page components, and parsing the source code; traversing the abstract syntax book in the parsed source code, extracting the target tags and the comment content corresponding to the target tags; performing magnetic annotation on the comment content using a preset natural language toolkit to generate a plurality of original keywords; splicing the plurality of original keywords based on a preset symbol to obtain the point keyword.

[0014] Optionally, the response data includes response frequency data; after analyzing the response data to obtain the response analysis result of the target point, the method further includes: obtaining the response frequency data of the target point in the response analysis result, and the response frequency data of the non-target point area in the display page; when there is response frequency data in the non-point area greater than or equal to the response frequency data of the target point, marking the non-point area and generating a point prompt message, where the prompt message is used to indicate adding a target point to the marked non-point area.

[0015] Optionally, the iterative optimization of the target point and the point keyword corresponding to the target point based on the response analysis result includes: obtaining the response analysis result based on a preset time interval, where the response analysis result includes multiple time dimensions and the point click data corresponding to the time dimension; generating a click data trend chart according to the multiple time dimensions and the point click data corresponding to the time dimension, and comparing the point click data with preset click data; when there is point click data lower than the preset click data, marking the target point corresponding to the point click data as a prompt point and generating an operation prompt message, where the operation prompt message is used to indicate updating or taking off the shelf of the prompt point.

[0016] Optionally, the iterative optimization of the target point and the point keyword corresponding to the target point based on the response analysis result includes: obtaining the page size of the display page and the point size of the target point; determining the display device of the display page and obtaining the screen size of the display device; calculating the proportional coefficient between the screen size and the page size, and adjusting the page size to the screen size to make the display page match the display device; adjusting the point size based on the proportional coefficient to make the target point match the display page.

[0017] According to the second aspect of the present invention, a page point automatic configuration device is provided. The device includes:

[0018] A point positioning module, configured to collect a page picture of a display page, perform feature recognition on the page picture, determine a target feature, and locate a target point based on the target feature;

[0019] A keyword generation module, configured to perform recognition and analysis on the display page to obtain a point keyword, and match the point keyword with the target point, where the point keyword is associated with a keyword editing area;

[0020] A response analysis module, configured to collect touch screen operations of a user on the target point, obtain response data of the target point, and analyze the response data to obtain a response analysis result of the target point;

[0021] A point iteration module, configured to iteratively optimize the target point and the point keyword corresponding to the target point based on the response analysis result, so as to update the display page.

[0022] According to a third aspect of the present invention, there is provided a storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned page point automatic configuration method is implemented.

[0023] According to a fourth aspect of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the above-mentioned page point automatic configuration method is implemented.

[0024] A page point automatic configuration method, device, storage medium and computer device provided by the present invention are based on picture recognition and other multiple recognition methods to determine target points and summarize point keywords, realizing automatic marking of readable points in the system, avoiding meaningless long codes generated by existing automatic dotting schemes, and reducing the cost of manually predefined points; the generated point keywords can be highly integrated with various medical business scenarios such as online pharmacy sales, hospital visit services, and medical information query, providing convenient usage conditions for both doctors and patients; on the basis of automatically generating points, a keyword editing area for manual supplementation is also provided to update or supplement the point keywords to meet actual application scenarios and improve the flexibility and fault tolerance of the system; through data collection and analysis of target points, it prompts operators to make corresponding configuration optimizations to the display page, and can also intervene in the target points and point keywords according to the data analysis results to realize automatic iterative optimization of the display page. The above method realizes efficient automatic point setting, reduces labor costs, improves point setting efficiency, highly integrates points with medical business scenarios using point keywords, the editable point keywords improve the fault tolerance of the system, and operators can also optimize points and pages according to the data analysis results of point clicks, improving the applicability of the display page in the medical field.

[0025] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the specific implementation manners of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings described herein are used to provide a further understanding of the present invention and form a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1 A flowchart showing a method for automatically configuring page points provided by an embodiment of the present invention is shown;

[0028] Figure 2 A schematic diagram showing the display page effect after manually setting points is shown;

[0029] Figure 3 A schematic diagram showing the display page effect after automatically setting points by the system is shown;

[0030] Figure 4 A schematic diagram showing a display page effect in a method for automatically configuring page points provided by an embodiment of the present invention is shown;

[0031] Figure 5 A flowchart showing another method for automatically configuring page points provided by an embodiment of the present invention is shown;

[0032] Figure 6 A schematic diagram showing another display page effect in a method for automatically configuring page points provided by an embodiment of the present invention is shown;

[0033] Figure 7 A schematic diagram showing the structure of a device for automatically configuring page points provided by an embodiment of the present invention is shown;

[0034] Figure 8 A schematic diagram showing the structure of another device for automatically configuring page points provided by an embodiment of the present invention is shown;

[0035] Figure 9 A schematic diagram showing the device structure of a computer device provided by an embodiment of the present invention is shown. Detailed implementation manners

[0036] Hereinafter, the exemplary embodiments of the present application will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be completely conveyed to those skilled in the art.

[0037] An embodiment of the present application provides a method for automatically configuring page points, as Figure 1 shown, the method includes the following steps:

[0038] 101. Collect the page image of the display page, perform feature recognition on the page image, determine the target feature, and locate the target point based on the target feature.

[0039] Among them, the application background of this application is introduced. Taking the CMS system of an online pharmacy as an example, the preset methods of points in the CMS system are mainly divided into the method of manually presetting points by humans and the method of automatically presetting points by the system. Among them, as Figure 2 shown in the schematic diagram of manually presetting points in the drug page sold by the online pharmacy. After the patient clicks to enter the drug page they want to purchase, they need to further fill in personal information to make a purchase. The preset point for the drug photo information is the banner, the preset point for the name column for the patient to fill in is the username, the preset point for the age column for the patient to fill in is the age, and the preset point for the submission column for submitting the drug purchase is the confirm. By manually presetting points, the labor cost is relatively high, and the setting of all points depends on manual definition, and there are scenarios of omission; the other is that the CMS system automatically sets points. Still taking the CMS system of the online pharmacy as an example, as Figure 3 shown, the preset system point for the drug photo information is image-1-130656, the preset system point for the name column for the patient to fill in is input-1-144178, the preset system point for the age column for the patient to fill in is input-2-160232, and the preset system point for the submission column for submitting the drug purchase information is button-1-178163. It can be seen that the finally preset points have no business logic, and are completely unable to be associated with the business scenario of drug purchase. They can only be recognized in the form of module types and random numbers, which are not easy to understand, do not carry business attributes, and are easy to be confused.

[0040] Specifically, when the user accesses a certain page, the system will capture the screenshot or image data of the page, and then perform feature recognition on the page image. Specifically, it can identify specific elements in the page through image processing algorithms. Then, according to predefined rules or models, it identifies the target features that need to be marked on the display page, such as specific types of buttons, picture areas, etc., and determines which are the target features that need to be marked. Once the target features are identified, the system will calculate the specific position coordinates of these features on the page and use them as the position of the marked area. The boundary box coordinates of the identified target features can be mapped back to the actual coordinate system of the page to generate the corresponding marked area.

[0041] In the embodiments of the present application, by automatically collecting page pictures and performing feature recognition, the workload of manually setting the dotting area is reduced, and the efficiency of point setting is improved; the computer vision technology can be used to more accurately identify the target features in the page, avoiding omissions or errors that may occur during manual setting, especially in complex page structures, ensuring that each area that needs to be dotted is correctly identified and recorded.

[0042] 102. Identify and analyze the display page to obtain point keywords, and match the point keywords with the target points, where the point keywords are associated with keyword editing areas.

[0043] Specifically, in the process of generating point keywords, it is also necessary to identify and analyze the display page or the page picture of the display page. By extracting the key elements in the display page and generating corresponding keywords for the key elements, that is, point keywords, the point keywords are usually descriptions of the functions or contents of the key elements. According to the generated point keywords, the system will then search for matching items in the previously obtained target point library to determine the specific positions and uses of each point, and match the point keywords with the target points. The point keywords are used to specifically explain the content and use of the corresponding target points; and the present application also reserves a keyword editing area at the point keywords for manual supplementation, which can be manually corrected when the point keywords are incorrect or missing.

[0044] Among them, taking Figure 4 as an example, in the drug page sold in an online pharmacy, the recognition preset point for the photo information of the drug is "Product - Ganmaoling", the recognition preset point for the name column for patients to fill in is "Patient - Name", the recognition preset point for the age column for patients to fill in is "Patient - Age", the recognition preset point for the submission column for submitting information is "Purchase - Drug", and a keyword editing area is associated with each recognition preset point for manual supplementation.

[0045] In the embodiments of the present application, by automatically identifying the key elements and corresponding point keywords in the display page, omissions or errors that may occur during manual setting are reduced, ensuring that each target point is correctly identified and recorded; and each point keyword is associated with an editable area, allowing operators to adjust and supplement according to actual needs, increasing the flexibility and fault tolerance of the system, and being able to adjust in a timely manner according to the actual situation of the target points, which is more in line with the actual business logic.

[0046] 103. Collect the touch operations of the user on the target points to obtain the response data of the target points, and analyze the response data to obtain the response analysis result of the target points.

[0047] Specifically, a transparent layer covering the entire page can be created above the display page to capture the user's touch screen operations, record all the user's click operations, and not interfere with the user's normal operation behavior. After recording all the user's touch screen operations, such as click position, click count, click time, etc., the data is sent to the server for storage and analysis. By identifying the high-frequency operation areas and inferring their potential functional requirements, a response analysis result for the target point is finally generated.

[0048] In the embodiment of the present application, the touch screen operations of the user on the target point are collected, all the user's click operations are recorded, and then further analyzed based on the collected data information to identify the high-frequency operation areas and discover the hot areas that the user is concerned about. Furthermore, according to the obtained response analysis result of the target point, the page layout is adjusted in a timely manner to improve the user interaction efficiency and optimize the page design. The entire process is always data-driven for decision support, automatically completed and the analysis result is obtained, effectively reducing the workload of manual statistics and analysis, improving the operation efficiency, and then providing a personalized service experience to meet the needs of different users.

[0049] 104. Iteratively optimize the target point and the point keyword corresponding to the target point based on the response analysis result to update the display page.

[0050] In the embodiment of the present application, by analyzing the user operation data, a response analysis result is generated, the high-frequency operation areas and potential functional requirements are identified. Specifically, machine learning algorithms such as clustering analysis and heat map analysis can be used to identify the high-frequency operation areas, and the functional requirements are inferred in combination with the business logic. Further, according to the response analysis result, the existing target points and their corresponding point keywords are optimized and adjusted. For example, the position, size or function of the target point is adjusted, and its corresponding point keyword is updated to ensure that these keywords accurately reflect the user's behavior and needs to better meet the user's needs. Finally, the optimized target points and their keywords are applied to the actual display page to improve the user experience and service quality.

[0051] A method, device, storage medium, and computer device for automatically configuring page points provided by the present invention first collect a page image of a display page, perform feature recognition on the page image to determine target features, and locate target points based on the target features. Then, the display page is recognized and analyzed to obtain point keywords, and the point keywords are matched with the target points. Among them, the point keywords are associated with keyword editing areas. Then, the touch operations of the user on the target points are collected to obtain response data of the target points, and the response data is analyzed to obtain a response analysis result of the target points. Finally, the target points and the point keywords corresponding to the target points are iteratively optimized based on the response analysis result to update the display page. The above method determines target points and summarizes point keywords based on image recognition and other multiple recognition methods, realizes automatic marking of readable points by the system, avoids the generation of meaningless long codes in the existing automatic dotting scheme, and also reduces the cost of manually predefined points; the generated point keywords can be highly integrated with various medical business scenarios such as online pharmacy sales, hospital visit services, and medical information query, providing convenient usage conditions for both doctors and patients; on the basis of automatically generating points, a keyword editing area for manual supplementation is also set to update or supplement the point keywords to meet the actual application scenarios and improve the flexibility and fault tolerance of the system; through the data collection and analysis of the target points, it prompts the operator to make corresponding configuration optimizations to the CMS page, and can also intervene in the target points and point keywords according to the data analysis results to realize automatic iterative optimization of the display page. The above method realizes efficient automatic point setting, reduces labor costs, improves point setting efficiency, highly integrates points with medical business scenarios using point keywords, the editable point keywords improve the fault tolerance of the system, and the operator can also optimize the points and the page according to the data analysis results of point clicks, improving the applicability of the display page in the medical field.

[0052] Another method for automatically configuring page points is provided in an embodiment of the present application. As Figure 5 shown, the method includes the following steps:

[0053] 201. Perform feature recognition on the page image in the display page to locate the target points.

[0054] Specifically, preprocess the page image, perform feature recognition on the preprocessed page image, and extract target features. Among them, the target features include text features, shape features, and color features; generate a feature data set according to the target features, input the feature data set into a preset point recognition model to obtain point area data; and locate the target points based on the point area data.

[0055] In the embodiments of the present application, in order to prepare high-quality input pictures for subsequent feature extraction and model training, it is first necessary to preprocess the page pictures, adjust all input pictures to the same size and resolution to ensure consistency, and enhance the contrast of the pictures through image enhancement algorithms such as histogram equalization to make the text and features more obvious. Finally, remove the noise in the pictures to improve the accuracy of feature extraction. After the picture preprocessing, it is necessary to extract the target features to extract features from the pictures that are helpful for identifying operation entrances, such as button, link and other features. The target features specifically include three categories: text features, shape features and color features. Among them, specific text has a higher weight: for example, common operation vocabulary such as "confirm", "cancel", "I got it", "submit" when purchasing drugs in offline pharmacies and registering in the hospital registration system has a higher weight. Then use edge detection and connected component analysis methods to frame the text area, and then use a recurrent neural network to recognize the shape and structure of the text and convert it into readable text; specific shape features have a higher weight, such as common operation entrance shapes such as square buttons and round buttons. Specifically, edge detection methods can be used to identify shape features; similarly, specific color features have a higher weight, such as the company's commonly used button theme color. In the hospital system, for example, cool colors such as blue and green are used more frequently, while in offline stores, warm colors such as orange and yellow are used more frequently. Furthermore, the color distribution characteristics of the pictures can be captured through color histograms, and finally texture features can be extracted using filters.

[0056] Further, after extracting the target features, convert the target features into a feature data set that can be used for training the model for model training. Specifically, use a support vector machine to perform basic classification on different operation entrances, such as buttons, links, etc., and then generate keywords or labels through a clustering algorithm based on the picture feature set and classification information. Finally, use the existing page marking data to test the model to verify the accuracy and effectiveness of the model. After ensuring the effectiveness of the model, display the results recognized by the model. The model outputs a specific position area data and prompts the user of the area that needs to be marked.

[0057] Using the technical means provided by the present application, the system can efficiently and automatically identify and set the marking positions, reduce the cost of manual intervention, and improve the accuracy and readability of the marking, ensuring the effectiveness and flexibility of the point confirmation.

[0058] 202. Use the image recognition method to recognize and analyze the display page to determine the point keywords.

[0059] Specifically, obtain the page image of the display page, preprocess the page image, and perform text extraction on the preprocessed page image to obtain the first keyword segment; perform feature recognition on the preprocessed page image and extract effective features, where the effective features include color features, texture features, and shape features; generate a feature training set based on the effective features, use the feature training set to train a preset keyword generation model, and obtain the second keyword segment output by the keyword generation model; based on the first weight coefficient corresponding to the first keyword segment and the second weight coefficient corresponding to the second keyword, splice the first keyword segment and the second keyword segment to obtain the location keyword.

[0060] In one implementation, the image recognition method can be used to identify and analyze to determine the location keyword. The first step is also to preprocess the page image to ensure that all input images have a consistent format and enhanced quality, preparing for subsequent processing steps. The specific process of image preprocessing can refer to the process of image preprocessing in step 201 and will not be elaborated here; then text extraction is required. First, text detection is performed. The text areas in the picture are framed through methods such as edge detection and connected region analysis. Then, a recurrent neural network is used to recognize the shape and structure of the text and convert it into readable text, and the model is corrected according to the feedback information updated manually to improve the recognition accuracy, obtaining the first keyword segment; on the basis of text extraction, further extract features such as color, texture, and shape from the picture. The color feature is extracted through a color histogram, the texture feature is extracted through a filter, and the shape feature is extracted through an edge detection method. Then, a feature training set is generated by calculating the gray-level co-occurrence matrix for subsequent model training. Among them, the training process first realizes classification training, and basic classifications such as commodities, drugs, promotional pictures, medical services, and electronic cards are realized through a support vector machine. Then, based on the picture feature set and classification information, a second keyword segment is generated through a clustering algorithm; then keyword weight division is performed. The weight of the first keyword segment extracted from the text accounts for 70%, and the weight of the second keyword segment summarized from the features accounts for 30%. The first weight coefficient is much larger than the second weight coefficient to ensure that the keywords used in most medical application scenarios are more inclined to commodity names, drug names, promotional slogans of promotional pictures, etc., and finally complete the entire process of generating the location keyword.

[0061] 203. Determine the location keyword by analyzing the code of the display page.

[0062] Specifically, determine the page components in the display page, obtain the source code corresponding to the page components, and parse the source code; traverse the abstract syntax tree in the parsed source code, extract the target tags and the annotation content corresponding to the target tags; use the preset natural language toolkit to perform part-of-speech tagging on the annotation content to generate multiple original keywords; perform splicing processing on the multiple original keywords based on the preset symbols to obtain the dot position keywords.

[0063] In the embodiment of the present application, in the display page of the CMS system, the modules are dragged in according to the page components. Taking the hospital system as an example, it includes components such as "appointment registration", "doctor introduction", and "drug information". When an operator drags in a page component, the system will automatically load the corresponding source code of the page and perform subsequent processing. By parsing the source code, traversing the parsed abstract syntax tree, extracting relevant target tags, and corresponding function names, variable names, etc., especially the annotation-related content. For example, in the source code of the "appointment registration" module, there may be annotations on how to obtain the department list, doctor information, etc. By parsing the annotation content, the function and purpose of the module can be understood more accurately, thereby generating more representative dot position keywords, which helps the operator better understand and maintain the code; after parsing the annotation content, use the natural language toolkit to perform part-of-speech tagging and extract representative nouns as keywords, such as "appointment confirmation", "online consultation", "drug instruction manual", etc., avoiding the interference of irrelevant words and improving the quality of the keywords; finally, through the multiple keywords that may be obtained in the previous steps, splice them with a specific delimiter as the dot position keywords. For example, in the "appointment registration" page component, the extracted keywords may be "appointment", "confirmation", "doctor selection", etc. Through specific delimiters, such as underscores or hyphens, these keywords are spliced into a complete dot keyword according to certain rules, such as "appointment_confirmation", "doctor_selection", etc., which is convenient for subsequent data collection and analysis.

[0064] 204. Collect the dot operation of the user, obtain the response data of the target dot position, and analyze the response data to obtain the response analysis result of the target dot position.

[0065] In the embodiment of the present application, it is necessary to collect the touch screen operation of the user on the target dot position, record all the click operation data of the user and further analyze it to identify the high-frequency operation area, and then adjust the page layout in a timely manner according to the obtained response analysis result of the target dot position to improve the user interaction efficiency and optimize the page design; the whole process is completed automatically and the analysis result is obtained, effectively reducing the workload of manual statistics and analysis and improving the operation efficiency. For the specific operation process, refer to step 103, and details will not be elaborated here.

[0066] 205. Add target dot positions to the non-target dot position areas with high-frequency dotting.

[0067] Specifically, obtain the response frequency data of the target point in the response analysis result, as well as the response frequency data of the non-target point area in the display page; when there is non-point area response frequency data greater than or equal to the response frequency data of the target point, mark the non-point area and generate a point prompt message, where the prompt message is used to indicate adding the target point to the marked non-point area.

[0068] In the embodiment of the present application, taking Figure 6 as an example, when there is no return point configured in the upper left corner of the display page, most users will frequently click on the upper left corner based on usage habits, but the page will not have a corresponding response feedback. The response analysis result shows that the non-point area in the upper left corner receives a high frequency of clicks, which will then prompt the operator to add a corresponding return button.

[0069] 206. Iteratively optimize the target point and the point keyword corresponding to the target point based on the response analysis result.

[0070] Specifically, obtain the response analysis result based on a preset time interval, where the response analysis result includes multiple time dimensions and the point click data corresponding to the time dimensions; generate a click data trend chart according to the multiple time dimensions and the point click data corresponding to the time dimensions, and compare the point click data with the preset click data; when there is point click data lower than the preset click data, mark the target point corresponding to the point click data as a prompt point and generate an operation prompt message, where the operation prompt message is used to indicate updating or taking off the shelf of the prompt point.

[0071] In the embodiment of the present application, obtaining the response analysis result including multiple time dimensions and the point click data corresponding to the time dimensions based on a preset time interval can be specifically recorded through a daily scheduled task, and data analysis is performed in weekly, monthly, and annual dimensions. For example, in the promotional activities of a shopping festival, during the promotion period in the first half of the month when the shopping festival is held, the access data volume of the point is high, and the data volume is low after the event, which prompts corresponding updates or taking off the shelf of relevant points; while the access data volume of points for cold and wind-cold medicines during the spring and autumn seasons is high, and the access data volume of points for heatstroke medicines in summer is high, etc., to achieve intelligent prompts in different seasons, taking off the shelf or updating the target points with low point click data according to different seasons to ensure that the display page always maintains accurate display content, is closer to the usage habits and needs of users, and is convenient for users to use efficiently.

[0072] In another implementation manner, still taking Figure 6For example, when a user needs to manually fill in their home address on a page, it usually takes a lot of time. At this time, the response analysis result of the target point will feedback the situation, identifying that the user spends a long time. To reduce the time spent by the user and improve the system response efficiency, the operator can be informed to replace the home address input field with a selection box for province, city, and district, thereby improving the user's filling efficiency.

[0073] 207. Optimize the size adaptation of the target point based on the display device of the display page.

[0074] Specifically, obtain the page size of the display page and the point size of the target point; determine the display device of the display page and obtain the screen size of the display device; calculate the proportionality coefficient between the screen size and the page size, and adjust the page size to the screen size to make the display page match the display device; adjust the point size based on the proportionality coefficient to make the target point match the display page.

[0075] In the embodiments of the present application, among the screen sizes of different display devices, the position of the point will continuously change dynamically. Specifically, the page size information of the display page can be collected, and the proportional relationship between the page size and the screen size of the display device can be obtained. The point area can be dynamically adjusted according to the proportional relationship, that is, the width and height conversion of the dotting area can be realized by using a fixed ratio to achieve the intelligent optimization of adjusting the point area according to the display device size.

[0076] Another method for automatic page point configuration provided by the present invention is as follows: First, perform feature recognition on the page image in the display page to locate the target point, then use image recognition or code analysis to identify and analyze the display page to determine the point keyword. After that, collect the user's dotting operation, obtain the response data of the target point, and analyze the response data to obtain the response analysis result of the target point. Add target points to the non-target point areas with high-frequency dotting, and then iteratively optimize the target point and the corresponding point keyword based on the response analysis result. Finally, optimize the size adaptation of the target point based on the display device of the display page.

[0077] Further, as Figure 1 a specific implementation of the method, the embodiments of the present application provide a device for automatic page point configuration, as Figure 7 shown. The device includes: a point positioning module 301, a keyword generation module 302, a response analysis module 303, and a point iteration module 304.

[0078] The point positioning module 301 is used to collect the page image of the display page, perform feature recognition on the page image to determine the target feature, and locate the target point based on the target feature;

[0079] The keyword generation module 302 is used to identify and analyze the display page to obtain point keywords, and match the point keywords with the target points. Among them, the point keywords are associated with keyword editing areas;

[0080] The response analysis module 303 is used to collect the touch screen operations of the user for the target point to obtain the response data of the target point, and analyze the response data to obtain the response analysis result of the target point;

[0081] The point iteration module 304 is used to iteratively optimize the target point and the point keywords corresponding to the target point based on the response analysis result to update the display page.

[0082] In a specific application scenario, the point positioning module 301 can specifically be used to preprocess the page image, perform feature recognition on the preprocessed page image, and extract target features. Among them, the target features include text features, shape features, and color features; generate a feature data set according to the target features, input the feature data set into a preset point recognition model to obtain point area data; and locate the target point based on the point area data.

[0083] In a specific application scenario, the keyword generation module 302 can specifically be used to obtain the page image of the display page, preprocess the page image, and perform text extraction on the preprocessed page image to obtain the first keyword segment; perform feature recognition on the preprocessed page image and extract effective features. Among them, the effective features include color features, texture features, and shape features; generate a feature training set based on the effective features, use the feature training set to train a preset keyword generation model, and obtain the second keyword segment output by the keyword generation model; and splice the first keyword segment and the second keyword segment based on the first weight coefficient corresponding to the first keyword segment and the second weight coefficient corresponding to the second keyword to obtain the point keyword.

[0084] In a specific application scenario, the keyword generation module 302 can specifically be used to determine the page components in the display page, obtain the source code corresponding to the page components, and parse the source code; traverse the abstract syntax book in the parsed source code, extract the target tags and the annotation content corresponding to the target tags; perform magnetic annotation on the annotation content using a preset natural language toolkit to generate multiple original keywords; and splice and process the multiple original keywords based on a preset symbol to obtain the point keyword.

[0085] In a specific application scenario, the response data includes response frequency data; the above device further includes a point adding module 305, which is specifically configured to obtain the response frequency data of the target point in the response analysis result, as well as the response frequency data of the non-target point area in the display page; when there is a response frequency data of a non-point area greater than or equal to the response frequency data of the target point, mark the non-point area and generate a point prompt message, where the prompt message is used to indicate adding a target point to the marked non-point area.

[0086] In a specific application scenario, the point iteration module 304 is specifically configured to obtain a response analysis result based on a preset time interval, where the response analysis result includes multiple time dimensions and the point click data corresponding to the time dimensions; generate a click data trend chart according to the multiple time dimensions and the point click data corresponding to the time dimensions, and compare the point click data with preset click data; when there is point click data lower than the preset click data, mark the target point corresponding to the point click data as a prompt point and generate an operation prompt message, where the operation prompt message is used to indicate updating or taking off the shelf of the prompt point.

[0087] In a specific application scenario, the above device further includes a point adaptation adjustment module 306, which is specifically configured to obtain the page size of the display page and the point size of the target point; determine the display device of the display page and obtain the screen size of the display device; calculate the proportional coefficient between the screen size and the page size, and adjust the page size to the screen size to make the display page match the display device; adjust the point size based on the proportional coefficient to make the target point match the display page.

[0088] It should be noted that for other corresponding descriptions of each functional unit involved in the page point automatic configuration device provided in this embodiment, reference can be made to Figure 1 and Figure 5 the corresponding descriptions therein, which will not be elaborated here.

[0089] Based on the above method as Figure 1 shown, correspondingly, this embodiment further provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, the above page point automatic configuration method is implemented.

[0090] Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product. The software product to be recognized can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the page point automatic configuration method in each implementation scenario of the present application.

[0091] Based on the above as Figure 1and Figure 5 the method shown, and Figure 7 and Figure 8 the embodiment of the page point position automatic configuration device shown. To achieve the above object, as Figure 9 shown, this embodiment also provides an entity device for automatic page point position configuration. The device includes a communication bus, a processor, a memory, and a communication interface, and may further include an input / output interface and a display device. Among them, each functional unit can complete mutual communication through the bus. The memory stores a computer program, and the processor is used to execute the program stored on the memory to execute the page point position automatic configuration method in the above embodiment.

[0092] Optionally, the entity device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, and so on. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), etc. Optionally, the user interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface), etc.

[0093] Those skilled in the art can understand that the structure of an entity device for automatic page point position configuration provided in this embodiment does not constitute a limitation on the entity device, and may include more or fewer components, or combine some components, or arrange different components.

[0094] The storage medium may further include an operating system and a network communication module. The operating system is a program for managing the hardware and software resources to be recognized of the above entity device, and supports the operation of the information processing program and other software and / or programs to be recognized. The network communication module is used to implement communication between components inside the storage medium, and communication between other hardware and software in the information processing entity device.

[0095] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform, or can also be implemented by hardware. By applying the technical solution of the present application, first, a page image of the display page is collected, the page image is subjected to feature recognition to determine the target feature, and the target point is located based on the target feature. Then, the display page is recognized and analyzed to obtain the point keyword, and the point keyword is matched with the target point. Among them, the point keyword is associated with the keyword editing area. Then, the touch operation of the user on the target point is collected to obtain the response data of the target point, and the response data is analyzed to obtain the response analysis result of the target point. Finally, based on the response analysis result, the target point and the point keyword corresponding to the target point are iteratively optimized to update the display page. The above method is based on image recognition and other multiple recognition methods to determine the target point and summarize the point keyword, realizing the automatic marking of readable points by the system, avoiding the generation of meaningless long codes in the existing automatic dotting scheme, and also reducing the cost of manually predefined points; the generated point keywords can be highly integrated with various medical business scenarios such as online pharmacy sales, hospital visit services, and medical information query, providing convenient usage conditions for both doctors and patients; on the basis of automatically generating points, a keyword editing area for manual supplementation is also set to update or supplement the point keyword to meet the actual application scenario and improve the flexibility and fault tolerance of the system; through the data collection and analysis of the target point, it prompts the operator to make corresponding configuration optimizations to the CMS page, and can also intervene in the target point and the point keyword according to the data analysis result to realize the automatic iterative optimization of the display page. The above method realizes efficient automatic point setting, reduces labor costs, improves the efficiency of point setting, highly integrates the points with medical business scenarios by using point keywords, the point keywords are editable to improve the fault tolerance of the system, and the operator can also optimize the points and the page according to the data analysis result of the point click, improving the applicability of the display page in the medical field.

[0096] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred embodiment scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the device in the embodiment scenario can be distributed in the device in the embodiment scenario according to the description of the embodiment scenario, or can be correspondingly changed and located in one or more devices different from this embodiment scenario. The modules in the above embodiment scenario can be combined into one module, or can be further split into multiple sub-modules.

[0097] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the embodiment scenarios. The above disclosure is only several specific embodiment scenarios of the present application. However, the present application is not limited thereto, and any change that can be thought of by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A method for automatically configuring page points, characterized in that: The method comprises: Collecting page images of displayed pages, performing feature recognition on the page images, determining target features, and locating target points based on the target features; Identify and analyze the display page to obtain a point keyword, and match the point keyword with the target point, wherein the point keyword is associated with a keyword editing area; Collecting the user's touch screen operation on the target point to obtain response data of the target point, and analyzing the response data to obtain a response analysis result of the target point; The target point and the point keyword corresponding to the target point are iteratively optimized based on the response analysis result to update the display page.

2. The method according to claim 1, characterized in that The performing feature recognition on the page image, determining the target feature, and locating the target point based on the target feature includes: Preprocessing the page image, and performing feature recognition on the preprocessed page image to extract target features, wherein the target features include text features, shape features, and color features; Generate a feature data set according to the target feature, input the feature data set into a preset point recognition model to obtain point area data; The target point is located based on the point area data.

3. The method according to claim 1, characterized in that The identifying and analyzing the display page to obtain the point keywords includes: Acquire a page image of the display page, preprocess the page image, and extract text from the preprocessed page image to obtain a first keyword segment; Performing feature recognition on the preprocessed page image and extracting effective features, wherein the effective features include color features, texture features and shape features; Generate a feature training set based on the effective features, use the feature training set to train a preset keyword generation model, and obtain a second keyword segment output by the keyword generation model; Based on a first weight coefficient corresponding to the first keyword segment and a second weight coefficient corresponding to the second keyword, the first keyword segment and the second keyword segment are concatenated to obtain the point keyword.

4. The method according to claim 1, characterized in that: The identifying and analyzing the display page to obtain the point keywords includes: Determine a page component in the display page, obtain source code corresponding to the page component, and parse the source code; Traversing the abstract grammar book in the parsed source code, extracting the target tag and the comment content corresponding to the target tag; Using a preset natural language toolkit to magnetically annotate the annotation content to generate a plurality of original keywords; The multiple original keywords are concatenated based on preset symbols to obtain the point keyword.

5. The method according to claim 1, characterized in that The response data includes response frequency data; after analyzing the response data to obtain the response analysis result of the target point, the method further includes: Acquire the response frequency data of the target point in the response analysis result, and the response frequency data of the non-target point area in the display page; When the response frequency data of the non-point area is greater than or equal to the response frequency data of the target point, the non-point area is marked and point prompt information is generated, wherein the prompt information is used to indicate adding a target point to the marked non-point area.

6. The method according to claim 1, characterized in that The iterative optimization of the target point and the point keyword corresponding to the target point based on the response analysis result includes: Acquire the response analysis result based on a preset time interval, wherein the response analysis result includes multiple time dimensions and point click data corresponding to the time dimensions; Generate a click data trend chart according to the multiple time dimensions and the point click data corresponding to the time dimensions, and compare the point click data with the preset click data; When the point click data is lower than the preset click data, the target point corresponding to the point click data is marked as a prompt point, and operation prompt information is generated, wherein the operation prompt information is used to instruct the prompt point to be updated or removed from the shelves.

7. The method according to claim 1, characterized in that The iterative optimization of the target point and the point keyword corresponding to the target point based on the response analysis result includes: Acquire the page size of the display page and the point size of the target point; Determine a display device for displaying the page, and obtain a screen size of the display device; Calculating a ratio coefficient between the screen size and the page size, and adjusting the page size to the screen size so that the display page matches the display device; The point size is adjusted based on the scale factor so that the target point matches the display page.

8. A device for automatically configuring page points, characterized in that: The device comprises: A point positioning module is used to collect page images of displayed pages, perform feature recognition on the page images, determine target features, and locate target points based on the target features; A keyword generation module, used for identifying and analyzing the display page to obtain a point keyword, and matching the point keyword with the target point, wherein the point keyword is associated with a keyword editing area; A response analysis module, used to collect the user's touch screen operation on the target point, obtain the response data of the target point, and analyze the response data to obtain the response analysis result of the target point; A point iteration module is used to iteratively optimize the target point and the point keyword corresponding to the target point based on the response analysis result to update the display page.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images