A teaching content pushing method and system based on multi-source data analysis

Abstract

The present application relates to the technical field of data analysis, in particular to a teaching content pushing method and system based on multi-source data analysis, the method provided by the present application mainly includes the selection frequency of the currently selected target type and the test result after the currently pushed teaching content, and the test result is sent to an evaluation terminal; a first selection model is established to obtain an evaluation index of the current type, a judgment threshold is set based on the average evaluation index, it is judged whether the current evaluation index is greater than the judgment threshold, if not, the target type of the current teaching content is replaced as the default type of the next teaching content pushing. Through the above method, the learner, i.e. the user of the learning terminal, is provided with a more suitable type of teaching content to help improve the learning efficiency of the learner and obtain better results, instead of blindly providing all suitable content for the learner without selection.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a teaching content pushing method and system based on multi-source data analysis. BACKGROUND

[0002] Current teaching content pushing mainly relies on big data analysis, artificial intelligence algorithms and learning behavior modeling technology. The system collects historical behavior data of learners (such as answer records, viewing time, interaction frequency), constructs personalized user portraits combined with knowledge graphs, and evaluates knowledge mastery and cognitive preferences in real time. Based on collaborative filtering and deep neural network recommendation algorithm, dynamically generate content sequence that adapts to individual ability level, learning pace and interest direction, realize accurate pushing. At the same time, with the help of adaptive learning engine, continuously optimize the pushing strategy according to real-time feedback, ensure the pertinence, coherence and progression of teaching content, and effectively improve the learning efficiency.

[0003] The prior art has made a relatively perfect analysis of the data of the current learners themselves, but the prior art does not make targeted differentiation, and generally pushes all suitable content, while teaching content can have different types for display. After pushing all kinds of teaching content, the learner does not know his own learning level, and does not adapt to the appropriate teaching content type, which may reduce the learning efficiency. SUMMARY

[0004] The purpose of the present application is to provide a teaching content pushing method and system based on multi-source data analysis to solve the above-mentioned problems in the prior art.

[0005] The present application is realized by the following technical solutions:

[0006] A teaching content pushing method based on multi-source data analysis, comprising:

[0007] Obtain the basic information of the learning terminal user, match the corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in multiple types for the learning terminal user to select;

[0008] Obtain the test results based on the teaching content input by the learning terminal, feed back the test results to the evaluation terminal, and obtain the scores based on the test results fed back by the evaluation terminal;

[0009] Establish a first selection model, and give an evaluation index through the first selection model based on the scores and the selection frequency of the selected target type;

[0010] Obtain the average evaluation index of the current learning terminal user group, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If so, the next teaching content will be pushed according to the current target type by default.

[0011] If not, the target type of the current teaching content is changed as the default type for the next teaching content push; if the evaluation index of all types of teaching content is not greater than the judgment threshold, then facial recognition is performed on the current user, and the effective learning time of the current user based on each type of teaching content is obtained based on the facial recognition, and the effective learning rate is obtained through the effective learning time and the total learning time.

[0012] A second selection model is established. Based on the effective learning rate and evaluation index of each type of teaching content, a ranking index is output through the second selection model. Based on the ranking index, one type of teaching content is selected as the default type to be sent next time, and an alarm signal is sent to the evaluation terminal.

[0013] Preferably, the matching of corresponding teaching content based on basic information located in the teaching content database includes:

[0014] Set the initial push frequency and push time of teaching content. When the push execution signal is received, obtain the basic information of the user of the learning terminal that is currently being pushed.

[0015] Establish a teaching content database, which includes content modules at different levels based on difficulty.

[0016] The difficulty level of the content sections is used as an index entry. The user's basic information is used to evaluate the current user's matching level. The teaching content database is indexed based on the matching level to obtain the corresponding target content section, and the target content section is pushed to the user.

[0017] Preferably, the process of evaluating the current user's basic information to obtain the current user's matching level includes:

[0018] Based on the aforementioned basic information, the initial matching level of the current learning terminal user and the learning score of the content section of the current initial matching level are obtained, and an evaluation model is established.

[0019]

[0020] when Less than At that time, based on the level difference, set from arrive The weight ratio of all content modules is determined, and the random acquisition amount from the content modules of the initial matching level to the content modules of the current matching level is determined based on the weight ratio of the acquisition amount. All the random acquisition amounts are then combined into the teaching content.

[0021] when Greater than When, then send and The corresponding content modules serve as the teaching content.

[0022] when equal When, then send or The corresponding content modules serve as the teaching content.

[0023] In the formula, The current matching level. The learning score for the content section at the current initial matching level. The current initial matching level content section passing score, This represents the current initial matching level.

[0024] Preferably, the step of combining the random acquisition amounts from the content sections at the initial matching level to the content sections at the current matching level based on the acquisition weight ratio into the teaching content includes:

[0025]

[0026] In the formula, The total data size of the teaching content. For the first The content sections at the top level acquire a higher weighting percentage. For the first The data size of the content sections at each level.

[0027] Preferably, sending the test results to the evaluation terminal further includes encrypting the sent test results, including:

[0028] The test results are divided into several different data packets, each data packet is encrypted, and the encryption order of each data packet is recorded.

[0029] Several encrypted data packets are sent randomly, and the encryption order is sent through a terminal that is different from the terminal that sent the test results.

[0030] If the decryption order of the data packet received by the terminal is consistent with the encryption order, the decryption is successful; otherwise, an alarm signal is sent.

[0031] Preferably, encrypting each data packet includes:

[0032] Generate the first-level master key bit of the AES algorithm using a coin-operated method. .

[0033] Calling the AES algorithm's Constructor function yields... The array is used to obtain N sets of round keys, i.e., subkey vectors, through key expansion. .

[0034] use Encrypt sequentially In each round of AES encryption, byte substitution, row shifting, column mixing, and round key addition operations are performed, resulting in a total of 100 working keys. , as a shared key.

[0035] use Encrypt data packets to complete the encryption process for data packets.

[0036] Preferably, establishing the first selection model includes:

[0037] The scores and selection frequency are used to provide an evaluation index through a type selection model;

[0038]

[0039]

[0040] In the formula, For the first The evaluation index of the o-th type of teaching content for each user. For the first User ratings For the selection frequency of the current type, Total number of times the user's type was selected. To determine the threshold, The coefficient is used for calculation, and its value ranges from 0.4 to 0.5. This represents the total number of groups to which the user belongs.

[0041] Preferably, the step of obtaining the current user's effective learning time for each type of teaching content based on facial recognition includes:

[0042] Obtain the execution signal used to start the teaching content, record the current first time point, and obtain the current learning screen data;

[0043] Face recognition is performed on the current learning screen data. When face data is recognized and matches the face data of the current learning terminal user, the learning time node is marked. When no face data is detected or the face data does not match the face data of the previous learning terminal user, the learning end time node is marked. The temporary effective learning time is obtained through the learning time node and the learning end time node.

[0044] Obtain the execution signal used to end the teaching content, record the current second time node, and obtain the total effective learning time through several segments of temporary effective learning time. The total learning time is obtained based on the first and second time nodes.

[0045] Preferably, establishing the second selection model includes:

[0046]

[0047] In the formula, For the first Ranking index of each type of teaching content For the current type of the first The duration of a temporary effective learning time. As the second time point, As the first point in time, This represents the total number of temporary effective study times.

[0048] Secondly, the present invention also provides a teaching content push system based on multi-source data analysis, used to execute the above-mentioned teaching content push method based on multi-source data analysis, including:

[0049] The matching module is configured to obtain basic information of the learning terminal user, match corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in various types for the learning terminal user to choose from; obtain test results based on the teaching content input by the learning terminal, feed the test results back to the evaluation terminal, and obtain the score based on the test results fed back by the evaluation terminal; establish a first selection model, and give an evaluation index based on the score and the selection frequency of the selected target type through the first selection model;

[0050] The analysis module is configured to obtain the average evaluation index of the user group of the current learning terminal, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If it is, the next teaching content will be pushed with the current target type by default; if not, the target type of the current teaching content will be changed as the default type for the next teaching content push. If the evaluation index of all types of teaching content is not greater than the judgment threshold, facial recognition will be performed on the current user, and the effective learning time of the current user for each type of teaching content will be obtained based on the facial recognition. The effective learning rate will be obtained from the effective learning time and the total learning time. A second selection model will be established, and a ranking index will be output through the second selection model based on the effective learning rate and evaluation index of each type of teaching content. Based on the ranking index, one type of teaching content will be selected as the default type for the next push, and an alarm signal will be sent to the evaluation terminal.

[0051] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0052] The method provided by this invention mainly includes: the selection frequency of the currently selected target type and the test results after the currently pushed teaching content; sending the test results to the evaluation terminal; establishing a first selection model to obtain the evaluation index of the current type; setting a judgment threshold based on the average evaluation index; judging whether the current evaluation index is greater than the judgment threshold; if not, changing the target type of the current teaching content as the default type for the next teaching content push. Through this method, learners, i.e., users of the learning terminal, are provided with more suitable types of teaching content, thereby helping to improve their learning efficiency and achieve better results, instead of blindly providing learners with all suitable content without any choice. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a schematic diagram of the process of the present invention;

[0055] Figure 2 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0057] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. The naming or numbering of steps in this application does not imply that the steps in the method flow must be executed in the chronological / logical order indicated by the naming or numbering. The execution order of named or numbered process steps can be changed according to the desired technical objective, as long as the same or similar technical effect is achieved.

[0058] The independently described modules or sub-modules may or may not be physically separated; they may be implemented in software or hardware, and some modules or sub-modules may be implemented in software, with the processor calling the software to implement the function of these modules or sub-modules, while other modules or sub-modules may be implemented in hardware, such as through hardware circuits. Furthermore, some or all of the modules can be selected to achieve the purpose of this application's solution according to actual needs.

[0059] Please refer to Figure 1 This invention provides a method for pushing teaching content based on multi-source data analysis, including:

[0060] S101: Obtain the basic information of the learning terminal user, match the corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in various types for the learning terminal user to choose from.

[0061] In this invention, the teaching content can be presented in various formats, such as videos, interactive quizzes, situational cases, and micro-lecture PPTs. When sending the teaching content, it will be sent to the teaching terminal user in various formats for selection.

[0062] S102: Obtain the test results based on the teaching content input from the learning terminal, feed the test results back to the evaluation terminal, and obtain the score based on the test results fed back by the evaluation terminal;

[0063] The test results primarily reflect the effectiveness of current users using the current type of teaching content, and are used for adjustments to subsequent types.

[0064] S103: Establish a first choice model, and give an evaluation index based on the score and the selection frequency of the selected target type through the first choice model;

[0065] In this embodiment, the learning terminal can be the student's terminal, and the evaluation terminal can be the teacher's terminal. After analyzing the student's test results, the teacher inputs score data into the evaluation terminal as one of the criteria for judging the effectiveness of the current type of use.

[0066] S104: Obtain the average evaluation index of the group to which the current learning terminal user belongs, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If so, the next teaching content will be pushed according to the current target type by default.

[0067] By comparing the threshold and the pre-evaluation index, the learning effect of the current user on the current type of teaching content is determined. If it is good, the current type is kept as the default for push. In this way, it is not necessary to download all other types of teaching content during the transmission process, reducing transmission time. Other types are then obtained when the user selects them. If it is not satisfactory, the type is changed.

[0068] S105: If not, change the target type of the current teaching content as the default type for the next teaching content push. If the evaluation index of all types of teaching content is not greater than the judgment threshold, perform facial recognition on the current user, obtain the effective learning time of the current user based on each type of teaching content based on facial recognition, and obtain the effective learning rate through the effective learning time and total learning time.

[0069] In order to consider whether various types of teaching content are suitable for the current user, this embodiment also incorporates an effective learning rate to reflect the current user's focus on the current type of teaching content.

[0070] S106: Establish a second selection model. Based on the effective learning rate and evaluation index of each type of teaching content, output a ranking index through the second selection model. Based on the ranking index, select one type of teaching content as the default type to be sent next time, and send an alarm signal to the evaluation terminal.

[0071] The second selection model yields a ranking index that can quantify the learning effectiveness of each type for the current user. By comparison, one type is selected as the default type for the next delivery. Since all types of teaching content are unsatisfactory for the current user, feedback can be given to the evaluation terminal to determine whether manual intervention is necessary.

[0072] The method provided by this invention mainly includes: the selection frequency of the currently selected target type and the test results after the currently pushed teaching content; sending the test results to the evaluation terminal; establishing a first selection model to obtain the evaluation index of the current type; setting a judgment threshold based on the average evaluation index; judging whether the current evaluation index is greater than the judgment threshold; if not, changing the target type of the current teaching content as the default type for the next teaching content push. Through this method, learners, i.e., users of the learning terminal, are provided with more suitable types of teaching content, thereby helping to improve their learning efficiency and achieve better results, instead of blindly providing learners with all suitable content without any choice.

[0073] In one exemplary embodiment of the present invention, the matching of corresponding teaching content based on basic information located in the teaching content database includes:

[0074] S201: Set the initial push frequency and push time of teaching content. When the push execution signal is obtained, obtain the basic information of the user of the learning terminal currently being pushed.

[0075] S202: Establish a teaching content database, which includes content modules of different levels based on difficulty.

[0076] S203: Using the difficulty level of the content section as an index entry, the current user's basic information is used to evaluate and obtain the current user's matching level. Based on the matching level, the teaching content database is indexed to obtain the corresponding target content section, and the target content section is pushed.

[0077] In this embodiment, the database of the current teaching content is divided into different levels according to the difficulty level, such as Level 1, Level 2, Level 3, etc. The higher the level number, the greater the difficulty.

[0078] Specifically, the evaluation based on the current learning terminal user's basic information to obtain the current user's matching level includes:

[0079] Based on the aforementioned basic information, the initial matching level of the current learning terminal user and the learning score of the content section of the current initial matching level are obtained, and an evaluation model is established.

[0080]

[0081] when Less than At that time, based on the level difference, set from arrive The weight ratio of all content modules is determined, and the random acquisition amount from the content modules of the initial matching level to the content modules of the current matching level is determined based on the weight ratio of the acquisition amount. All the random acquisition amounts are then combined into the teaching content.

[0082] when Greater than When, then send and The corresponding content modules serve as the teaching content.

[0083] when equal When, then send or The corresponding content modules serve as the teaching content.

[0084] In the formula, The current matching level. The learning score for the content section at the current initial matching level. The current initial matching level content section passing score, This represents the current initial matching level.

[0085] Among these features, the system evaluates the user's current learning score and passing score for the content section at their current level, and determines whether the difficulty of the next teaching content can be increased or decreased to better suit the current user and make the use of the teaching content more flexible.

[0086] Specifically, the method of combining the random acquisition amounts from the content sections at the initial matching level to the content sections at the current matching level, based on the acquisition weight ratio, into the teaching content includes:

[0087]

[0088] In the formula, The total data size of the teaching content. For the first The content sections at the top level acquire a higher weighting percentage. For the first The data size of the content sections at each level.

[0089] For example, if the initial user uses a second-level content section, and the current matching level is a fourth-level content section, then when i=2, When i=3, When i=4, To obtain a weighted average, simply ensure that the total weight percentage equals 1.

[0090] In one exemplary embodiment of the present invention, to better protect user privacy, the content sent by the user is encrypted. Sending the test results to the evaluation terminal further includes encrypting the sent test results, including:

[0091] The test results are divided into several different data packets, each data packet is encrypted, and the encryption order of each data packet is recorded.

[0092] Several encrypted data packets are sent randomly, and the encryption order is sent through a terminal that is different from the terminal that sent the test results.

[0093] If the decryption order of the data packet received by the terminal is consistent with the encryption order, the decryption is successful; otherwise, an alarm signal is sent.

[0094] Specifically, encrypting each data packet includes:

[0095] Generate the first-level master key bit of the AES algorithm using a coin-operated method. ;

[0096] Calling the AES algorithm's Constructor function yields... The array is used to obtain N sets of round keys, i.e., subkey vectors, through key expansion. , It consists of several subkeys;

[0097] use Encrypt sequentially In each round of AES encryption, byte substitution, row shifting, column mixing, and round key addition operations are performed, resulting in a total of 100 working keys. , as a shared key;

[0098] use Encrypt data packets to complete the encryption process for data packets.

[0099] In one exemplary embodiment of the present invention, establishing the first selection model includes:

[0100] The scores and selection frequency are used to provide an evaluation index through a type selection model;

[0101]

[0102]

[0103] In the formula, For the first The evaluation index of the o-th type of teaching content for each user. For the first User ratings For the selection frequency of the current type, Total number of times the user's type was selected. To determine the threshold, The coefficient is used for calculation, and its value ranges from 0.4 to 0.5. This represents the total number of groups to which the user belongs.

[0104] In this invention, selection frequency and rating are mainly used to evaluate the learning effect of the current user, and the group to which the current user belongs is used as the rating standard to obtain a relatively objective evaluation result, which serves as a basis for whether to change the type in the future.

[0105] An exemplary embodiment of the present invention, which obtains the effective learning time of the current user for each type of teaching content based on facial recognition, includes:

[0106] Obtain the execution signal used to start the teaching content, record the current first time point, and obtain the current learning screen data;

[0107] Face recognition is performed on the current learning screen data. When face data is recognized and matches the face data of the current learning terminal user, the learning time node is marked. When no face data is detected or the face data does not match the face data of the previous learning terminal user, the learning end time node is marked. The temporary effective learning time is obtained through the learning time node and the learning end time node.

[0108] Obtain the execution signal used to end the teaching content, record the current second time node, and obtain the total effective learning time through several segments of temporary effective learning time. The total learning time is obtained based on the first and second time nodes.

[0109] In one exemplary embodiment of the present invention, establishing the second selection model includes:

[0110]

[0111] In the formula, For the first Ranking index of each type of teaching content For the current type of the first The duration of a temporary effective learning time. As the second time point, As the first point in time, This represents the total number of temporary effective study times.

[0112] The above model evaluates each type except the one being replaced, and selects the type with the highest ranking index as the default type for the next push. This improves transmission efficiency and provides users with a relatively objective choice.

[0113] Please refer to Figure 2 A teaching content delivery system based on multi-source data analysis, used to execute the aforementioned teaching content delivery method based on multi-source data analysis, includes:

[0114] The matching module is configured to obtain basic information of the learning terminal user, match corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in various types for the learning terminal user to choose from; obtain test results based on the teaching content input by the learning terminal, feed the test results back to the evaluation terminal, and obtain the score based on the test results fed back by the evaluation terminal; establish a first selection model, and give an evaluation index based on the score and the selection frequency of the selected target type through the first selection model;

[0115] The analysis module is configured to obtain the average evaluation index of the user group of the current learning terminal, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If it is, the next teaching content will be pushed with the current target type by default; if not, the target type of the current teaching content will be changed as the default type for the next teaching content push. If the evaluation index of all types of teaching content is not greater than the judgment threshold, facial recognition will be performed on the current user, and the effective learning time of the current user for each type of teaching content will be obtained based on the facial recognition. The effective learning rate will be obtained from the effective learning time and the total learning time. A second selection model will be established, and a ranking index will be output through the second selection model based on the effective learning rate and evaluation index of each type of teaching content. Based on the ranking index, one type of teaching content will be selected as the default type for the next push, and an alarm signal will be sent to the evaluation terminal.

[0116] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for pushing teaching content based on multi-source data analysis, characterized in that, include: Obtain basic information about the learning terminal user, match the corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in various types for the learning terminal user to choose from. Obtain test results based on teaching content input from the learning terminal, feed the test results back to the evaluation terminal, and obtain the score based on the test results fed back by the evaluation terminal; A first selection model is established, and an evaluation index is output based on the scores and the selection frequency of the selected target types, including: ; In the formula, For the first The evaluation index of the o-th type of teaching content for each user. For the first The score of each user, For the selection frequency of the current type, Total number of type selections for the current user. To determine the threshold, The coefficient is used for calculation, and its value ranges from 0.4 to 0.

5. The total number of user groups; Obtain the average evaluation index of the current learning terminal user group, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If so, the next teaching content will be pushed according to the current target type by default. If not, the target type of the current teaching content is changed as the default type for the next teaching content push; if the evaluation index of all types of teaching content is not greater than the judgment threshold, then facial recognition is performed on the current user, and the effective learning time of the current user based on each type of teaching content is obtained based on the facial recognition, and the effective learning rate is obtained through the effective learning time and the total learning time. A second choice model is established, which outputs a ranking index based on the effective learning rate and evaluation index of each type of teaching content. This index includes: In the formula, For the first o Ranking index of each type of teaching content For the current type of the first The duration of a temporary effective learning time. As the second time point, As the first point in time, To determine the total number of temporary effective learning times, select one type of teaching content based on the ranking index as the default type to be sent next time, and send an alarm signal to the evaluation terminal. Based on the aforementioned basic information, the initial matching level of the current learning terminal user and the learning score of the content section of the current initial matching level are obtained, and an evaluation model is established. when Less than At that time, based on the level difference, set from arrive The weighting percentage of all content sections is used to determine the random acquisition amount from the initial matching level to the current matching level of the content sections. Based on this weighting percentage, all randomly acquired amounts are combined to form the teaching content, including: In the formula, The total data size of the teaching content. For the first The content sections at the top level acquire a higher weighting percentage. For the first The data size of the content sections at each level; when Greater than When, then send and The corresponding content modules serve as the teaching content. when equal When, then send or The corresponding content modules serve as the teaching content. In the formula, The current matching level. The learning score for the content section at the current initial matching level. The current initial matching level content section passing score, This represents the current initial matching level.

2. The teaching content delivery method based on multi-source data analysis according to claim 1, characterized in that, The matching of corresponding teaching content based on basic information located in the teaching content database includes: Set the initial push frequency and push time of teaching content. When the push execution signal is received, obtain the basic information of the user of the learning terminal that is currently being pushed. Establish a teaching content database, which includes content modules at different levels based on difficulty. The difficulty level of the content sections is used as an index entry. The user's basic information is used to evaluate the current user's matching level. The teaching content database is indexed based on the matching level to obtain the corresponding target content section, and the target content section is pushed to the user.

3. The teaching content delivery method based on multi-source data analysis according to claim 1, characterized in that, Sending the test results to the evaluation terminal also includes encrypting the sent test results, including: The test results are divided into several different data packets, each data packet is encrypted, and the encryption order of each data packet is recorded. Several encrypted data packets are sent randomly, and the encryption order is sent through a terminal that is different from the terminal that sent the test results. If the decryption order of the data packet received by the terminal is consistent with the encryption order, the decryption is successful; otherwise, an alarm signal is sent.

4. The teaching content delivery method based on multi-source data analysis according to claim 3, characterized in that, The encryption of each data packet includes: Generate the first-level master key bit of the AES algorithm using a coin-operated method. ; Calling the AES algorithm's Constructor function yields... The array is used to obtain N sets of round keys, i.e., subkey vectors, through key expansion. ; use Encrypt sequentially In each round of AES encryption, byte substitution, row shifting, column mixing, and round key addition operations are performed, resulting in a total of 100 working keys. , as a shared key; use Encrypt data packets to complete the encryption process for data packets.

5. The teaching content delivery method based on multi-source data analysis according to claim 4, characterized in that, The method for obtaining the effective learning time of the current user for each type of teaching content based on facial recognition includes: Obtain the execution signal used to start the teaching content, record the current first time point, and obtain the current learning screen data; Face recognition is performed on the current learning screen data. When face data is recognized and matches the face data of the current learning terminal user, the learning time node is marked. When no face data is detected or the face data does not match the face data of the previous learning terminal user, the learning end time node is marked. The temporary effective learning time is obtained through the learning time node and the learning end time node. Obtain the execution signal used to end the teaching content, record the current second time node, and obtain the total effective learning time through several segments of temporary effective learning time. The total learning time is obtained based on the first and second time nodes.

6. A teaching content delivery system based on multi-source data analysis, characterized in that, A method for pushing teaching content based on multi-source data analysis as described in claim 1, comprising: The matching module is configured to obtain basic information of the learning terminal user, match corresponding teaching content in the teaching content database based on the basic information, and send the teaching content in various types for the learning terminal user to choose from; obtain test results based on the teaching content input by the learning terminal, feed the test results back to the evaluation terminal, and obtain the score based on the test results fed back by the evaluation terminal; establish a first selection model, and give an evaluation index based on the score and the selection frequency of the selected target type through the first selection model; The analysis module is configured to obtain the average evaluation index of the user group of the current learning terminal, set a judgment threshold based on the average evaluation index, and determine whether the current evaluation index is greater than the judgment threshold. If it is, the next teaching content will be pushed with the current target type by default; if not, the target type of the current teaching content will be changed as the default type for the next teaching content push. If the evaluation index of all types of teaching content is not greater than the judgment threshold, facial recognition will be performed on the current user, and the effective learning time of the current user for each type of teaching content will be obtained based on the facial recognition. The effective learning rate will be obtained from the effective learning time and the total learning time. A second selection model will be established, and a ranking index will be output through the second selection model based on the effective learning rate and evaluation index of each type of teaching content. Based on the ranking index, one type of teaching content will be selected as the default type for the next push, and an alarm signal will be sent to the evaluation terminal.

Images