A system, method and medium for APP information push for one-card

Abstract

The present invention belongs to the field of data processing technology, and relates to an APP information push system, method and medium for a one-card pass, including: a data acquisition module, a database, an ETL module, a model module and a push module, the data acquisition module is used to acquire the user's data and behavior; the database is used to store and evaluate the user's data and behavior; the ETL module is used to process the user's data and behavior according to the data processing model; the model module is used to bring the user's data and behavior processed by the ETL module into the data processing model for clustering calculation, and divide the data and behavior into several categories; the push module is used to push information to the current user according to the data and behavior category. It analyzes the behavior analysis data generated inside the APP, improves the accuracy of the random demand analysis, and thus improves the user's experience of using the APP.

Description

Technical Field

[0001] The present invention relates to an APP information push system, method and medium for an all-in-one card, and belongs to the technical field of data processing. Background Art

[0002] The number of active users is the basis for the survival of an APP. Users can bring value to the product. In addition to the strong demand functions of the APP, information push is also a good way to gain special attention from users and start APP activation. Most users will only use a few functional points in the APP. Messaging can become a part of the APP content service to a certain extent, increasing user usage and user experience. Wake up dormant users and improve retention rate.

[0003] At present, APP has not yet conducted correlation analysis on the data generated by users to generate additional value and provide services to users. If the data is not utilized through an effective knowledge system, then static information cannot enable creators, users and managers to obtain greater benefits. In addition, among the data used for analysis, there is less behavioral analysis data generated by the APP, and a large amount of data is factual attributes generated by external non-user behavior. User needs are divided into inherent needs, long-term needs and random needs. At present, the basic attributes and needs of users are relatively clear through external third-party data (diagnostic information, work location, property certificates, etc.). The long-term needs of users can also be understood through information such as schools and children. However, the random needs of users require the user's recent behavior as production materials, and random needs are less related to each other, so it is difficult to make accurate predictions. Summary of the invention

[0004] In view of the above problems, the purpose of the present invention is to provide an APP information push system, method and medium for a one-card pass, which analyzes the behavior analysis data generated inside the APP, improves the accuracy of random demand analysis, and thus improves the user's experience of using the APP.

[0005] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme: an APP information push system for one-card pass, comprising: a data acquisition module, a database, an ETL module, a model module and a push module, the data acquisition module is used to acquire user data and behavior; the database is used to store and evaluate user data and behavior; the ETL module is used to process the user data and behavior according to the data processing model; the model module is used to bring the user data and behavior processed by the ETL module into the data processing model for clustering calculation, and divide the data and behavior into several categories; the push module is used to push information to the current user according to the data and behavior category.

[0006] Furthermore, the data acquired by the data acquisition module includes static data and dynamic data, wherein the static data is extracted from the one-card pass, using the data from the source database in the one-card pass, and is acquired for the first time by full processing, and subsequent daily data collection is acquired by inquiring about data changes.

[0007] Furthermore, the ETL module includes processing the data and behaviors to synchronize them with the data processing model before the data and behaviors are calculated; and standardizing the output data format of the data and behaviors after the data and behaviors are calculated.

[0008] Furthermore, the ETL process of the ETL module is completed through batch processing or streaming processing.

[0009] Furthermore, the data extraction scope of the ETL module in the database is determined by the one-card business system. The one-card business system and the ETL module realize data interaction by determining data specifications. The data extraction process is: the one-card business system determines the physical structure list of the database, and generates files for the data through the ETL module. After the files are landed, the data is pulled into the ETL module via FTP or under the premise of permission. The ETL module extracts the data of the one-card business system in incremental form every day or at regular intervals, and forms a data file.

[0010] Furthermore, before data extraction, data checking and cleaning are required. The data checking and cleaning work includes the following steps: checking the task status, setting the loading job flag, and when an error occurs, the program automatically reports and stops processing this job and subsequent operations of related jobs; checking the execution log, and understanding the omissions by comparing the data entry and processing numbers; data cleaning is limited to serious errors that affect data loading, cleaning the contents of the error field, and generating a cleaning log.

[0011] Furthermore, several categories of tags are generated through the tags of content browsed by users, while clicks where users stay for a short time are filtered out; corresponding weights are set for different user behaviors, with the weight of behaviors with a short time interval with the test being higher than the weight of behaviors with a long time interval with the test; if the user has not viewed the recommended content, the weight of the corresponding tag is reduced.

[0012] Furthermore, the push module uses a short link to push information. The push module cannot be actively used by users. When a user logs in to the App, the user client implicitly sends a request to the server. After receiving the request, the server responds to the request and calls the push module to send push information to the user.

[0013] The present invention also discloses a method for pushing APP information for an all-in-one card, which adopts any of the above-mentioned APP information push systems for an all-in-one card, and includes the following steps: S1 pulls the user's data and behavior to a specified directory of the server; S2 loads and saves the user's data and behavior in the specified directory of the server into a database; S3 batch processes the data and behavior according to the logic of the data processing model, and inputs them into the data processing model for calculation, so as to obtain the classification of the data and behavior; S4 obtains push information according to the classification result, and pushes the push information to the corresponding user.

[0014] The present invention also discloses a computer-readable storage medium, on which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned APP information push method for the one-card pass.

[0015] Since the present invention adopts the above technical scheme, it has the following advantages: 1. This scheme analyzes the behavior analysis data generated inside the APP, improves the accuracy of random demand analysis, and thus improves the user's experience of using the APP. 2. In order to obtain the user's needs, in addition to accessing the user's inherent information, the present invention also needs to collect user behaviors at various functions inside the APP. 3. The final way to obtain data is to collect what information the user uses when clicking on a certain module. The system persists this information and uses it as a parameter to fall into the corresponding model after algorithm processing, and recommends messages that meet the user's preferences to the user. 4. The present invention can obtain statistics on the request interface, that is, it can understand that users have stronger needs for certain functions inside the APP, and then analyze the information with low real-time requirements in such information, create fixed data models and dimensional data, and use data batch processing to first meet business classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of an APP information push system for an all-in-one card in one embodiment of the present invention;

[0017] Figure 2 This is an interface diagram of an APP information push system for an all-in-one card in one embodiment of the present invention;

[0018] Figure 3 It is a flow chart of a method for pushing APP information for an all-in-one card in one embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make those skilled in the art better understand the technical direction of the present invention, the present invention is described in detail through specific embodiments. However, it should be understood that the provision of specific embodiments is only for a better understanding of the present invention, and they should not be understood as limitations of the present invention. In the description of the present invention, it should be understood that the terms used are only for the purpose of description and cannot be understood as indicating or suggesting relative importance.

[0020] The present invention relates to an APP information push system, method and medium for an all-in-one card, which collects data and behaviors generated inside the APP and brings them into a data processing model, calculates data and behavior types through the data processing model, and sends corresponding push information to users of this type according to the determined behavior type, so that the push information is more in line with the user's interests and needs, thereby improving the user's usage experience.

[0021] Embodiment 1

[0022] This embodiment discloses an APP information push system for a one-card system. Figure 1 As shown, it includes: data acquisition module, database, ETL module, model module and push module.

[0023] The data acquisition module is used to obtain user data and behavior.

[0024] The data acquired by the data acquisition module includes static data and dynamic data. The static data is extracted from the one-card pass through the connection of the App with the peripheral system and part of the user information that the user needs to submit when using the App. The data in the source database of the one-card pass is first acquired by full processing. The subsequent daily data collection is acquired by asking for data changes. The way of asking for data changes produces the least pressure on the system. In data calculation, the pressure on the operation program is also reduced because of the small amount of data sets. The data acquisition module can provide relevant information such as users, access modules, time, etc., and uses statistical user requests for a certain APP to obtain user data as a supplementary dimension analysis.

[0025] Database, used to store and evaluate user data and rows. The database saves push strategy configuration, success / failure history, sent information, etc.

[0026] The ETL module is used to process user data and behaviors according to the data processing model.

[0027] After the collected data and behaviors are stored in the database, they need to be processed and rule-processed according to the data model. There are two intervention points for data processing in this system: the ETL module processes the data and behaviors before they are calculated to synchronize them with the data processing model; and standardizes the output data format of the data and behaviors after they are calculated.

[0028] The ETL process of the ETL module is completed through batch processing or streaming processing. Among them, batch processing is mainly suitable for querying or processing all or most of the static offline data that has been landed, such as querying information already in the data structure and using static data for analysis. It is aimed at large batches of data (such as GB or PB level), and the time required is generally several minutes to several hours. Stream processing is more suitable for real-time or instant information, such as real-time or instant analysis and data processing of dynamic data generated by users during search or query, targeting a single record or micro-batch data containing several records (such as KB or MB). The time required for stream processing is a few milliseconds to a few seconds. Stream processing is smaller and faster, and batch processing corresponds to a larger and more extensive data range. The two correspond to two methods, one is batch processing and the other is instant processing.

[0029] Batch processing can be understood as unified processing, and it is often run in batches or in large quantities using a scheduling method when business requests are idle. This method will not have a functional impact on business systems and programs, nor will it have an impact on the database, but it will have a certain impact on the timeliness of data applications. Generally, this method is implemented in scenarios with low real-time requirements, and various technical means are used, such as tools: KETTLE; components such as Sqoop; development languages ​​such as JAVA / PYTHON / PERL, etc.; the SQL language of relational or non-relational databases themselves can also be used, and the technical redundancy is relatively high. In this embodiment, batch processing is preferably used.

[0030] Instant processing mode, the object is the data interface. When the user requests or enters data, the data is classified and standardized according to the established standard format, and the data is landed. This method will have a certain impact on the timeliness of system response when a large number of data are requested, but it saves a lot of time and resources for subsequent data analysis. Generally, this method is used in scenarios with high real-time requirements, and the technical requirements are relatively high. The general technical application is Storm or Flume batch processing, and the Map layer of Hadoop can also be implemented to achieve custom cleaning.

[0031] Considering security and confidentiality, the data extraction scope of the ETL module in the database is determined by the one-card business system. The one-card business system and the ETL module realize data interaction by determining data specifications. The data extraction process is as follows: the one-card business system determines the physical structure list of the database, and the ETL module generates files for the data. After landing, the data is pulled into the ETL module by FTP or under the premise of permission. The ETL module extracts the data of the one-card business system in incremental form every day or regularly, and forms a data file. The benefits of interface extraction are: if the reading fails, it will not affect the data in the ETL database, and only needs to be re-extracted. Tasks that fail using data files can be re-run, which has little impact on other tasks. Interface files can be saved for a long time. Using interface files can easily verify data quality.

[0032] Before data extraction, data inspection and cleaning are required. The purpose of data inspection is to prevent errors in data sources and generate information reports. The benefits of this are to solve problems as soon as possible, avoid meaningless work and improve the quality of system data. Data inspection during data loading is to check the correctness of data loading according to rules. Data inspection and cleaning work includes the following steps: check the task status, set the loading job flag, and when an error occurs, the program automatically reports and stops processing the subsequent operations of this job and related jobs; check the execution log, and grasp the omissions by comparing the data warehouse and the number of data processed; clearly understand the amount of data that a job should process, the actual amount of data processed, which data is not processed and reports the error result, and common error results include violation of constraints, formatting errors, data duplication and other problems. Data cleaning is limited to serious errors that affect data loading, clean the contents of the error field, and generate a cleaning log. The cleaning log includes information such as the number of rows and fields, so as to correct the data back to the source system, and try to keep the original state of the data during the loading process, and do not clean the data.

[0033] The model module is used to bring the user data and behaviors processed by the ETL module into the data processing model for clustering calculation, and divide the data and behaviors into several categories. The model module stores the analyzed library table attributes, classification labels, weight attributes, label types, implementation data, etc.

[0034] Several categories of tags are generated through the tags of content that users have browsed, while filtering out clicks where users have stayed for a short time. Corresponding weights are set for different user behaviors, with the weight of behaviors that are closely related to the test time interval being higher than the weight of behaviors that are closely related to the test time interval being higher. If the user has not viewed the recommended content, the weight of the corresponding tag is reduced.

[0035] The push module is used to push information to the current user based on their data and behavior categories.

[0036] The data processed by ETL falls into different data models. The data model will classify the data, classify such users, and push the same type of information to the corresponding users. The push method combines the functions, performance, hardware resources and user operations of the one-period one-card system. The push module uses a short link to push information. The push module cannot be actively used by users. When the user logs in to the App, the user client implicitly sends a request to the server. After receiving the request, the server responds to the request and calls the push module to send push information to the user. The backend classifies users and completes the content to be pushed in advance through user analysis, without adding too many new development elements. This method is not only beneficial to the stability of the current system, but also does not occupy additional data and service resources: the short link solution is based on the user's request through the interface, pulls the server data to the user's local application and displays it, and follows the HTTP protocol. The advantage is that it takes up less server resources; the disadvantage is that it depends on the user's operation on the application. If the user does not use the App, the precise push operation will not occur.

[0037] The following is a specific example to illustrate the specific process of HTTP protocol short link: First, instantiate the URL object, and pass in the URL of the requested data as a parameter. Calling the open Connection method of the URL object will return a URLConnection object, and the URL Connection class is the parent class of the Http URL Connection class, which can be forced to convert to the required Http URL Connection object. Set the connection properties through the obtained Http URL Connection object, such as setting the connection type "GET" or "POST" by set Request Method, setting the read timeout limit by set Read Timeout, etc. Get the response code, which is used to confirm whether the connection is connected. If the return value is Http URL Connection.HTTP_OK (200), the connection is successful. Read and parse the input stream: You can get an input stream through the Http URL Connection object, read the content of the input stream locally, and then parse it. You can directly use JSONObject for parsing, or use a third-party framework. It is recommended to use gson. Gson parses ordinary json objects: The use of gson depends on JSONObject. Through the getString method of the JSONObject object, the corresponding data is obtained in the form of a string, and then it is parsed into a specified class. gson parses array data: The steps for parsing array data are basically the same as those for ordinary json objects. The difference is that the first parameter of the fromJson method is a string that satisfies the json array format, and the second parameter is a Type object, which needs to be obtained through the getType method of the TypeToken object. Pass data back to the main thread: Since network operations cannot be performed in the main thread, and child threads are not allowed to operate the UI, the parsed data needs to be passed back to the main thread.

[0038] like Figure 2As shown, it is the operation interface of the system in this embodiment. The data system should currently adopt the forms of directional information capture, directional information classification (user and push information), online user editing, etc., that is, regularly maintain push information by category, regularly maintain push groups by category, and send pre-compiled information according to conditions. The push platform is presented in a visual form and can be used as an auxiliary system plug-in on the one-card application. The push content of the message push system can be a self-edited web page, and it should also be able to guide to the details page of the original content by sending content. The different messages classified by the model and different user groups are organically linked together through a series of operations such as checking to ensure that the group receiving a certain tag information is the one that this part of the users cares about. Reverse operation, using users as groups can also be associated with information of the same tag. The system can customize the sending of messages and send customizable information to a certain group or all groups; when the information is sent incorrectly, the message can be withdrawn in time. The frequency of sending information in the information library to the group can be configured in units of hours / days and custom time frequency, but the same information cannot be sent repeatedly. The user clicks to view the content of the brief information sent by the system, and can directly jump to the original URL of the system information source through the URL in the content.

[0039] The following is an example of popular outpatient recommendations. For example, it is recommended that patients choose hospitals with rich experience. The classification labels are hospital, outpatient department, department, follow-up visit, and patient. Taking the disease classification of the department visited by the user as the target, the number of outpatient visits for each disease in each hospital is counted. It is believed that the outpatient department of the hospital with the most outpatient visits has rich clinical experience, and it is recommended to users who often perform outpatient operations. The recommended table is shown in Table 1.

[0040] Table 1 Popular clinic recommendation table

[0041] HOUSEHOLDER_IDCARD ID card / health card number of household head NAME Name

[0042] DIS_NAME Disease diagnosis name ORGAN_NAME Hospital Name

[0043]

[0044]

[0045] Embodiment 2

[0046] Based on the same inventive concept, this embodiment discloses a method for pushing APP information for a one-card pass, using any of the above-mentioned APP information push systems for a one-card pass, such as Figure 3 As shown, the following steps are included:

[0047] S1 pulls the user's data and behavior to the specified directory of the server;

[0048] S2 loads and saves the user's data and behavior in the directory specified by the server into the database;

[0049] S3 batch processes data and behaviors according to the logic of the data processing model, and inputs them into the data processing model for calculation, thereby obtaining the classification of data and behaviors;

[0050] S4 obtains push information according to the classification result, and pushes the push information to the corresponding user.

[0051] Embodiment 3

[0052] Based on the same inventive concept, this embodiment discloses a computer-readable storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the above-mentioned APP information push method for the one-card pass.

[0053] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0054] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0055] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the specific implementations of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not deviate from the spirit and scope of the present invention should be included in the protection scope of the claims of the present invention. The above content is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An APP information push system for a one-card system, characterized in that: include: Data acquisition module, database, ETL module, model module and push module, Data acquisition module, used to obtain user data and behavior; Databases for storing and evaluating the data and behavior of said users; An ETL module is used to process the user's data and behavior according to a data processing model; A model module is used to bring the user's data and behavior processed by the ETL module into the data processing model for clustering calculation, and classify the data and behavior into several categories; Push module, used to push information to the current user based on their data and behavior categories; The data acquired by the data acquisition module includes static data and dynamic data, wherein the static data is extracted from the one-card pass, using the data of the source database in the one-card pass, and is acquired by full processing for the first time, and the subsequent daily data collection is acquired by querying data changes; The ETL module includes processing the data and behaviors to synchronize them with the data processing model before the data and behaviors are calculated; and standardizing the output data format of the data and behaviors after the data and behaviors are calculated; The data extraction range of the ETL module in the database is determined by the one-card business system. The one-card business system and the ETL module realize data interaction by determining data specifications. The data extraction process is as follows: the one-card business system determines the physical structure list of the database, generates files for the data through the ETL module, and then pulls the data into the ETL module in FTP mode or under the premise of permission. The ETL module extracts the data of the one-card business system in the form of increments every day or regularly, and forms a data file; The data processed by ETL falls into different data models. The data model will classify the data, classify such users, and push the same type of information to the corresponding users. The push method is based on the one-period-one-card function, performance, hardware resources and user operation reference factors. The push module uses a short link to push information. The push module cannot be actively used by users. When the user logs in to the App, the user client implicitly sends a request to the server. After receiving the request, the server responds to the request and calls the push module to send push information to the user.

2. The APP information push system for the all-in-one card according to claim 1, characterized in that: The ETL process of the ETL module is completed by batch processing or streaming processing.

3. The APP information push system for the one-card system as claimed in claim 2, characterized in that: Before performing the data extraction, data checking and cleaning work is required, and the data checking and cleaning work includes the following steps: checking the task status, setting the loading job flag, and when an error occurs, the program automatically reports and stops processing this job and subsequent operations of related jobs; checking the execution log, and grasping the omissions by comparing the data entry and processing numbers; data cleaning is limited to serious errors that affect data loading, cleaning the contents of the error field, and generating a cleaning log.

4. The APP information push system for one-card according to any one of claims 1 to 3, characterized in that: The tags of the several categories are generated by the tags of the content browsed by the user, and the clicks of the user with short stay time are filtered out; and corresponding weights are set for different behaviors of the user, and the weight of the behavior with a short time interval with the test time is higher than the weight of the behavior with a long time interval with the test time; If the user does not view the recommended content, the weight of the corresponding tag will be reduced.

5. The APP information push system for one-card according to any one of claims 1 to 3, characterized in that: The push module uses a short link to push information. The push module cannot be actively used by users. When a user logs in to the App, the user client implicitly sends a request to the server. After receiving the request, the server responds to the request and calls the push module to send push information to the user.

6. A method for pushing APP information for a one-card system, characterized in that: The APP information push system for the one-card system as described in any one of claims 1 to 5 is adopted, comprising the following steps: S1 pulls the user's data and behavior to the specified directory of the server; S2 loads and saves the user's data and behavior in the directory specified by the server into the database; S3 batch processes the data and behaviors according to the logic of the data processing model, and inputs them into the data processing model for calculation, thereby obtaining the classification of the data and behaviors; S4 obtains push information according to the classification result, and pushes the push information to the corresponding user.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the APP information push method for a one-card pass as described in claim 6.

Images