Information push method, related device, and computer medium

Abstract

The present application discloses an information push method, a related device and a computer storage medium, wherein the method comprises: obtaining a teaching push strategy and a user's change data; preprocessing the user's change data to obtain the user's training data; inputting the user's training data into a prediction training model, and outputting the user's prediction result; and pushing teaching information to the user based on the user's prediction result and the teaching push strategy. Thus, the embodiment of the present application can adjust the teaching progress for the user in real time according to the user's learning status and the pre-formulated teaching push strategy, which not only solves the problems of delay and the like existing in the prior art and improves the user experience, but also can formulate corresponding learning plans for different teaching programs, thereby effectively improving the user's enthusiasm for learning.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an information push method, related devices, and computer media. Background Art

[0002] Most online education platforms use a teaching process that combines teaching, practice, testing, and evaluation to push teaching content. Currently, most online education platforms use a data warehouse system between the business system and the model system to process and store student data, but the data warehouse system has a certain delay and cannot dynamically adjust the pushed learning content according to each student's current learning status. Summary of the invention

[0003] The embodiments of the present application provide an information push method, related devices and computer storage media to solve the problems of delay and so on existing in the prior art.

[0004] In a first aspect, an embodiment of the present application provides an information push method, the method comprising:

[0005] Obtaining teaching push strategy and user change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data;

[0006] Preprocessing the changed data of the user to obtain training data of the user;

[0007] Input the training data of the user into the prediction training model, and output the prediction result of the user; wherein the prediction training model is generated based on the sample change data after preprocessing and training, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample operation change data;

[0008] Teaching information is pushed to the user based on the prediction result of the user and the teaching push strategy.

[0009] In a second aspect, an embodiment of the present application provides an information push device, the device comprising:

[0010] An acquisition module is used to acquire teaching push strategies and user change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data;

[0011] An obtaining module, used for preprocessing the changed data of the user to obtain the training data of the user;

[0012] An output module, used to input the training data of the user into the prediction training model and output the prediction result of the user; wherein the prediction training model is generated by training based on the sample change data after preprocessing, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample job change data;

[0013] A push module is used to push teaching information to the user based on the prediction result of the user and the teaching push strategy.

[0014] In a third aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.

[0015] In a fourth aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory;

[0016] The memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above method steps.

[0017] The beneficial effects brought about by the technical solutions provided by some embodiments of the present application include at least:

[0018] In the embodiment of the present application, the teaching push strategy and the user's change data can be obtained; the user's change data can be preprocessed to obtain the user's training data; the user's training data can be input into the prediction training model, and the user's prediction result can be output; and the teaching information can be pushed to the user based on the user's prediction result and the teaching push strategy. Therefore, the embodiment of the present application can adjust the teaching progress for the user in real time according to the user's learning status and the pre-formulated teaching push strategy, which not only solves the problems of delay in the prior art and improves the user experience, but also can formulate corresponding learning plans for different teaching programs, thereby effectively improving the user's enthusiasm for learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 An application scenario diagram of an information push method provided in an embodiment of the present application;

[0021] Figure 2 A system architecture diagram of an information push method provided in an embodiment of the present application;

[0022] Figure 3 A flowchart of an information push method provided in an embodiment of the present application;

[0023] Figure 4 A flowchart of another information push method provided in an embodiment of the present application;

[0024] Figure 5 A schematic diagram of a preprocessing process of change data in an information push method provided in an embodiment of the present application;

[0025] Figure 6 A flowchart of another information push method provided in an embodiment of the present application;

[0026] Figure 7 A schematic diagram of a storage process of a prediction training model in an information push method provided in an embodiment of the present application;

[0027] Figure 8 A schematic diagram of the structure of an information push device provided in an embodiment of the present application;

[0028] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0030] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0031] Figure 1The schematic diagram of the application scenario of the information push method provided by the embodiment of the present application is exemplified. Among them: the embodiment of the present application can input the training sample change data in the training sample set into the pre-established prediction training model to obtain the prediction result corresponding to the training sample set, combine the prediction result with the business strategy to obtain the teaching information of the training sample set, further, push the teaching information to each user system, and train the prediction training model according to the feedback information of each user on the teaching information, until the feedback information of each user meets the prediction threshold range corresponding to the business strategy.

[0032] Specifically, the training sample set may include a variety of training sample data sources, such as but not limited to: sample video change data, sample content change data, sample homework change data. Among them, the sample video change data may be the number of interactions between the two sides of the terminal recorded by the camera device when collecting training samples, the collected expression status and other data, the sample content change data may be teaching content and other data, and the sample homework change data may be homework, test and other data corresponding to the teaching content. The prediction result may be the score or accuracy of users of different levels corresponding to the training sample set completing the same teaching content. The business strategy may be a pre-set teaching goal, so that the embodiment of the present application can push teaching tasks, homework questions, test questions and other data of different difficulty levels to each user according to the learning ability of users of different levels, and further adjust the accuracy of the prediction training model according to the feedback information of each user, that is, the completion of teaching tasks, homework questions and test questions by each user.

[0033] Figure 2The system architecture diagram of the information push method applied to the embodiment of the present application is shown. Among them, the information push method provided by the embodiment of the present application can be applied in a server. Specifically, the server can be connected to the terminal through a network. The network is used to provide a communication link between the terminal and the server. The network may include various connection types, such as wired, wireless communication links or optical fiber cables, etc. The terminal includes but is not limited to: wearable devices, monitoring devices, handheld devices, personal computers, tablet computers, vehicle-mounted devices, smart phones, computing devices or other processing devices connected to wireless modems, etc. In different networks, terminal devices can be called different names, such as: monitoring equipment, user equipment, access terminals, user units, user stations, mobile stations, mobile stations, remote stations, remote terminals, mobile devices, user terminals, terminals, wireless communication devices, user agents or user devices, cellular phones, cordless phones, personal digital assistants (PDA), terminal devices in fifth generation mobile communication technology (5th generation mobile networks, 5G) networks or future evolution networks, etc. The terminal system refers to an operating system that can be run on a terminal, a program for managing and controlling terminal hardware and terminal applications, and an indispensable system application for the terminal. The system includes but is not limited to Android system, IOS system, Windows phone (WP) system and Ubuntu mobile operating system.

[0034] It should be understood that Figure 2 The number of terminals, networks and servers in the example is only illustrative. Any number of terminals, networks and servers may be provided according to actual needs. For example, a server may be a server cluster composed of multiple servers. Users may use the server to interact with the terminal through the network to obtain an optimized version, etc.

[0035] Next, combine Figure 1 Application scenario diagram of the information push method introduced and Figure 2 The system architecture diagram introduced here is used to introduce the information push method provided in the embodiment of the present application.

[0036] In one embodiment, Figure 3 As shown in FIG, a flow chart of an information push method is provided. Figure 3 As shown, the information push method may include the following steps:

[0037] S301, obtaining teaching push strategy and user change data.

[0038] Among them, the user's change data may include any one or more of the following: teaching video change data, teaching content change data, and teaching assignment change data.

[0039] Specifically, the teaching information strategy is used to represent the pre-set teaching objectives. For example, the teaching information strategy of the English word spelling test module can be to determine the words that students have not mastered in the word library. The user's change data represents the data that has changed in the scene. For example, during the video class between the student and the teacher, the teaching video change data can be the number of times the student frowns, the number of times the student answers, the length of time the student answers, the number of times the teacher smiles, and other data. The teaching content change data can be the course content (such as the 8th lesson of children's English), and the evaluation data of the teaching content (such as the mutual evaluation of the content of this lesson by the student and the teacher). The teaching homework change data can be the knowledge points corresponding to the homework title, the time for the student to complete each homework question, whether the answer to the homework is correct, each unit test question, and whether the answer to the unit test question is correct.

[0040] S302, pre-processing the user's changed data to obtain the user's training data.

[0041] Preprocessing is used to indicate operations such as filtering the user change data to remove possible sample data or other irrelevant data.

[0042] Possibly, the embodiment of the present application can monitor the current data of the user; if the current data of the user is detected to have changed during the monitoring process, the changed data of the user can be obtained. Furthermore, the changed data of the user can be filtered, data completed, data format aligned, etc., so that the obtained training data of the user meets the data format requirements of the prediction training model and does not include possible sample change data or other irrelevant data.

[0043] Possibly, the current data may include any one or more of the following: video data, classroom data, homework data.

[0044] S303: Input the user's training data into the prediction training model, and output the user's prediction result.

[0045] Among them, the prediction training model is generated based on the sample change data after preprocessing and training. The sample change data may include any one or more of the following: sample video change data, sample content change data, and sample operation change data.

[0046] Possibly, the prediction training model can be a deep learning model built using a variety of training algorithms, such as the Deep Knowledge Tracing (DKT) model. Since the DKT model uses a recurrent neural network (RNN) network model to model users, the DKT model does not need to label the input data, and the more training data there is, the better the DKT model effect.

[0047] It is understandable that, since the structure of RNN can memorize historical input information and apply it to the prediction result of the current output, RNN is essentially an output input that can process a sequence of varying lengths. In RNN, each neuron inputs the output of the previous moment into the current moment state through self-connection, so the neuron input at the current moment includes two: one is the input change data at the current moment, and the other is the prediction result of the neuron output at the previous moment. Therefore, the more training data the DKT model using RNN has, the better the DKT model effect.

[0048] S304: Push teaching information to the user based on the user's prediction result and teaching push strategy.

[0049] Among them, the prediction result is used to indicate the user's ability to complete a certain course content. For example, for course A, the prediction result of student A with strong learning ability may be that he can successfully complete course A, while the prediction result of student B with average learning ability may be that it is difficult to complete course A. The teaching push strategy is used to indicate the pre-set teaching content completion goal. For example, the teaching push strategy can be to enable all students to successfully complete the pushed teaching content. The teaching information is used to indicate the teaching content, teaching homework and other information that the user needs to complete in the next stage. For example, based on the prediction result of student B (it is difficult to complete course A) and the teaching push strategy (to enable all students to successfully complete the pushed teaching content), course B with a lower course difficulty can be pushed.

[0050] In a specific example, in the grammar learning module, the teaching push strategy is to find the students' weak points in knowledge as much as possible. The change data of user Xiao Zhang includes: Intermediate Grammar Course 1, class duration 60 minutes, break time 5 minutes, student smiles 10 times, student frowns 3 times, teacher smiles 12 times, students are very satisfied with this course, teachers are very satisfied with this course, and teaching homework for this course is excellent. The change data of the above users is preprocessed to filter out unnecessary data: after a break of 5 minutes, the training data of user Xiao Zhang is obtained, and these training data are input into the prediction training model, and the prediction result can be output for the next teaching goal: it is expected that 90% of the content of Intermediate Grammar Course 2 can be completed; further, according to the prediction result and the teaching push strategy, the teaching content, teaching homework, and comprehensive test questions of Intermediate Grammar Course 1 and Intermediate Grammar Course 2 of Intermediate Grammar Course 2 are pushed to Xiao Zhang.

[0051] In the embodiment of the present application, the teaching push strategy and the user's change data can be obtained; the user's change data can be preprocessed to obtain the user's training data; the user's training data can be input into the prediction training model, and the user's prediction result can be output; and the teaching information can be pushed to the user based on the user's prediction result and the teaching push strategy. Therefore, the embodiment of the present application can adjust the teaching progress for the user in real time according to the user's learning status and the pre-formulated teaching push strategy, which not only solves the problems of delay in the prior art and improves the user experience, but also can formulate corresponding learning plans for different teaching programs, thereby effectively improving the user's enthusiasm for learning.

[0052] In some embodiments, Figure 4 The following is a flow chart showing an example of an information push method provided by an embodiment of the present application. Figure 4 As shown, the information push method may at least include the following steps:

[0053] S401, obtaining teaching push strategy and user change data.

[0054] Specifically, S401 is consistent with S301 and will not be described in detail here.

[0055] S402, obtaining the user's changed data through the message middleware, and outputting a plurality of first data streams.

[0056] Among them, the message middleware can be a distributed publish-subscribe message system Kafka that supports thousands of user-end reading and writing and can process hundreds of thousands of messages per second. In this way, the change data of a large number of users can be obtained through Kafka, and log processing and log aggregation can be performed to output multiple first data streams. The first data stream is used to represent all the data generated when completing a task. For example, the first data stream generated when the user completes the 5th homework question may include: user identification, question data, completion time, question duration, user answer, whether the answer is correct, etc.

[0057] Possibly, the embodiments of the present application can use Databus to obtain change data in different data sources. Among them, Databus is a low-latency distributed database synchronization system that provides reliable data capture, flow and data processing functions. For example, the user's change data generated by the business system falls into the relational database management system Mysql, and Databus can monitor these change data, process and encapsulate them in binary format binlog, and perform other operations. In addition, Databus can also write to the corresponding Kafka topic in real time and sequentially, so that the server can quickly obtain multiple first data streams of users by consuming data in Kafka.

[0058] S403: Obtain first training data based on multiple first data streams.

[0059] Possibly, the embodiment of the present application can clean and / or transform the data in each first data stream to obtain multiple second data streams; and extract data from the multiple second data streams based on preset extraction rules to obtain first training data.

[0060] Among them, the preset extraction rules can be pre-set to extract which change data in multiple second data streams. For example, student A completes Exercise 1 and Exercise 2 to generate two second data streams, respectively: ① student ID 111, question ID 12321, completion date 2021-3-26, completion time 15.00, question duration 2 minutes, correct answer; ② student ID 111, question ID 12322, completion date 2021-3-26, completion time 15.03, question duration 3 minutes, correct answer. The preset extraction rules can be to extract only the student ID, question ID, question duration, and whether to answer, then the extracted first training data includes: student ID 111, question ID 12321 and question ID 12322, question duration 2 minutes and 3 minutes respectively, and the answers are all correct.

[0061] Specifically, the embodiment of the present application can use a Data-app-engine based on the real-time data stream processing system Flink to distribute the data in multiple first data streams to the Flink Cluster for cleaning and / or conversion processing to obtain multiple second data streams respectively. Among them, the cleaning task is to delete the data that does not meet the requirements and complete and align the incomplete data. The data that does not meet the requirements mainly include sample data, erroneous data, and duplicate data. Sample data is mainly the data that may be left over when creating a prediction training model; erroneous data may be caused by the business system being not sound enough, and directly writing to the background database without judgment after receiving the input, such as numerical data input as full-width numeric characters, incorrect date format, date out of bounds, etc.; duplicate data may be repeated fields caused by unstable network signals, etc. Data conversion mainly involves encapsulating the original format of data into a new format. For example, when the second data stream is input into the data conversion system Data-Tranfer-sys, it needs to be converted into a data format customized by Data-Tranfer-sys.

[0062] S404, obtaining historical data of the user.

[0063] Among them, historical data can be obtained from the user database, for example, the data warehouse tool Hive, and the server can query the user's historical data from Hive, such as the student Xiao Wang's test records and video data last week.

[0064] S405: Verify the format information of the first training data based on the format information of the user's historical data to obtain second training data.

[0065] For example, the format of a historical data can be "question_id:1,student_id:1course_id:"MC1-LC1-U1-L1",concept_ids:[1,2,3],isCorrect:1,time_cost:100ms,NoOfRevision:1,type:1". Among them, question_id refers to the question ID; student_id refers to the user ID; concept_ids refers to the knowledge points contained in the current question; isCorrect refers to whether the current question is correct; time_cost refers to the time it takes to answer the question; NoOfRevision: refers to the process of students going back and forth on the answer while answering the question (for example, the process of students going back and forth on the answer before finally submitting the answer to the multiple-choice question); type refers to whether the current question is a homework or an assessment result.

[0066] Specifically, the embodiment of the present application can verify the format information of the first training data according to the above-mentioned format information of the historical data. If errors are found during the verification, adjustments can be made, such as adjusting the order of the changed data in the first training data, so that the obtained second training data can not only be input according to the predefined data format of the predictive training model, but also can accurately fall into the preset first database to store the second training data, so as to facilitate future queries by users or servers.

[0067] S406: Input the second training data into the prediction training model to output the prediction result of the user.

[0068] Specifically, S406 is consistent with S303 and will not be described in detail here.

[0069] S407, pushing teaching information to the user based on the user's prediction result and teaching push strategy.

[0070] Specifically, S407 is consistent with S304 and will not be described in detail here.

[0071] See also Figure 5 In a specific example, Databus is used to monitor the teaching video change data 51 of student A: the number of times the student spoke was 20, the number of times he smiled was 5, and the length of time he spoke was 15 minutes; the teaching content change data 52: the teaching content is the 8th lesson; the teaching homework change data 53: the length of time the student spent on the question was 20 minutes, the knowledge points corresponding to the question, and the answers were 10 correct and 3 wrong. These change data are binlog processed through Databus and then encapsulated into multiple first data streams of Kafka. Furthermore, Data-App-Engine_server is used to clean, transform, and process the multiple first data streams output by Kafka to filter out irrelevant data and transcapsulate the original data format into a new format customized by Data-Tranfer-sys. Since the amount of teaching change data obtained is large, the embodiment of the present application needs to first submit multiple first data streams to Flink's Data-app-engine to distribute multiple first data streams to Flink Cluster for cleaning and / or transformation processing to obtain multiple second data streams. Data-Tranfer-sys is further used to extract the teaching change data that the server is concerned about in each second data stream according to the preset extraction rules, thereby obtaining the first training data of student A. In addition, after verifying the data format of the first training data using the historical data obtained by querying Hive by the management end Geae of the data warehouse, the second training data that meets the data format requirements can be obtained. After the second training data is saved in the distributed database TIDB, it is input into the prediction training system to obtain the prediction result of the next learning stage of student A.

[0072] In some embodiments, Figure 6 The following is a flow chart showing an example of an information push method provided by an embodiment of the present application. Figure 6 As shown, the information push method may at least include the following steps:

[0073] S601, obtaining teaching push strategy and user sample change data.

[0074] Among them, the sample change data is used to represent the data used to establish the initial prediction training model, which may include any one or more of the following: sample video change data, sample content change data, and sample operation change data.

[0075] S602, pre-processing the sample change data to obtain sample training data.

[0076] Possibly, the preprocessing of the sample change data in the embodiment of the present application may include: using the message middleware to obtain and perform log processing and log aggregation to output multiple first sample data streams. The first sample data stream is used to represent all the data generated by completing a sample task. Furthermore, the data in each first sample data stream can be cleaned and / or transformed to obtain multiple second sample data streams; and data can be extracted from the multiple second sample data streams based on preset extraction rules to obtain first sample training data. Then, the format information of the first sample training data can be verified based on the preset data format information to obtain the second sample training data.

[0077] S603, creating an initial prediction training model, training the initial prediction training model based on sample training data, and generating a prediction training model.

[0078] Possibly, the embodiment of the present application can train the initial prediction training model based on the second sample training data to generate a prediction training model.

[0079] Possibly, the initial prediction training model may include: collecting sample data, including course education objectives, learning resources, course structure, teaching strategies, and practice test question banks in various fields, and extracting and annotating ontology terms for learning resources. According to the collected initial learning data, after entering the connection and sequence between each meta-knowledge point, a directed knowledge model is constructed using an ontology editor based on the extracted and annotated ontology terms and the connection and sequence between meta-knowledge points. Further, according to the acquired sample change data, the knowledge space of the sample knowledge model is initialized, the learning process status data is updated in real time, and the DKT model of multiple knowledge points is modeled on the learning process status data through the DKT algorithm, the mastery status of each knowledge point is fed back, and the learner's knowledge space is dynamically updated; further, the mastery status of each knowledge point is returned from the DKT to form the learner's knowledge space; the knowledge space is compared with the knowledge structure of the knowledge model to match the corresponding learning path.

[0080] S604, obtaining teaching push strategy and user change data.

[0081] Specifically, S604 is consistent with S301 and will not be described in detail here.

[0082] S605: Preprocess the user's changed data to obtain the user's training data.

[0083] Specifically, S605 is consistent with S302 and will not be described in detail here.

[0084] S606: Input the user's training data into the prediction training model, and output the user's prediction result.

[0085] Specifically, S606 is consistent with S303 and will not be described in detail here.

[0086] S607, pushing teaching information to the user based on the user's prediction result and teaching push strategy.

[0087] Specifically, S607 is consistent with S304 and will not be described in detail here.

[0088] S608, based on the user's teaching information, obtaining change data corresponding to the teaching information.

[0089] The teaching information may include teaching video data, teaching content data, and teaching homework data that the current user needs to complete. When the user completes the teaching information, the change data corresponding to the teaching information may be generated.

[0090] S609, determining the accuracy of the prediction training model based on the change data corresponding to the teaching information.

[0091] Among them, the accuracy of the prediction training model is used to represent the comprehensive score obtained when the user completes the pushed teaching information. For example, according to the change data of user Xiao Zhang completing Intermediate Grammar Course 2: the student smiled 3 times, the student frowned 5 times, the teacher smiled 6 times, the student's evaluation of this course was average, the teacher's evaluation of this course was satisfactory, the teaching assignments for this course were good, etc., it can be determined that the accuracy of the prediction training model is 70%.

[0092] S610: When the accuracy of the prediction training model does not reach the accuracy threshold, the prediction training model is adjusted based on the change data corresponding to the teaching information.

[0093] The accuracy threshold is used to indicate the preset accuracy that can be achieved when the user completes the teaching information. For example, based on the pushed teaching information, the accuracy threshold can be preset to be 80% for the correctness of the homework exercises.

[0094] It is understandable that when the accuracy of the prediction training model does not reach the accuracy threshold, the embodiment of the present application needs to train the prediction training model based on the change data corresponding to the teaching information. For example, according to the prediction result of student B (strong learning ability) and the teaching push strategy (discovering weak knowledge points), teaching content with increased difficulty level is pushed. The prediction threshold range is that the evaluation result is at least greater than 80 points. If the evaluation result of student B is 78 points, it is necessary to adjust the parameters of the prediction training model based on the various change data corresponding to the teaching content of student B this time, such as adjusting the learning path in the model, reducing the difficulty of the questions, so that students can successfully complete the pushed teaching information and achieve the expected goals.

[0095] It can be understood that when the accuracy of the prediction training model reaches the accuracy threshold, the training is completed and returns to step S604.

[0096] Possibly, in an embodiment of the present application, when the change data corresponding to the teaching information meets the prediction threshold range, the unmastered knowledge point status in the teaching information can continue to be pushed again as the current knowledge point; when the knowledge point status is mastered, the system finds its successor knowledge point, and searches whether all the predecessor knowledge points of the successor knowledge point have been mastered. If not, the unmastered predecessor knowledge points of the successor knowledge point are pushed; if mastered, the learning resources corresponding to the successor knowledge point are pushed.

[0097] Possibly, the embodiment of the present application can store the prediction training model in a second database (e.g., MongoDB) and / or an external storage device when the accuracy of the prediction training model reaches the accuracy threshold. The model on the external storage device is mainly used for service restart and recovery.

[0098] See also Figure 7 In a specific example, the embodiment of the present application can store the trained prediction training system and the prediction results output by the prediction training system in the second database and the external storage device. Furthermore, the business system can push the corresponding teaching information to the user by reading the prediction results in the second database and the pre-set teaching push strategy.

[0099] Figure 8 1 is a schematic diagram of the structure of an information push device provided by an exemplary embodiment of the present application. The information push device can be set in a device such as a server to execute the information push method of any of the above embodiments of the present application. Figure 8 As shown, the information push device may include:

[0100] The acquisition module 81 is used to acquire the teaching push strategy and the user's change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data;

[0101] An obtaining module 82 is used to pre-process the user's changed data to obtain the user's training data;

[0102] The output module 83 is used to input the training data of the user into the prediction training model and output the prediction result of the user; wherein the prediction training model is generated by training based on the sample change data after preprocessing, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample job change data;

[0103] The push module 84 is used to push teaching information to the user based on the prediction result of the user and the teaching push strategy.

[0104] In the embodiment of the present application, the teaching push strategy and the user's change data can be obtained; the user's change data can be preprocessed to obtain the user's training data; the user's training data can be input into the prediction training model to output the user's prediction result; and the teaching information can be pushed to the user based on the user's prediction result and the teaching push strategy. Thus, the embodiment of the present application can adjust the teaching progress for the user in real time according to the user's learning status and the pre-formulated teaching push strategy, which not only solves the problems of delay in the prior art and improves the user experience, but also can formulate corresponding learning plans for different teaching programs, thereby effectively improving the user's enthusiasm for learning.

[0105] In some implementations, before acquiring module 81, the device further includes:

[0106] A current data acquisition module, used to acquire the current data of the user; wherein the current data includes any one or more of the following: teaching video data, teaching content data, and teaching homework data;

[0107] The changed data obtaining module is used to obtain the changed data of the user when it is detected that the current data of the user is changed.

[0108] In some embodiments, the user's training data includes first training data and second training data;

[0109] The obtaining module 82 comprises:

[0110] An output unit, configured to obtain the user's change data through a message middleware and output a plurality of first data streams;

[0111] A first obtaining unit, configured to obtain the first training data based on the multiple first data streams;

[0112] An acquisition unit, used for acquiring the historical data of the user;

[0113] a second obtaining unit, configured to verify the format information of the first training data based on the format information of the user's historical data to obtain the second training data;

[0114] The output module 83 is specifically used to: input the second training data into the prediction training model to output the prediction result of the user. In some embodiments, the first obtaining unit is specifically used to:

[0115] Cleaning and / or transforming the data in each first data stream to obtain a plurality of second data streams;

[0116] Extracting data from the plurality of second data streams based on a preset extraction rule to obtain first training data. In some implementations, the device further includes: a first storage module, configured to store the second training data in a first database.

[0117] In some implementations, before the acquisition module 81, the following further includes:

[0118] The sample change data acquisition module is used to acquire the teaching push strategy and the user's sample change data; wherein the sample change data includes any one or more of the following: sample video change data, sample content change data, sample homework change data;

[0119] A sample training data obtaining module is used to pre-process the sample change data to obtain sample training data;

[0120] The generation module is used to create an initial prediction training model, train the initial prediction training model based on the sample training data, and generate the prediction training model.

[0121] In some embodiments, after the push module 84, the device further includes:

[0122] A teaching information change data module, used for obtaining change data corresponding to the teaching information based on the teaching information of the user;

[0123] A determination module, used to determine the accuracy of the prediction training model based on the change data corresponding to the teaching information;

[0124] An adjustment module, configured to adjust the prediction training model based on the change data corresponding to the teaching information when the accuracy of the prediction training model does not reach the accuracy threshold;

[0125] The prediction training model module is used to complete the training when the accuracy of the prediction training model reaches an accuracy threshold.

[0126] In some embodiments, the device further includes: a second storage module, configured to store the prediction training model in a second database and / or an external storage device when the change data corresponding to the teaching information satisfies a prediction threshold range.

[0127] It should be noted that the information push device provided in the above embodiment only uses the division of the above functional modules as an example when executing the information push method. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the information push device provided in the above embodiment and the information push method embodiment belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.

[0128] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0129] See also Fig. 9 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device 90 may include: at least one processor 91 , at least one network interface 94 , a user interface 93 , a memory 95 , and at least one communication bus 92 .

[0130] The communication bus 92 is used to realize the connection and communication between these components.

[0131] The user interface 93 may include a display screen (Display) and a camera (Camera), and the optional user interface 93 may also include a standard wired interface and a wireless interface.

[0132] The network interface 94 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0133] Among them, the processor 91 may include one or more processing cores. The processor 91 uses various interfaces and lines to connect various parts of the entire electronic device 90, and executes various functions and processes data of the electronic device 90 by running or executing instructions, programs, code sets or instruction sets stored in the memory 95, and calling data stored in the memory 95. Optionally, the processor 91 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 91 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 91, and it can be implemented separately through a chip.

[0134] Among them, the memory 95 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 95 includes a non-transitory computer-readable storage medium. The memory 95 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 95 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 95 may also be optionally at least one storage device located away from the aforementioned processor 91. As Fig. 9As shown, the memory 95 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an information push application program.

[0135] exist Fig. 9 In the electronic device 90 shown, the user interface 93 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 91 can be used to call the information push application stored in the memory 95 and specifically perform the following operations:

[0136] Obtaining teaching push strategy and user change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data;

[0137] Preprocessing the changed data of the user to obtain training data of the user;

[0138] Input the training data of the user into the prediction training model, and output the prediction result of the user; wherein the prediction training model is generated based on the sample change data after preprocessing and training, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample operation change data;

[0139] Teaching information is pushed to the user based on the prediction result of the user and the teaching push strategy.

[0140] In some implementations, before executing the acquisition of the teaching push strategy and the user's change data, the processor 91 further executes:

[0141] Acquire the current data of the user; wherein the current data includes any one or more of the following: teaching video data, teaching content data, and teaching homework data;

[0142] When it is detected that the current data of the user is changed, the changed data of the user is obtained.

[0143] In some embodiments, the user's training data includes first training data and second training data;

[0144] When the processor 91 performs the preprocessing of the user's changed data to obtain the user's training data, the processor 91 specifically performs:

[0145] Obtaining the user's changed data through the message middleware, and outputting a plurality of first data streams;

[0146] Obtaining the first training data based on the multiple first data streams;

[0147] Obtaining historical data of the user;

[0148] Verifying the format information of the first training data based on the format information of the user's historical data to obtain the second training data;

[0149] The step of inputting the training data of the user into a prediction training model and outputting the prediction result of the user includes: inputting the second training data into a prediction training model and outputting the prediction result of the user.

[0150] In some implementations, when the processor 91 executes the step of obtaining the first training data based on the multiple first data streams, the processor 91 specifically executes:

[0151] Cleaning and / or transforming the data in each first data stream to obtain a plurality of second data streams;

[0152] Data is extracted from the plurality of second data streams based on a preset extraction rule to obtain first training data.

[0153] In some implementations, the processor 91 further executes storing the second training data into the first database.

[0154] In some implementations, before executing the step of obtaining the teaching push strategy and the user's change data, the processor 91 further executes:

[0155] Obtaining teaching push strategies and user sample change data; wherein the sample change data includes any one or more of the following: sample video change data, sample content change data, sample homework change data;

[0156] Preprocessing the sample change data to obtain sample training data;

[0157] An initial prediction training model is created, and the initial prediction training model is trained based on the sample training data to generate the prediction training model.

[0158] In some implementations, after executing the pushing of teaching information to the user based on the prediction result of the user and the teaching push strategy, the processor 91 further executes:

[0159] Based on the teaching information of the user, obtaining change data corresponding to the teaching information;

[0160] Determining the accuracy of the prediction training model based on the change data corresponding to the teaching information;

[0161] When the accuracy of the prediction training model does not reach the accuracy threshold, adjusting the prediction training model based on the change data corresponding to the teaching information;

[0162] When the accuracy of the prediction training model reaches the accuracy threshold, the training is completed.

[0163] In some embodiments, the processor 91 further executes: when the change data corresponding to the teaching information satisfies a prediction threshold range, storing the prediction training model in a second database and / or an external storage device.

[0164] The embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a computer or a processor, the computer or the processor executes the above-mentioned Figure 3-Figure 4 ,as well as Figure 6 One or more steps in the illustrated embodiment. If the components of the above-mentioned information push device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0165] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted by the computer-readable storage medium. The computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server, a data center, etc. that contains one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0166] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The aforementioned storage medium includes: various media that can store program codes, such as system memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk. In the absence of conflict, the technical features in this embodiment and the implementation scheme can be combined arbitrarily.

[0167] The embodiments described above are merely preferred embodiments of the present application and are not intended to limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements made to the technical solutions of the present application by ordinary technicians in this field should fall within the protection scope determined by the claims of the present application.

Claims

1. An information push method, characterized in that: The method comprises: Acquire the teaching push strategy and the user's change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data; the change data indicates data that has changed in the scene; the teaching push strategy includes a strategy for enabling the user to complete the pushed teaching content and a strategy for discovering the user's knowledge weaknesses; Obtaining the user's changed data through the message middleware, and outputting a plurality of first data streams; Cleaning and / or transforming the data in each first data stream to obtain a plurality of second data streams; Extracting data from the plurality of second data streams based on a preset extraction rule to obtain first training data; wherein the preset extraction rule is a preset extraction of target change data from the plurality of second data streams; Obtaining historical data of the user; Verifying the format information of the first training data based on the format information of the user's historical data, and adjusting the order of the first training data to obtain second training data; Input the second training data into a prediction training model, and output a prediction result of the user; wherein the prediction training model is generated based on sample change data after preprocessing and training, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample homework change data; the prediction result is used to indicate the user's ability to complete a certain course content; Teaching information is pushed to the user based on the prediction result of the user and the teaching push strategy.

2. The method according to claim 1, characterized in that Before obtaining the teaching push strategy and the user's change data, the method further includes: Acquire the current data of the user; wherein the current data includes any one or more of the following: teaching video data, teaching content data, and teaching homework data; When it is detected that the current data of the user is changed, the changed data of the user is obtained.

3. The method according to claim 1, characterized in that The method further includes storing the second training data in a first database.

4. The method according to claim 1, characterized in that Before obtaining the teaching push strategy and the user's change data, the method further includes: Obtaining teaching push strategies and user sample change data; wherein the sample change data includes any one or more of the following: sample video change data, sample content change data, sample homework change data; Preprocessing the sample change data to obtain sample training data; An initial prediction training model is created, and the initial prediction training model is trained based on the sample training data to generate the prediction training model.

5. The method according to claim 4, characterized in that The prediction result based on the user After the teaching information is pushed to the user according to the teaching push strategy, the method further includes: Based on the teaching information of the user, obtaining change data corresponding to the teaching information; Determining the accuracy of the prediction training model based on the change data corresponding to the teaching information; When the accuracy of the prediction training model does not reach the accuracy threshold, adjusting the prediction training model based on the change data corresponding to the teaching information; When the accuracy of the prediction training model reaches the accuracy threshold, the training is completed.

6. The method according to claim 4 or 5, characterized in that The method further includes: when the accuracy of the prediction training model reaches an accuracy threshold, storing the prediction training model in a second database and / or an external storage device.

7. An information push device, characterized in that: The device comprises: The acquisition module is used to acquire the teaching push strategy and the user's change data; wherein the change data includes any one or more of the following: teaching video change data, teaching content change data, and teaching homework change data; the change data represents the data that has changed in the scene; the teaching push strategy includes a strategy for enabling the user to complete the pushed teaching content and a strategy for discovering the user's knowledge weaknesses; A obtaining module is used to obtain the user's change data through the message middleware and output multiple first data streams; clean and / or transform the data in each first data stream to obtain multiple second data streams; extract data from the multiple second data streams based on a preset extraction rule to obtain first training data; obtain the user's historical data; verify the format information of the first training data based on the format information of the user's historical data, and adjust the order of the first training data to obtain second training data; wherein the preset extraction rule is a preset extraction of target change data in the multiple second data streams; an output module, configured to input the second training data into a prediction training model and output a prediction result of the user; wherein the prediction training model is generated by training based on sample change data after preprocessing, and the sample change data includes any one or more of the following: sample video change data, sample content change data, and sample homework change data; and the prediction result is used to indicate the user's ability to complete a certain course content; A push module is used to push teaching information to the user based on the prediction result of the user and the teaching push strategy.

8. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 6.

9. An electronic device, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 6.

Images