Information Push Methods and Systems Based on Smart Healthcare Big Data

Abstract

This invention relates to an information push method and system based on smart healthcare big data. The method includes: calculating the activation rate of single-category projects for each user category and project category combination; calculating a comprehensive project activation rate by using a weighted average method for multiple single-category project activation rates; setting a project activity factor based on the value of the single-category project activation rate; setting a user activity factor based on the number of project types involved in a single user; and generating an optimization scheme based on the single-category project activation rate, combined with the project activity factor and the user activity factor. By calculating the single-category project activation rate, it is possible to more accurately match users with smart healthcare projects, taking into account user interests and needs, combining project characteristics and target user groups, and better meeting the characteristics and preferences of different user groups, thereby improving the accuracy of push notifications and user experience.

Description

Technical Field

[0001] This invention relates to the field of smart medical big data technology, specifically to an information push method and system based on smart medical big data. Background Technology

[0002] With the continuous development of computer science and information technology, big data has been widely applied across various industries. For example, in the smart healthcare industry, big data-based applications are emerging rapidly. In a typical smart healthcare application scenario, information push based on smart healthcare big data has been widely used. However, most common information push methods rely on predetermined rules to send information to the recipient, resulting in poor push effectiveness and user experience.

[0003] Chinese invention patent application number CN202111349320.4 provides an information push method and a smart healthcare cloud server based on smart healthcare big data. The method uses the session feature data of smart healthcare projects without preset session tags corresponding to the target push service scope and the target smart healthcare session, as well as the push service information statistics results corresponding to the target push service scope, to obtain an optimized push strategy scheme for the target smart healthcare session based on the information push strategy optimization network. Then, based on the push service information statistics results, information is pushed to the target smart healthcare session through the optimized information push strategy.

[0004] Existing patented technologies use a session item feature dataset without pre-defined session tags to analyze the item activation frequency and obtain an optimized push strategy for the target smart healthcare session based on the item activation frequency analysis data. However, the item activation frequency used is an analysis and optimization of all items and users, without calculating the item activation frequency according to different user and item categories. This results in inaccurate and untargeted item activation frequency calculation, which in turn reduces the accuracy of information push. Summary of the Invention

[0005] This invention addresses the technical problems existing in the prior art by providing an information push method and system based on smart medical big data.

[0006] The technical solution of this invention to solve the above-mentioned technical problems is as follows: an information push method based on smart medical big data, the method comprising:

[0007] S100: Obtain smart healthcare project information and corresponding session feature data, and at the same time obtain push business information statistics results, and classify users and projects according to smart healthcare project information and corresponding session feature data;

[0008] S200: For each user category and item category combination, calculate the single category item activation rate, and use a weighted average method to calculate the comprehensive item activation rate for multiple single category item activation rates;

[0009] S300: Set the project activity factor based on the activation rate of the single category project; set the user activity factor based on the number of project types involved by a single user;

[0010] S400: Based on the activation rate of the single-category item, combined with the item activity factor and the user activity factor, generate an optimization plan; optimize the information push strategy according to the optimization plan, push information according to the optimized information push strategy, and monitor and obtain push feedback data in real time.

[0011] S500: Evaluate the push optimization effect based on the activation rate of the comprehensive project, and generate a user trend change report for data analysis to adjust the optimization strategy in real time.

[0012] Furthermore, the optimization scheme categorizes users into high-activity, medium-activity, and low-activity users based on the single-category item activation rate, item activity factor, and user activity factor; it also divides items into high, medium, and low priorities; the optimization scheme sets a basic push strategy and a cross-pre-test push strategy. The basic push strategy pushes information at different frequencies for different active users and different priority items; the cross-pre-test push strategy pushes high-priority items to low-activity users and low-priority items to high-activity users, sets a time limit to collect user feedback data, analyzes the feedback results collected by the cross-pre-test push strategy, and combines it with the basic push strategy to determine the push frequency and push scope, generating the final information push optimization strategy.

[0013] Furthermore, the optimization scheme also includes a related project recommendation rule. The related project recommendation rule selects complementary projects that are strongly related to the project with the highest current attention from the user based on the project association matrix, and pushes information based on the complementary projects to guide and optimize the user's access path.

[0014] Furthermore, the project association matrix is ​​constructed by analyzing the degree of association between smart healthcare projects using an association rule learning algorithm, and the calculated degree of association is used to construct the project association matrix; the strongly associated complementary projects refer to projects with an association degree greater than the association threshold.

[0015] Furthermore, the user access path optimization refers to identifying key nodes in the user access path and inserting complementary items at the key nodes to achieve a natural transition; the key nodes refer to the project pages that the user visits and stays on, and complementary items are inserted during the time the user stays on the project page.

[0016] Furthermore, the push feedback data is obtained, the user push service results are integrated, the matching value of each push item received by the user is calculated, the matching value of all push items for each user is counted and the median is calculated, and the initial item matching value range is set with the median as the origin; the initial item matching value range is debugged according to the debugging rules to obtain the accurate item matching value range.

[0017] Furthermore, the debugging rule involves performing pre- and post-tests on the lower limit of the initial project matching value range until the maximum number of pre-tests is reached, observing user feedback and push effects, and adjusting the project matching value range based on the pre-test results; the maximum number of pre-tests is preset.

[0018] Furthermore, the precise project matching value range is obtained, and the project matching value range is dynamically adjusted in combination with a random exploration mechanism to form a fluctuating project matching value range. Information push is optimized based on the fluctuating project matching value range.

[0019] Furthermore, the random exploration mechanism randomly selects a predetermined number of untouched items to fill the precise item matching value range, and sets a time interval to periodically change the items and remove invalid items from the precise item matching value range; the predetermined number is determined by the number of items in the precise item matching value range and the random perturbation factor to increase the number of untouched items.

[0020] The information push system based on smart healthcare big data, along with the information push method based on smart healthcare big data as described above, includes: a data acquisition module for acquiring smart healthcare project information, session feature data, and statistical results of push business information;

[0021] The classification calculation module receives data from the data acquisition module, classifies users and items, and calculates the activation rate of single-category items.

[0022] The activity factor setting module is used to receive the activation rate of a single category item from the category calculation module and set the item activity factor and user activity factor.

[0023] The optimization scheme module is used to receive the activation rate of single-category projects, project activity factors, and user activity factors to generate optimization schemes.

[0024] The feedback adjustment module is used to monitor and obtain push feedback data in real time, and adjust and optimize strategies in real time.

[0025] The beneficial effects of this invention are as follows: By classifying users and projects, and calculating the activation rate of single-category projects for each user category and project category combination, it is possible to more accurately match users with smart healthcare projects. This takes into account users' interests and needs, combines project characteristics and target user groups, and is more in line with the characteristics and preferences of different user groups, thereby improving the accuracy of push notifications and user experience. By setting user activity factors and project activity factors, it is possible to identify and actively push information to users who may be interested in the project but whose current activity level is low, thereby improving overall user activity. Attached Figure Description

[0026] Figure 1 This is a flowchart of the information push method based on smart medical big data of the present invention;

[0027] Figure 2 This is an architecture diagram of the information push system based on smart medical big data of the present invention. Detailed Implementation

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0030] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0031] Example 1: As Figure 1 As shown, this application provides an information push method based on smart healthcare big data, the method comprising:

[0032] S100: Obtain smart healthcare project information and corresponding session feature data, and at the same time obtain push business information statistics results, and classify users and projects according to smart healthcare project information and corresponding session feature data;

[0033] Specifically, the session feature data includes consultation content, consultation items, consultation type, and number of sessions; the smart healthcare project information includes the smart healthcare project name, sub-project name, project description, target user group, project type, project service content, and project feedback; the project type represents whether the project is a general project or a special project; the push service information statistics include: push project statistics, the name of the pushed project and the number of times it was pushed; user feedback statistics, the click-through rate, conversion rate, and satisfaction rate of users with the push information; push time statistics, the sending time and frequency of the push information; user behavior score, the user's behavior after receiving the push information, such as whether to view it or consult it; and push effect analysis, which is a comprehensive analysis of the above content to obtain the push effect, expressed using a specific score or percentage, without any specific limitation.

[0034] Specifically, the user classification is based on users' historical conversation feature data. Key features, such as users' interests, health needs, and health status, are extracted from the conversation data. Clustering algorithms are used to group users, classifying them based on factors such as users' interests, needs, and health status, such as "cardiovascular disease patients," "diabetic patients," and "health consultants." The project classification is divided into two main categories: general projects and specialized projects. General projects are applicable to most users, such as health consultations and routine physical examinations. Specialized projects target specific user groups or specific health issues, such as cardiovascular disease treatment and diabetes management. Each project is analyzed and classified based on its target user group, project type, and project service content.

[0035] S200: For each user category and item category combination, calculate the single category item activation rate, and use a weighted average method to calculate the comprehensive item activation rate for multiple single category item activation rates;

[0036] Specifically, each user category is combined with each project category to form a user-project category combination. For each user-project category combination, the number of sessions a user has under that project category is counted, and the single-category project activation rate is calculated. The single-category project activation rate = the number of sessions a user has under that project category / the total number of sessions a user has. The weighted average method is used to calculate the comprehensive project activation rate by combining multiple single-category project activation rates. By combining user categories and project categories, different weights are set for different category combinations, and the weighted average method is used to calculate the comprehensive project activation rate. The comprehensive project activation rate can clarify the trend of project opening changes, and the single-category project activation rate can accurately match users and projects, improving the push effect.

[0037] S300: Set the project activity factor based on the activation rate of the single category project; set the user activity factor based on the number of project types involved by a single user;

[0038] Specifically, the steps for setting the project activity factor based on the activation rate of the single-category project are as follows: S301: Determine a series of single-category project activation rate threshold ranges based on business needs or historical data, and set different levels according to the threshold ranges; S302: Compare the activation rate of each single-category project with the set thresholds to determine the level to which each single-category project activation rate belongs; S303: Pre-assign a project activity factor to each different level, and match the corresponding project activity factor according to the actual level. For example, the thresholds are set to three levels: low (0%-20%), medium (21%-50%), and high (51%-100%). The project activity factor for the low level is 1, the project activity factor for the medium level is 2, and the project activity factor for the high level is 3. For example, there are the following three single-category project activation rate data: Activation rate of project A: 45%; Activation rate of project B: 75%; Activation rate of project C: 15%. Based on step 1, we set the following thresholds: Low threshold: 0%-20%; Medium threshold: 21%-50%; High threshold: 51%-100%. Next, we categorized the project activation rates according to step 2: Project A's activation rate is in the medium range (21%-50%); Project B's activation rate is in the high range (51%-100%); Project C's activation rate is in the low range (0%-20%). Finally, according to step 3, we set a project activity factor for each project: Project A's project activity factor: 2; Project B's project activity factor: 3; Project C's project activity factor: 1.

[0039] Specifically, the user activity factor is set based on the number of project types a single user is involved in, reflecting the user's enthusiasm for participating in smart healthcare projects. For each user, the number of different project types they are involved in is counted, and a user activity factor is assigned to the user based on the number of project types involved. For example, a baseline value (such as 1) is set, and then for each new project type added, the user activity factor increases by a certain value (such as 0.5). A high project activity factor indicates that the user has a high level of interest in that project, and a high user activity factor indicates that the user is highly enthusiastic about participating in smart healthcare projects.

[0040] S400: Based on the activation rate of the single-category item, combined with the item activity factor and the user activity factor, generate an optimization plan; optimize the information push strategy according to the optimization plan, push information according to the optimized information push strategy, and monitor and obtain push feedback data in real time.

[0041] Specifically, the optimization scheme categorizes users into high-activity, medium-activity, and low-activity users based on the single-category project activation rate, project activity factor, and user activity factor; it also categorizes projects into high, medium, and low priorities; the optimization scheme sets up a basic push strategy and a cross-pre-test push strategy. The basic push strategy pushes information at different frequencies for different active users and different priority projects. The cross-pre-test push strategy pushes high-priority projects to low-activity users and low-priority projects to high-activity users. A time limit is set to collect user feedback data. The feedback results collected by the cross-pre-test push strategy are analyzed, and the push frequency and scope are determined in conjunction with the basic push strategy to generate the final information push optimization strategy.

[0042] Specifically, based on user activity factors, users are segmented into highly active, moderately active, and inactive users. Combining project activity factors and single-category project activation rates, projects are categorized into high, medium, and low priorities. The basic push strategy involves high-frequency, personalized information pushes for high-priority projects and highly active users; a moderate information push strategy for medium-priority projects and moderately active users; and low-frequency but precise information pushes for low-priority projects and inactive users. Push frequency refers to the frequency of push notifications; the high and low frequencies are set based on actual conditions and adjusted according to user habits.

[0043] The cross-testing push strategy involves conducting cross-tests based on priority and activity levels, analyzing the results to formulate a more precise optimized push strategy. Specifically, high-priority projects are pushed to some inactive users, and their feedback is collected to understand their reactions to these projects, further analyze the reasons for their inactivity, and identify potential problems with the projects. Simultaneously, low-priority projects are pushed to some highly active users, and their feedback is collected to identify potential shortcomings and areas for improvement in these projects. The timeframe for the cross-testing push can be set according to actual conditions; preferably, the timeframe should be no less than one week.

[0044] Analyze feedback data collected from cross-test push notifications to identify specific reasons for user disinterest or low activity. Adjust project content, presentation, or push strategies based on feedback to improve project appeal and user engagement. Regularly review the entire push strategy, including the effectiveness of basic push notifications and cross-test push notifications, and iterate and optimize as necessary. Continuously update user segmentation and project categorization methods to adapt to changing market demands and user preferences.

[0045] Specifically, for example, there are two smart healthcare projects, A and B, and two user groups, C and D. Data calculations yield the following results: Project A has a high single-category project activation frequency and a high project activity factor, indicating that the project is popular and has high user interest. Project B has a medium single-category project activation frequency and a low project activity factor, indicating that the project is used less frequently and has low user interest.

[0046] User group C has a higher user activity factor, indicating that this group actively participates in smart healthcare projects. User group D has a lower user activity factor, indicating that this group is not very enthusiastic about participating in smart healthcare projects.

[0047] Based on the above data, the following push strategy can be formulated: For project A, push it to both user group C and user group D at the same time, because the project is popular and user group C has high activity, so the push effect is expected to be better. Push it to user group D at the same time to form a comparison group with user group C. Analyze the feedback of subsequent push results to further determine the reasons for the low activity factor of user group D.

[0048] For Project B, push it to user group C. Although the project itself is not of high interest, the activity level of user group C may stimulate their interest in the project. At the same time, appropriately push it to highly active users in user group D to expand the project's influence.

[0049] Set appropriate push frequencies and times to ensure that information reaches users at the right time. The forms of push information can be diversified, including SMS, email, and APP notifications. The specific form should be selected based on user needs and platform characteristics.

[0050] S500: Evaluate the push optimization effect based on the activation rate of the comprehensive project, and generate a user trend change report for data analysis to adjust the optimization strategy in real time;

[0051] Specifically, the effectiveness of push notification optimization is evaluated by comparing the changes in the overall project activation rate before and after optimization. A user trend change report is generated based on these changes, and analysis of this report reveals current user trends, which plays a crucial role in the addition and reduction of subsequent smart healthcare projects. Simultaneously, user feedback data after push notifications is collected, including metrics such as click-through rate, conversion rate, and user satisfaction, to further analyze the push notification effect and evaluate the effectiveness of the optimization strategy. Based on the evaluation results, the optimization strategy is adjusted, and the information push effect is continuously iterated and improved.

[0052] In this application, preferably, the smart healthcare project can be a single project or a comprehensive project containing multiple sub-projects. If it contains multiple sub-projects, the categories of the sub-projects must be consistent.

[0053] This embodiment categorizes users and projects, and calculates the activation rate of single-category projects for each user and project category combination. This allows for more accurate matching of users with smart healthcare projects, taking into account user interests and needs, and combining project characteristics with target user groups. This approach better aligns with the characteristics and preferences of different user groups, thereby improving the accuracy of push notifications and user experience. Evaluating the push optimization effect through the comprehensive project activation rate helps understand the effectiveness of the push strategy and provides data support for subsequent project development. By setting user activity factors and project activity factors, information can be identified and actively pushed to users who may be interested in the project but currently have low activity levels, thereby improving overall user activity.

[0054] Example 2: In the above examples, users and projects are classified in detail, and different activity levels for different users and different priorities for projects are set. An optimization scheme is set to optimize and adjust the push strategy. This example further limits the scope of the example.

[0055] The optimization scheme also includes a related project recommendation rule, which selects complementary projects that are strongly related to the project with the highest current attention from the user based on the project association matrix, and pushes information based on the complementary projects to guide and optimize the user's access path.

[0056] The project association matrix is ​​constructed by analyzing the degree of association between smart healthcare projects using an association rule learning algorithm. The strongly associated complementary projects refer to projects with an association degree greater than the association threshold.

[0057] Specifically, association rule learning algorithms (such as Apriori or FP-Growth) are used to identify frequent itemsets. These algorithms are well-known and common techniques in the field. The algorithm identifies frequently occurring item combinations, analyzes the association between items based on these frequent itemsets, and constructs an item association matrix based on the calculated association strength. This matrix is ​​a two-dimensional matrix where each cell represents the association strength between two items. The association rule learning algorithm is then applied to find association rules between items, and the association matrix is ​​populated based on these rules. A threshold is set, and the association strength between two items is calculated. When the association strength exceeds the threshold, a strong association is considered to exist between the two items. When pushing information to users, if a user visits a specific item, other items in the association matrix with an association strength exceeding the threshold are searched and pushed as secondary recommended items.

[0058] The user access path optimization refers to identifying key nodes in the user access path and inserting complementary items at these key nodes to achieve a natural transition. The key nodes refer to the project pages that the user visits and stays on, and complementary items are inserted during the time the user stays on the project page.

[0059] Specifically, user access path optimization refers to inserting complementary items as guidance and prompts to enable users to transition more naturally to other related items during their visit, thereby increasing the overall activation frequency of items and user retention rate. This involves identifying key nodes in the user access path—pages or items that users frequently visit or linger on—and inserting guidance and prompts for other strongly related items at these key nodes, such as pop-up recommendations, sidebar prompts, or bottom navigation bar recommendations. Furthermore, based on users' historical behavior and preferences, the inserted related guidance is personalized to improve its targeting and effectiveness.

[0060] This embodiment introduces project correlation analysis to construct a project correlation matrix, quantifying and clarifying the inherent connections between projects. This deepens the understanding of relationships between projects and provides a more accurate basis for subsequent information push and user guidance. By identifying strong correlation rules between projects, when a user visits a project, closely related complementary projects can be intelligently recommended, achieving a highly accurate second recommendation and greatly improving the user experience. The insertion of guidance and prompts for complementary projects into the user's access path allows users to transition more naturally to other related projects, effectively increasing the overall activation frequency of projects and user retention. By guiding users to explore more smart healthcare projects, user stickiness and satisfaction with the platform are enhanced.

[0061] Example 3 introduces project correlation analysis and constructs a project correlation matrix in the above examples. This enables secondary push during the information push process, improving the user experience and increasing the project activation rate. This example further limits the above content.

[0062] The push feedback data is obtained, the user push service results are integrated, the matching value of each push item received by the user is calculated, the matching value of all push items for each user is counted and the median is calculated, and the initial item matching value range is set with the median as the origin; the initial item matching value range is debugged according to the debugging rules to obtain the accurate item matching value range;

[0063] Specifically, the push notification feedback data includes user click counts, user dwell time, user conversion rate, and user feedback. It records whether a user clicked on the push notification, tracks the time a user spends on the push content page, and collects user satisfaction with the push content through surveys, rating systems, or direct feedback channels. If a user initiates a consultation, appointment, or check-up due to the push notification, a conversion rate is generated. The integrated user push notification business results are categorized according to dimensions such as push project, user group, and time, and statistical measures such as the average, median, and standard deviation of various indicators are calculated to reflect the overall situation.

[0064] The matching value is an indicator used to measure the degree of fit between the pushed items and the user's interests and needs. A standard value is set for each indicator in the push feedback data (number of user clicks, user dwell time, user conversion rate, etc.). The standard value is based on the average value of each indicator in a certain item of the user. Each indicator of the item is compared with the average value to obtain the ratio of each indicator. The matching value is generated based on the ratio of each indicator.

[0065] For example, if the average score of all metrics for a certain project across all users is 50, and user A scores 80 for all metrics for that project, with a ratio of 1.6 for each metric, the combined score of 4.8 is the final matching value.

[0066] Specifically, the matching values ​​of all push items for each user are counted and the median is calculated. The initial item matching value range is set with the median as the origin. For example, the lower limit of the initial item matching value range is set by floating 30% downward from the median as the origin, and the highest value is used to form the initial item matching value range.

[0067] The debugging rule involves performing pre- and post-tests on the lower limit of the initial project matching value range until the maximum number of pre-tests is reached. User feedback and push effects are observed, and the project matching value range is adjusted based on the pre-test results. The maximum number of pre-tests is preset.

[0068] Specifically, the debugging rules are based on the initial project matching value range. Pre-testing is performed before and after the lower limit. The pre-test range can be preset according to the actual situation. Two sets of test projects are created: one set with matching values ​​slightly lower than the lower limit (e.g., 5% lower), and the other set slightly higher than the lower limit (e.g., 5% higher). Other variables (e.g., push time, content type, etc.) are kept consistent. Pushes are sent to users, and user feedback on the pushed projects is recorded, including click-through rate, conversion rate, and dwell time. Simultaneously, other users similar to the first user are selected, and pre-testing is performed using the same method, with data recorded. User feedback data from the two sets of test projects is compared to observe the different user reactions to projects with slightly lower and slightly higher matching values, as well as the different feedback effects among similar users. Based on the pre-test results of the two sets of experiments above, if multiple users respond positively to items with slightly lower matching values, the lower limit of the range can be appropriately lowered; if multiple users respond better to items with slightly higher matching values, the lower limit of the range can be appropriately raised. Select items within the new matching value range for push notifications, and repeat the above steps until the maximum number of pre-tests (e.g., set to 5 times). Based on the results of the five pre-tests, determine the final item matching value range, i.e., the precise item matching value range.

[0069] In this embodiment, the matching value of all projects involving a user is calculated by analyzing the project results. After debugging, the most accurate range of project matching values ​​is obtained. Within this range, the projects are those that are most interesting to the user and highly match their needs. Based on the projects within the accurate project matching value range, the user is more likely to receive projects that match their interests and needs, thereby improving user satisfaction and acceptance, and ultimately enhancing user stickiness and activity.

[0070] Example 4: In the above examples, the most accurate range of project matching values ​​is obtained, and the user is recommended projects based on the accurate range of project matching values. This example further limits the above.

[0071] The precise project matching value range is obtained, and the project matching value range is dynamically adjusted in combination with a random exploration mechanism to form a fluctuating project matching value range. Information push is optimized based on the fluctuating project matching value range.

[0072] The random exploration mechanism randomly selects a predetermined number of untouched items to fill the precise item matching value range, and sets a time interval to periodically change the items and remove invalid items from the precise item matching value range; the predetermined number is determined by the number of items in the precise item matching value range and the random perturbation factor to increase the number of untouched items.

[0073] Specifically, the designated number in the random exploration mechanism is determined by combining historical data and user feedback, specifically based on the number of items within the precise item matching value range and the random perturbation factor. This ensures that while maintaining a certain level of recommendation accuracy, diversity is increased. The random perturbation factor should be set based on historical user behavior data, actual test results, and business objectives. The initial value can be set to a small percentage, such as 5% or 10%, and then adjusted according to the actual effect. If the recommendation accuracy is too high, causing users to always see similar content, the value of the random perturbation factor can be appropriately increased. The number of newly added untouched items is determined by the random perturbation factor and the number of items within the precise item matching value range. For example, if the random perturbation factor is 5% and the number of items within the precise item matching value range is 100, then the designated number of newly added untouched items is 5.

[0074] Specifically, a reasonable time interval is set, which can be daily, weekly, or monthly, depending on the timeliness of the content and the frequency of user behavior. If user interaction is frequent, the update cycle can be shortened. In each update cycle, some recommended content is replaced based on user behavior and feedback data to ensure that users can continuously access fresh and popular content. The invalidation of a project is determined by a comprehensive assessment of user feedback, the project's timeliness, and user behavior data. Specifically, negative user feedback, such as low ratings, negative comments, or quickly skipping; outdated timeliness, such as a seasonal science popularization or temporary medical lecture; and user behavior data, if a project has not been interacted with by users for an extended period. A comprehensive assessment is made based on these three factors, and if any of these conditions are met, the project is deemed invalid. Preferably, the user behavior data assessment can be based on a time threshold. For example, if a project has not been clicked by users or has a rating below a certain value (e.g., 2 points) for a continuous period (e.g., one month), it is deemed invalid.

[0075] This embodiment introduces a random exploration mechanism based on the above embodiments, which can increase the diversity of recommendations while ensuring the accuracy of the recommendations. This diversity is reflected not only in the types of recommended content, but also in the recommendation of new content, thereby helping users discover more content of interest; by dynamically adjusting the content of items within the range of item matching values, it can continuously adapt to changes in user interests and content updates. This dynamic adaptability can always remain in sync with user needs; setting time intervals ensures that users can access the latest smart healthcare projects in a timely manner.

[0076] Example 5: This example provides an information push system based on smart medical big data, such as... Figure 2 As shown, the system includes: a data acquisition module, used to acquire smart healthcare project information, session feature data, and statistical results of push business information;

[0077] The classification calculation module receives data from the data acquisition module, classifies users and items, and calculates the activation rate of single-category items.

[0078] The activity factor setting module is used to receive the activation rate of a single category item from the category calculation module and set the item activity factor and user activity factor.

[0079] The optimization scheme module is used to receive the activation rate of single-category projects, project activity factors, and user activity factors to generate optimization schemes.

[0080] The feedback adjustment module is used to monitor and obtain push feedback data in real time, and adjust and optimize strategies in real time.

[0081] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0082] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0087] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An information push method based on smart healthcare big data, characterized in that, The method includes: S100: Obtain smart healthcare project information and corresponding session feature data, and at the same time obtain push business information statistics results, and classify users and projects according to smart healthcare project information and corresponding session feature data; S200: For each user category and item category combination, calculate the single category item activation rate, and use a weighted average method to calculate the comprehensive item activation rate for multiple single category item activation rates; S300: Set the project activity factor based on the activation rate of the single category project; set the user activity factor based on the number of project types involved by a single user; S400: Based on the activation rate of the single-category item, combined with the item activity factor and the user activity factor, generate an optimization plan; optimize the information push strategy according to the optimization plan, push information according to the optimized information push strategy, and monitor and obtain push feedback data in real time. The optimization scheme is to classify users and projects into levels based on the single-category project activation rate, project activity factor, and user activity factor, and set a cross-pre-test push strategy; the cross-pre-test push strategy is to push high-priority projects to low-activity users and low-priority projects to high-activity users, thereby generating the final information push optimization strategy. Based on push feedback data, calculate the matching value of each push item received by the user and set an initial item matching value range; debug the initial item matching value range according to debugging rules to obtain an accurate item matching value range; dynamically adjust the item matching value range by combining a random exploration mechanism to form a fluctuating item matching value range, and optimize information push based on the fluctuating item matching value range; The random exploration mechanism randomly selects a predetermined number of untouched items to fill the precise item matching value range, and sets a time interval to periodically change the items and remove invalid items from the precise item matching value range; the predetermined number is determined by the number of items in the precise item matching value range and the random perturbation factor to increase the number of untouched items. S500: Evaluate the push optimization effect based on the activation rate of the comprehensive project, and generate a user trend change report for data analysis to adjust the optimization strategy in real time; Specifically, the matching value of all push items for each user is calculated and the median is determined. The initial range of item matching values ​​is set with the median as the origin. The matching value is used to measure the degree of fit between the push items and the user's interests and needs. A standard value is set for each indicator in the push feedback data, and it is compared with the average value. The matching value is generated based on the ratio of each indicator. The indicators include the number of user clicks, user dwell time, and user conversion rate.

2. The information push method based on smart medical big data according to claim 1, characterized in that, Users and projects are categorized into levels based on single-category project activation rate, project activity factor, and user activity factor, including: dividing users into highly active, moderately active, and low-active users; and dividing projects into three priorities: high, medium, and low. The optimization scheme sets a basic push strategy, which is to push information at different frequencies for different active users and different priority items. The final information push optimization strategy is generated, including: setting time limits to collect user feedback data, analyzing the feedback results collected from the cross-test push strategy, and determining the push frequency and scope in combination with the basic push strategy.

3. The information push method based on smart medical big data according to claim 2, characterized in that, The optimization scheme also includes a related project recommendation rule, which selects complementary projects that are strongly related to the project with the highest current attention from the user based on the project association matrix, and pushes information based on the complementary projects to guide and optimize the user's access path.

4. The information push method based on smart medical big data according to claim 3, characterized in that, The project association matrix is ​​constructed by analyzing the degree of association between smart healthcare projects using an association rule learning algorithm. The strongly associated complementary projects refer to projects with an association degree greater than the association threshold.

5. The information push method based on smart medical big data according to claim 3, characterized in that, The optimization of the user access path refers to identifying key nodes in the user access path and inserting complementary items at the key nodes to achieve a natural transition; the key nodes refer to the project pages that the user visits and stays on, and complementary items are inserted during the time the user stays on the project page.

6. The information push method based on smart medical big data according to claim 1, characterized in that, The debugging rule involves performing pre- and post-tests on the lower limit of the initial project matching value range until the maximum number of pre-tests is reached. User feedback and push effects are observed, and the project matching value range is adjusted based on the pre-test results. The maximum number of pre-tests is preset.

7. An information push system based on smart medical big data, accompanied by an information push method based on smart medical big data as described in any one of claims 1 to 6, characterized in that, The system includes: a data acquisition module, used to acquire smart healthcare project information, session feature data, and statistical results of push business information; The classification calculation module receives data from the data acquisition module, classifies users and items, and calculates the activation rate of single-category items. The activity factor setting module is used to receive the activation rate of a single category item from the category calculation module and set the item activity factor and user activity factor. The optimization scheme module is used to receive the activation rate of single-category projects, project activity factors, and user activity factors to generate optimization schemes. The feedback adjustment module is used to monitor and obtain push feedback data in real time, and adjust and optimize strategies in real time.

Images