CN111914176A

CN111914176A - Method and device for recommending subjects

Info

Publication number: CN111914176A
Application number: CN202010792916.0A
Authority: CN
Inventors: 李习华; 李俊宁; 赵学敏; 曹云波
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-08-07
Filing date: 2020-08-07
Publication date: 2020-11-10
Anticipated expiration: 2040-08-07
Also published as: CN111914176B

Abstract

The embodiment of the application provides a title recommendation method and device. The title recommendation method comprises the following steps: obtaining a question to be pushed and the corresponding accuracy of the question to be pushed; acquiring a current answer behavior record of the target object, and determining the type information of the answer of the target object and answer results according to the current answer behavior record, wherein the answer results comprise correct answers and wrong answers; updating the accuracy of the questions to be pushed corresponding to the category information in the questions to be pushed according to the category information and the answer result; and determining a target topic in the topics to be pushed according to the updated accuracy of the topics to be pushed so as to push the target topic to the target object. According to the technical scheme, the questions can be pushed according to the real learning level of the target object, and therefore the question making effect of the target object is guaranteed.

Description

Method and device for recommending subjects

Technical Field

The application relates to the technical field of computers, in particular to a title recommendation method and device.

Background

With the research and progress of artificial intelligence technology, artificial intelligence technology has been developed and applied in various fields. In the education field, internet online education is widely used. The user can carry out online learning, online question making, online examination and the like through the online education learning system. After the user learns online, the learning system provides the user with exercise questions related to the learning content. Therefore, how to provide corresponding exercise questions according to the real learning level of the user and further ensure the question making effect of the user becomes a technical problem to be solved urgently.

Disclosure of Invention

The embodiment of the application provides a question recommending method and device, so that corresponding exercise questions can be provided at least to a certain extent according to the real learning level of a user, and the question making effect of the user is further ensured.

Other features and advantages of the present application will be apparent from the following detailed description, or may be learned by practice of the application.

According to an aspect of an embodiment of the present application, there is provided a method for recommending a topic, the method including:

acquiring a question to be pushed and the accuracy corresponding to the question to be pushed, wherein the accuracy is used for describing the possibility of answering the question to be pushed by a target object;

acquiring a current answer behavior record of the target object, and determining the type information of the answer of the target object and answer results according to the current answer behavior record, wherein the answer results comprise correct answers and wrong answers;

updating the accuracy of the questions to be pushed corresponding to the category information in the questions to be pushed according to the category information and the answer result;

and determining a target topic in the topics to be pushed according to the updated accuracy of the topics to be pushed so as to push the target topic to the target object.

According to an aspect of an embodiment of the present application, there is provided an apparatus for recommending a topic, the apparatus including:

the device comprises a first obtaining module, a second obtaining module and a third obtaining module, wherein the first obtaining module is used for obtaining a question to be pushed and a correct rate corresponding to the question to be pushed, and the correct rate is used for describing the possibility of a target object answering the question to be pushed;

the second acquisition module is used for acquiring the current answer behavior record of the target object and determining the type information of the answer of the target object and answer results according to the current answer behavior record, wherein the answer results comprise correct answers and wrong answers;

the updating module is used for updating the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question according to the category information and the answer result;

and the processing module is used for determining a target topic in the topics to be pushed according to the updated accuracy of the topics to be pushed so as to push the target topic to the target object.

In some embodiments of the present application, based on the foregoing, the first obtaining module is configured to: acquiring a plurality of to-be-processed titles and category information and attribute information corresponding to the plurality of to-be-processed titles; dividing according to the category information of the plurality of to-be-processed questions to obtain to-be-processed question sets corresponding to the category information; determining the quality score of each topic to be processed in the topic set to be processed according to the attribute information of each topic to be processed contained in the topic set to be processed; selecting a question to be pushed from the question set to be processed according to the quality score; and predicting the questions to be pushed by adopting a pre-trained prediction model to obtain the corresponding accuracy of the questions to be pushed.

In some embodiments of the present application, based on the foregoing, the first obtaining module is configured to: acquiring weights corresponding to all parameters in the attribute information; and calculating the quality scores of the to-be-processed questions contained in the to-be-processed question set according to the attribute information and the weight of each to-be-processed question contained in the to-be-processed question set.

In some embodiments of the present application, based on the foregoing scheme, before the obtaining of the multiple to-be-processed titles and the category information and the attribute information corresponding to the multiple to-be-processed titles, the first obtaining module is further configured to: displaying a category information editing interface of the to-be-processed question according to the category information editing request of the to-be-processed question; and associating the category information with the corresponding to-be-processed title according to the category information of the to-be-processed title acquired by the category information editing interface.

In some embodiments of the present application, based on the foregoing, the processing module is configured to: dividing according to the category information of the questions to be pushed to obtain a set of the questions to be pushed corresponding to each category information; calculating the mastery degree of the target object to the knowledge points corresponding to the category information according to the accuracy of each to-be-pushed question contained in each to-be-pushed question set; and selecting a target topic from the topics to be pushed according to the mastery degree and a preset threshold value for determining the lower limit of the mastery degree of the target object.

In some embodiments of the present application, based on the foregoing, the processing module is configured to: acquiring a knowledge graph, wherein the knowledge graph comprises priority relations between knowledge points corresponding to all kinds of information; identifying the knowledge points with the mastery degree lower than the preset threshold value as knowledge points to be mastered; identifying the knowledge points with the priority higher than the knowledge points to be mastered as knowledge points to be pushed according to the knowledge graph; and determining the to-be-pushed question corresponding to the to-be-pushed knowledge point as the target question.

In some embodiments of the present application, based on the foregoing, the processing module is further configured to; and calculating the average value of the mastery degrees according to the mastery degrees of the knowledge points corresponding to the various kinds of information, wherein the average value is used as the preset threshold value.

In some embodiments of the present application, based on the foregoing, the update module is configured to: if the answer result is correct, updating the correctness of the question to be pushed corresponding to the category information in the questions to be pushed to be 1; and if the answer result is wrong, updating the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question to be 0.

In some embodiments of the present application, based on the foregoing, the update module is configured to: acquiring a historical answer behavior record of a target object; training a prediction model according to the historical answer behavior record and the current answer behavior record; and predicting the to-be-pushed questions corresponding to the category information in the to-be-pushed questions by adopting the trained prediction model to obtain the accuracy of the to-be-pushed questions corresponding to the category information in the to-be-pushed questions.

According to an aspect of an embodiment of the present application, there is provided a computer-readable medium on which a computer program is stored, the computer program, when executed by a processor, implementing a method of recommending titles as described in the above embodiments.

According to an aspect of an embodiment of the present application, there is provided an electronic device including: one or more processors; a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the method for recommending titles as described in the above embodiments.

According to an aspect of the application, a computer program product or computer program is provided, comprising computer instructions, the computer instructions being stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions to cause the computer device to execute the recommendation method for a topic provided in the above-described embodiment.

In the technical solutions provided in some embodiments of the present application, the question to be pushed and the correctness rate corresponding to the question to be pushed are obtained, the current answer behavior record of the target object is obtained, the category information and the answer result of the question answered by the target object are determined according to the current answer behavior record, the correctness rate of the question to be pushed corresponding to the category information in the question to be pushed is updated according to the category information and the answer result, and the target question in the question to be pushed is determined according to the updated correctness rate of the question to be pushed, so as to push the target question to the target object. Therefore, the accuracy of the corresponding questions to be pushed can be updated in real time according to the current answer behavior record of the user, so that the accuracy of the questions to be pushed can be updated in real time, and the accuracy of the questions to be pushed can correspond to the real learning level of the target object. And selecting and pushing a target question from the questions to be pushed according to the updated correctness of the questions to be pushed, so that the pushed question can correspond to the real learning level of the target object, the problem that the pushed question is too difficult or too simple is avoided, and the question making effect of the target object is improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application. It is obvious that the drawings in the following description are only some embodiments of the application, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:

FIG. 1 shows a schematic diagram of an exemplary system architecture to which aspects of embodiments of the present application may be applied;

FIG. 2 shows a flow diagram of a method of recommending items according to an embodiment of the present application;

FIG. 3 shows a flowchart of step S210 of the method for recommending titles of FIG. 2 according to an embodiment of the present application;

FIG. 4 shows a flowchart of step S330 in the method for recommending titles of FIG. 3 according to an embodiment of the present application;

FIG. 5 is a schematic flowchart illustrating editing of category information of a topic to be processed in a topic recommendation method according to an embodiment of the present application;

FIG. 6 shows a flowchart of step S240 of the method for recommending titles of FIG. 2 according to an embodiment of the present application;

FIG. 7 shows a flowchart of step S630 of the method for recommending titles of FIG. 6 according to an embodiment of the present application;

FIG. 8 shows a flowchart of step S230 of the method for recommending titles of FIG. 2 according to an embodiment of the present application;

FIG. 9 shows a flowchart of a method for recommending titles according to an embodiment of the present application;

FIG. 10 is a schematic flow chart illustrating step S920 of the title recommendation method of FIG. 9 according to an embodiment of the present application;

FIG. 11 shows a block diagram of a title recommendation device according to an embodiment of the present application;

FIG. 12 illustrates a schematic structural diagram of a computer system suitable for use in implementing the electronic device of an embodiment of the present application.

Detailed Description

Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art.

Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the subject matter of the present application can be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known methods, devices, implementations, or operations have not been shown or described in detail to avoid obscuring aspects of the application.

The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. I.e. these functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor means and/or microcontroller means.

The flow charts shown in the drawings are merely illustrative and do not necessarily include all of the contents and operations/steps, nor do they necessarily have to be performed in the order described. For example, some operations/steps may be decomposed, and some operations/steps may be combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.

Fig. 1 shows a schematic diagram of an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied.

As shown in fig. 1, the system architecture may include a terminal device (e.g., one or more of a smartphone 101, a tablet computer 102, and a portable computer 103 shown in fig. 1, but may also be a desktop computer, etc.), a network 104, and a server 105. The network 104 serves as a medium for providing communication links between terminal devices and the server 105. Network 104 may include various connection types, such as wired communication links, wireless communication links, and so forth.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. For example, server 105 may be a server cluster comprised of multiple servers, or the like.

It should be noted that the server 105 may be an independent physical server, may also be a server cluster or a distributed system formed by a plurality of physical servers, and may also be a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network service, cloud communication, a middleware service, a domain name service, a security service, a CDN, and a big data and artificial intelligence platform. The terminal may be, but is not limited to, a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and the like. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the application is not limited herein.

A user may use a terminal device to interact with the server 105 over the network 104 to receive or transmit information or the like. The server 105 may be a server that provides various services. For example, a user uploads a current answer behavior record to the server 105 by using the terminal device 103 (which may also be the terminal device 101 or 102), and the server 105 may obtain a question to be pushed and a correct rate corresponding to the question to be pushed, where the correct rate is used to describe a possibility that a target object answers the question to be pushed; acquiring a current answer behavior record of the target object, and determining the type information of the answer of the target object and answer results according to the current answer behavior record, wherein the answer results comprise correct answers and wrong answers; updating the accuracy of the questions to be pushed corresponding to the category information in the questions to be pushed according to the category information and the answer result; and determining a target topic in the topics to be pushed according to the updated accuracy of the topics to be pushed so as to push the target topic to the target object.

It should be noted that the recommendation method for topics provided in the embodiment of the present application is generally executed by the server 105, and accordingly, the recommendation device for topics is generally disposed in the server 105. However, in other embodiments of the present application, the terminal device may also have a similar function to the server, so as to execute the scheme of the title recommendation method provided in the embodiments of the present application.

The implementation details of the technical solution of the embodiment of the present application are set forth in detail below:

FIG. 2 shows a flowchart illustrating a title recommendation method according to an embodiment of the present application. Referring to fig. 2, the title recommendation method at least includes steps S210 to S240, which are described in detail as follows:

in step S210, a question to be pushed and a correctness rate corresponding to the question to be pushed are obtained, where the correctness rate is used to describe a possibility that a target object answers the question to be pushed.

The to-be-pushed question may be a question for pushing to a target object, and the to-be-pushed question may correspond to the learning content, for example, the to-be-pushed question may be divided into a language question, a math question, an english question, and the like according to the subject.

The accuracy may be a probability for describing the answer of the target object to the item to be pushed, and it should be understood that the accuracy may be a value between 0 and 1, for example, the accuracy may be 10%, 60%, or 90%. Each topic to be pushed can correspond to a correct rate, so as to represent the possibility that the target object answers the topic to be pushed.

In an embodiment of the present application, a server may obtain a topic to be pushed from a local storage space, specifically, a topic database to be pushed may be established in the server, and when a target object answers a question, a corresponding topic to be pushed may be obtained from the topic database to be pushed for subsequent pushing.

When the to-be-pushed question is stored locally, the correctness can be stored in association with the to-be-pushed question, for example, the correctness corresponding to the to-be-pushed question can be obtained by predicting in advance according to the to-be-pushed question, and then a corresponding relation table of the to-be-pushed question and the correctness is established. When the item to be pushed is obtained, the accuracy of the item to be pushed can be correspondingly inquired according to the corresponding relation table, so that the obtaining efficiency of the accuracy is improved. In other embodiments, the topic to be pushed obtained locally may also be predicted in real time, so as to obtain the accuracy corresponding to the topic to be pushed.

In another embodiment of the present application, the server may also obtain the topics to be pushed from a third-party organization through a network, for example, a topic database of an online education platform. Therefore, resources on the network can be gathered to increase the richness of the titles to be pushed. After the questions to be pushed are obtained through the network, the obtained questions to be pushed can be predicted, so that the accuracy of target object answering to the questions to be pushed is obtained.

In step S220, a current answer behavior record of the target object is obtained, and according to the current answer behavior record, category information of an answer of the target object and an answer result are determined, where the answer result includes an answer right and an answer error.

The answer behavior record may be information for recording the answer behavior of the target object, for example, the answer behavior record may include, but is not limited to, time when the target object answers (for example, XX of the target object starts answering separately in XX), time taken by the target object to answer each question to be pushed, terminal equipment (for example, a smart phone, a tablet computer, or a notebook computer, etc.) used when the target object answers the question, identification information of the question to be pushed (for example, a number of the question to be pushed, etc.), category information of the question answered by the target object, and an answer result, and the like. Therefore, the state of the target object in answering can be known according to the answering behavior record of the target object, so that the answering behavior of the target object can be known.

The category information may be feature information used to classify the titles to be pushed, and the titles to be pushed corresponding to each category information may be obtained by dividing different features corresponding to the titles to be pushed, for example, the titles to be pushed may be divided according to the subjects corresponding to the titles to be pushed, the knowledge points corresponding to the titles to be pushed, the question types of the individual titles to be pushed, or the difficulty level of the titles to be pushed.

In an embodiment of the present application, the to-be-pushed topics may be divided according to subjects, knowledge points, question types, and difficulty levels corresponding to the to-be-pushed topics, so as to obtain category information corresponding to each to-be-pushed topic, for example, the category information of the to-be-pushed topic a is: mathematics-trigonometric function-choice question-difficulty, category information of the question B to be pushed is: chinese-ancient poetry-merwriting-simple, etc.

It should be understood that one topic to be pushed can correspond to multiple category information, for example, if the topic a to be pushed is a composite topic which examines multiple knowledge points, the topic a to be pushed can correspond to multiple category information, and so on.

In an embodiment of the application, when a target object starts answering, a current answering behavior record of the target object can be obtained in real time, and category information and an answering result of the question answered by the target object are obtained according to the current answering behavior record, wherein the answering result includes correct answer and wrong answer, so that the grasping condition of the target object on the question to be pushed corresponding to the category information can be known according to the answering result.

In step S230, according to the category information and the answer result, the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question is updated.

In an embodiment of the application, the server may obtain category information of a question answered by the target object from the current answer behavior record, and compare the category information of the answered question with category information of a question to be pushed, so as to obtain the question to be pushed, which is the same as the category information of the answered question. And updating the accuracy of the questions to be pushed, which are the same as the type information of the answered questions, according to the answering results.

It should be understood that if the answer result is correct, the accuracy of the question to be pushed, which is the same as the type information of the question to be answered, should be increased; if the answer result is wrong, the accuracy of the question to be pushed, which is the same as the type information of the answered question, is required to be reduced. Therefore, the grasping condition of the target object to the to-be-pushed title corresponding to each category information can be reflected according to the accuracy.

In step S240, according to the updated accuracy of the to-be-pushed titles, determining a target title in the to-be-pushed titles, so as to push the target title to the target object.

In an embodiment of the application, after the accuracy of the to-be-pushed topic is updated, the mastering condition of the target object on the to-be-pushed topic corresponding to each category information can be known according to the accuracy of the to-be-pushed topic. Therefore, through comparison, the questions to be pushed with poor mastering conditions can be used as the target questions and pushed to the target object, so that the pushed questions can accord with the current learning level of the target object, and the question making effect of the target object is improved.

Based on the embodiment shown in fig. 2, fig. 3 shows a flowchart of step S210 in the title recommendation method of fig. 2 according to an embodiment of the present application. Referring to fig. 3, step S210 at least includes steps S310 to S320, which are described in detail as follows:

in step S310, a plurality of to-be-processed titles, and category information and attribute information corresponding to the plurality of to-be-processed titles are obtained.

Wherein the attribute information may be information for describing a quality of a title, the attribute information being associated with the title. The attribute information may include a plurality of parameters, for example, the attribute information may include, but is not limited to, the year of the topic, whether the topic is real, whether the topic is composite (i.e., whether multiple knowledge points are examined), the degree of distinction from other topics, and the accuracy rate, etc.

In an embodiment of the present application, a topic to be processed may be a topic stored in a local storage space of a server, and the server may obtain the topic to be processed from the local storage space and obtain category information and attribute information corresponding to the topic to be processed. It should be noted that the to-be-processed question may be a question that is not selected, and when the to-be-processed question is uploaded, the question provider may upload the category information and the attribute information of the to-be-processed question at the same time, and store the category information and the attribute information in association with the to-be-processed question for subsequent acquisition.

In step S320, the multiple to-be-processed topics are divided according to the category information to obtain a to-be-processed topic set corresponding to each category information.

In an embodiment of the present application, the obtained category information of a plurality of to-be-processed topics is compared, and to-be-processed topics corresponding to the same category information are determined, for example, the to-be-processed topics corresponding to the category information of "math-trigonometric function-choice question-difficulty" include: A. b, C, etc., the questions to be processed corresponding to the category of "Chinese-ancient poetry-write-simple" are: D. e, F et al, … …. Therefore, the set of to-be-processed topics corresponding to the category information can be obtained.

It should be noted that the "plurality" mentioned in this application may be any number of two or more, such as three, four or five hundred, etc., and the above are merely exemplary, and the present application is not limited to this.

In step S330, the quality score of each topic to be processed in the topic set to be processed is determined according to the attribute information of each topic to be processed included in the topic set to be processed.

Wherein the quality score may be information representing the quality of the topic to be processed. It should be understood that the higher the quality score is, the better the quality of the to-be-processed question is, and accordingly, the better the effect of checking the learning level of the target object by using the quality score is; if the quality score is lower, it indicates that the quality of the topic to be processed is lower, for example, the topic is too simple or the topic is not strict, the effect of checking the learning level of the target object by using the quality score is worse, and the learning level of the target object cannot be reflected.

In an embodiment of the present application, according to attribute information of a to-be-processed topic included in each to-be-processed topic set, a quality score corresponding to each to-be-processed topic in each to-be-processed topic set can be determined. Specifically, the server may display a quality score editing interface in the display device, so as to edit the quality score of each to-be-processed question, and the quality score editing interface may display each to-be-processed question and corresponding attribute information, and a professional may input the quality score according to the attribute information of the to-be-processed question through an input device (e.g., an input keyboard, a touch display screen, or a mouse) configured by the server. After the input is completed, the server can perform associated storage on the quality scores of the to-be-processed titles acquired according to the quality score editing interface and the corresponding to-be-processed titles.

In step S340, selecting a topic to be pushed from the topic set to be processed according to the quality score.

In an embodiment of the present application, according to the quality score of each to-be-processed topic in each to-be-processed set, a to-be-processed topic with a quality score in the front can be selected from each to-be-processed topic set as a to-be-pushed topic. In an example, a predetermined number of to-be-processed topics with quality scores arranged in front of each to-be-processed topic set may be selected as to-be-pushed topics, where the predetermined number may be 600, 800, or 1000, and the above is merely an exemplary example, and this is not limited in this application.

In another example, in the case that the number of to-be-pushed topics in the to-be-pushed topic set is large, a pre-determined proportion of the to-be-pushed topics with quality scores arranged in front in each to-be-pushed topic set can be selected as the to-be-pushed topics, and the pre-determined proportion can be 20%, 50%, 60%, or the like. Therefore, the corresponding number of to-be-pushed topics can be selected according to the number of the to-be-processed topics in each to-be-processed topic set, so that the richness of the to-be-pushed topics is ensured.

It should be understood that the questions to be pushed selected according to the quality scores are high in quality among the questions to be processed, namely, exquisite questions, and therefore the questions to be pushed are pushed to the target object, so that the question making effect of the target object can be guaranteed, and the purpose of checking the true learning level can be achieved.

In step S350, predicting the question to be pushed by using a pre-trained prediction model to obtain a correctness corresponding to the question to be pushed.

The prediction model can be a machine learning model used for automatically predicting the accuracy of the item to be pushed. When the prediction model is trained, the recorded answering behavior record of the target object can be used as the input of the prediction model, so that the prediction model outputs the accuracy of the answer of the target object to the question to be pushed. It should be noted that the answer behavior records of all target objects in the online learning platform may be used to train the prediction model, so as to improve the richness of the training data of the training model, and thus improve the accuracy of the prediction result of the training model.

In an embodiment of the application, the prediction model can be established by using a deep fm model, wherein the deep fm model effectively combines the points of a factorization machine and a neural network in feature learning, and can simultaneously extract low-order combined features and high-order combined features, so that the accuracy of the prediction result of the prediction model can be improved.

In other embodiments, a person skilled in the art may also use other machine learning models to predict the items to be pushed, so as to obtain the accuracy of the items to be pushed, or use a statistical model, for example, may count the accuracy of a certain number of items to be pushed in the same category information, so as to use the accuracy as the accuracy of the items to be pushed in the category information, and so on. The above are merely exemplary examples, and those skilled in the art may establish a corresponding prediction model according to implementation needs, which is not particularly limited in the present application.

In an embodiment of the application, according to the to-be-pushed question selected from each to-be-processed question set, for each target object, a trained prediction model is adopted to predict each to-be-pushed question, so that the accuracy of each target object answering each to-be-pushed question is obtained.

In the embodiment shown in fig. 3, the quality score of each to-be-processed question in each to-be-processed question set is determined, and the to-be-pushed question is selected from each to-be-processed question set according to the quality score, so that the quality of each to-be-pushed question can be ensured, a better question making effect can be obtained, the problem that the quality of the to-be-pushed question is poor is avoided, the learning time of the target object is wasted, and the real learning level of the target object cannot be detected. Meanwhile, only the questions to be pushed are predicted, so that the calculation efficiency can be improved, the calculation resources are saved, and the problems that the number of the questions to be processed is too large, the prediction time is too long, and too much calculation resources are occupied are avoided.

Based on the embodiments shown in fig. 2 and fig. 3, fig. 4 shows a flowchart of step S330 in the title recommendation method of fig. 3 according to an embodiment of the present application. Referring to fig. 4, step S330 at least includes steps S410 to S420, which are described in detail as follows:

in step S410, weights corresponding to the respective parameters in the attribute information are acquired.

In one embodiment of the present application, corresponding weights may be preset for each parameter (i.e., year, whether the question is true, whether the question is a composite question (i.e., whether a plurality of knowledge points are examined), the degree of distinction and the accuracy of the other questions, etc.) in the attribute information, for example, the weight corresponding to the year is 0.2, the weight corresponding to the question is true is 0.2, the weight corresponding to the composite question is 0.3, the weight corresponding to the degree of distinction of the other questions is 0.2, the weight corresponding to the accuracy is 0.1, and so on. The above are merely exemplary, and the present application is not limited thereto. The server may obtain the weight corresponding to each parameter in the attribute information from the local storage space for subsequent calculation.

In step S420, according to the attribute information and the weight of each topic to be processed included in the topic set to be processed, the quality score of each topic to be processed included in the topic set to be processed is calculated.

In an embodiment of the application, according to the weight corresponding to each parameter in the obtained attribute information, and according to the attribute information corresponding to each topic to be processed, weighting and calculation can be performed, so as to obtain a quality score corresponding to each topic to be processed. Therefore, the quality score of each to-be-processed topic set can be automatically calculated without manual calculation, and the determination efficiency of the quality score is improved.

Based on the embodiments shown in fig. 2 and fig. 3, fig. 5 is a schematic flowchart illustrating editing of category information of a topic to be processed in a topic recommendation method according to an embodiment of the present application. Referring to fig. 5, before step S310, editing category information of a to-be-processed topic at least includes steps S510 to S520, which are described in detail as follows:

in step S510, a category information editing interface of the to-be-processed question is displayed according to the category information editing request of the to-be-processed question.

In one embodiment of the present application, the category information editing request for the to-be-processed topic may be information for requesting editing (e.g., adding or modifying, etc.) of the category information of the to-be-processed topic. In an example, the practitioner can send the category information editing request by clicking on a particular area on the interface (e.g., a "category information editing" button, etc.).

The category information editing interface for the to-be-processed title may be an interface for editing the category information of the to-be-processed title. The corresponding relation between the to-be-processed item and the category information can be displayed in the category information editing interface, and a professional can select and edit the category information corresponding to the to-be-processed item through the configured input equipment.

In step S520, the category information is associated with the corresponding to-be-processed title according to the category information of the to-be-processed title acquired by the category information editing interface.

In an embodiment of the application, after the professional inputs the category information corresponding to the to-be-processed question in the category information editing interface, the server may associate and store the category information of the to-be-processed question acquired by the category information editing interface with the corresponding to-be-processed question. Specifically, a corresponding relation table between the to-be-processed question and the category information may be established, so that in the subsequent processing, the category information of the to-be-processed question may be obtained by querying the corresponding relation table between the to-be-processed question and the category information.

In the embodiment shown in fig. 5, a professional can edit the category information of the to-be-processed topic through the category information editing interface, so that the category information of the to-be-processed topic is modified conveniently and the accuracy of the category information is ensured.

Based on the embodiment shown in fig. 2, fig. 6 shows a flowchart of step S240 in the title recommendation method of fig. 2 according to an embodiment of the present application. Referring to fig. 6, step S240 at least includes steps S610 to S630, which are described in detail as follows:

in step S610, the titles to be pushed are divided according to the category information of the titles to be pushed, so as to obtain a set of titles to be pushed corresponding to each category information.

In an embodiment of the present application, category information corresponding to titles to be pushed is compared, and titles to be pushed corresponding to the same category information can be determined, so as to obtain a set of titles to be pushed corresponding to each category information. It should be noted that, since the to-be-pushed topic is selected from the to-be-pushed topics, the category information corresponding to the to-be-pushed topic can be obtained by querying the correspondence table between the to-be-pushed topic and the category information.

In step S620, according to the accuracy of each to-be-pushed topic included in each to-be-pushed topic set, the mastery degree of the target object on the knowledge point corresponding to each category information is calculated.

In an embodiment of the present application, it should be understood that the to-be-pushed topics in each to-be-pushed topic set correspond to the same category information, and each category information corresponds to a knowledge point, so that the to-be-pushed topics in the same to-be-pushed topic set correspond to the same knowledge point. Therefore, the mastery degree of the target object to each knowledge point can be obtained according to the accuracy of each topic to be pushed in the same topic set to be pushed. It should be understood that the higher the accuracy of each topic to be pushed contained in a certain topic set to be pushed is, the higher the mastery degree of the target object on the knowledge point is, and the lower the accuracy of each topic to be pushed contained in a certain topic set to be pushed is, the lower the mastery degree of the target object on the knowledge point is.

In step S630, a target topic is selected from the topics to be pushed according to the mastery level and a predetermined threshold used for determining a lower limit of the mastery level of the target object.

In one embodiment of the present application, the predetermined threshold may be a threshold for determining a lower limit of the mastery degree of the target object, which may be any value set in advance, for example, the predetermined threshold may be 60%, 80%, 90%, or the like. If the mastery degree of the target object to a certain knowledge point is lower than the preset threshold, it can be shown that the learning degree of the target object to the knowledge point is not enough, so that the question to be pushed corresponding to the knowledge point can be selected as the target question, and then pushed to the target object, thereby achieving the purpose of targeted exercise.

Based on the embodiments shown in fig. 2 and fig. 6, fig. 7 shows a flowchart of step S630 in the title recommendation method of fig. 6 according to an embodiment of the present application. Referring to fig. 7, step S630 includes at least steps S710 to S740, and is described in detail as follows:

in step S710, a knowledge graph is acquired, which includes the priority relationship between the knowledge points corresponding to the respective category information.

In one embodiment of the present application, the knowledge-graph may be a graph for representing relationships between knowledge points, and may contain a plurality of sequences of knowledge points consisting of related knowledge points. In each knowledge point sequence, the priority of the knowledge point arranged in the front is higher than that of the knowledge point arranged in the back, that is, the knowledge point arranged in the front is the basic knowledge point of the knowledge point arranged in the back. It should be understood that in the learning process, the knowledge points with higher priority, i.e. the basic knowledge points, should be mastered first, and the knowledge points with higher difficulty (i.e. the knowledge points with lower priority) can be mastered. The knowledge graph can be pre-established and stored in the server for later retrieval.

In step S720, a knowledge point whose degree of grasp is lower than the predetermined threshold is identified as a knowledge point to be grasped.

In one embodiment of the present application, the degree of grasp of each knowledge point by the target object may be compared with a predetermined threshold value, and the knowledge points whose degree of grasp is lower than the predetermined threshold value may be determined as knowledge points to be grasped, i.e., knowledge points whose degree of grasp is not high and need to be strengthened. For example, the degree of grasp of the knowledge point a by the target object is 59%, the degree of grasp of the knowledge point B is 80%, and the predetermined threshold value is 70%, and therefore, the knowledge point a may be determined as a knowledge point to be grasped, and so on.

In step S730, according to the knowledge graph, a knowledge point with a higher priority than the knowledge point to be mastered is identified as a knowledge point to be pushed.

In an embodiment of the application, according to the determined knowledge point to be mastered, a knowledge point which is located in the same knowledge point sequence as the knowledge point to be mastered and has a higher priority than the knowledge point to be mastered can be obtained by querying a knowledge map and is used as the knowledge point to be pushed. It should be noted that the knowledge point to be pushed should be a knowledge point that is related to and is the basis of the knowledge point to be mastered.

In step S740, the topic to be pushed corresponding to the knowledge point to be pushed is determined as the target topic.

In an embodiment of the application, according to the determined knowledge point to be pushed, a topic to be pushed corresponding to the knowledge point to be pushed is determined as a target topic from the topics to be pushed. Specifically, the category information corresponding to the knowledge point to be pushed may be determined according to the knowledge point to be pushed, and then, the query may be performed according to the determined category information to obtain a topic to be pushed corresponding to the category information, so as to use the topic as a target topic.

In the embodiment shown in fig. 7, a knowledge graph is established in advance, a knowledge point with a mastery degree lower than a predetermined threshold is determined as a knowledge point to be mastered, a knowledge point to be pushed with a priority higher than that of the knowledge point to be mastered is queried from the knowledge graph, and a topic to be pushed corresponding to the knowledge point to be pushed is determined as a target topic. Therefore, under the condition that the mastery degree of the target object to the knowledge point to be mastered is not high, the question to be pushed of the knowledge point to be pushed with the priority higher than that of the knowledge point to be mastered can be preferentially pushed to the target object, so that the target object can preferentially learn the basic knowledge point (namely the knowledge point to be pushed) of the knowledge point to be mastered, the learning level of the basic knowledge point of the target object is improved, and the target object can better and faster understand and master the knowledge point to be mastered.

Based on the embodiments shown in fig. 2 and fig. 6, in an embodiment of the present application, the title recommendation method further includes:

and calculating the average value of the mastery degrees according to the mastery degrees of the knowledge points corresponding to the various kinds of information, wherein the average value is used as the preset threshold value.

In this embodiment, the average of the degrees of grasp of the knowledge points corresponding to all the category information by the target object may be calculated based on the degree of grasp of the knowledge points corresponding to each category information by the target object. It should be understood that the average value may represent an average level of the degree of grasp of the knowledge points corresponding to all the category information by the target object, and may be taken as a predetermined threshold value. So that the determination of the predetermined threshold can be made to more closely match the current learning level of the target object based on the average level of the degree of grasp of the knowledge points of the target object. The preset threshold is prevented from being set too high, and the mastery degree of all knowledge points of the target object is required to be higher than the preset threshold, so that the actual learning condition of the target object is not met.

Based on the embodiment shown in fig. 2, in an embodiment of the present application, according to the category information and the answer result, updating a correctness of a to-be-pushed question corresponding to the category information in the to-be-pushed question includes:

if the answer result is correct, updating the correctness of the question to be pushed corresponding to the category information in the questions to be pushed to be 1;

and if the answer result is wrong, updating the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question to be 0.

In this embodiment, if the answer result of the question answered by the target object is correct, it may be considered that the target object can answer the question to be pushed corresponding to the category information of the answered question. Therefore, the accuracy of the to-be-pushed question corresponding to the type information of the answered question can be updated to 1, if the answer result of the question answered by the target object is wrong, the target object can be considered to be unable to answer the to-be-pushed question corresponding to the type information of the answered question, and the accuracy of the to-be-pushed question corresponding to the type information of the answered question can be updated to 0. Therefore, the accuracy of the questions to be pushed can meet the learning level of the target object, and the updating efficiency of the accuracy can be improved.

Based on the embodiment shown in fig. 2, fig. 8 shows a flowchart of step S230 in the title recommendation method of fig. 2 according to an embodiment of the present application. Referring to fig. 8, step S230 at least includes steps S810 to S830, which are described in detail as follows:

in step S810, a history of answer behavior of the target object is acquired.

In an embodiment of the present application, the historical answer behavior record may be historical answer behavior records of all target objects on the online education platform, or may be historical answer behavior records of a single target object. The historical answer behavior record can be stored in a storage space of the server for subsequent processing.

In step S820, a prediction model is trained according to the historical answer behavior records and the current answer behavior records.

In an embodiment of the application, the acquired historical answer behavior record of the target object and the current answer behavior record of the target object are used as training data of a prediction model, so that the prediction model can predict the accuracy of the answer of the target object to the question to be pushed. Therefore, the current answer behavior record of the target object is added to serve as training data, so that the prediction model can be updated in real time, the output result of the prediction model can better accord with the current learning level of the target object, and the accuracy of the output result of the prediction model is improved.

In step S830, the trained prediction model is used to predict the to-be-pushed question corresponding to the category information in the to-be-pushed questions, so as to obtain the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question.

In an embodiment of the application, after the retraining of the prediction model is completed, the retraining of the prediction model is adopted to perform retraining on the questions to be pushed corresponding to the category information of the questions answered by the target object, so as to obtain an updated accuracy, so that the updated accuracy can better meet the current learning level of the target object, and the accuracy of the accuracy is ensured.

In an embodiment of the application, the trained prediction model can also be used for predicting all questions to be pushed, so that the accuracy of all the questions to be pushed is updated, and the learning level corresponding to the target object in real time is ensured.

In an embodiment of the present application, the correctness rate corresponding to the topic to be pushed can also be updated every other predetermined period. Specifically, the prediction model may be trained according to the answer behavior record of the target object in the last predetermined period and the historical answer behavior record, and the prediction model is used to predict the to-be-pushed question every other predetermined period (for example, 12h, 24h, or 48 h), so as to update the accuracy rate corresponding to the to-be-pushed question.

Based on the technical solution of the above embodiment, a specific application scenario of the embodiment of the present application is introduced as follows:

referring to fig. 9, fig. 9 is a flowchart illustrating a title recommendation method according to an embodiment of the present application.

In step S910, the server may obtain an item library, where the item library may include a plurality of basic items, that is, to-be-processed items; in step S920, by dividing the basic questions and determining the quality scores, the basic questions with higher quality scores can be selected as the top-quality questions to form a top-quality question bank; in step S930, historical answer behavior records of all users may be obtained from the server; in step S940, the obtained historical answer behavior records are used to train and establish a prediction model, so that the prediction model can predict the accuracy of the user answering the fine questions; in step S950, predicting the accuracy of each of the essence questions included in the essence question library by the user using the trained prediction model; in step S960, according to the current answer behavior of the user, obtaining the current answer behavior record of the user, wherein the current answer behavior record includes the category information of the answer of the user and the answer result; in step S970, the accuracy of the top-quality questions corresponding to the category information of the answered questions in the top-quality question bank is updated according to the current answer behavior record of the user; in step S980, the mastery degree of the user on each knowledge point is calculated according to the updated accuracy of the essence questions and the correspondence between the essence questions and the knowledge points; in step S990, according to the mastery degree of each knowledge point by the user, the corresponding question is selected from the question bank as the target question to be pushed to the user.

Therefore, the question recommending method can update the accuracy of each fine question in the fine question library according to the real learning condition of the user, further update the mastery degree of the user on each knowledge point, enable the mastery degree to truly reflect the current learning level of the user, and push the corresponding fine question to the user according to the mastery degree, so that the question making effect of the user is guaranteed.

Referring to fig. 10, fig. 10 is a flowchart illustrating a step S920 in a title recommendation method of fig. 9 according to an embodiment of the present application.

Referring to fig. 10, step S920 includes at least steps S1010 to S1050, which are described in detail as follows:

in step S1010, basic questions in the question bank are acquired.

It should be understood that, since the basic questions are not screened, the basic questions are not only in large quantity, but also in variable quality. Some basic questions are too simple, so that the training effect cannot be achieved, and the time of the user is wasted.

In step S1020, the basic topic is divided according to the category information (subject-knowledge point-subject type-difficulty) of the basic topic to obtain a basic topic set corresponding to each category information.

Such as math-trigonometric function-choice question-difficult, math-trigonometric function-fill-in-the-blank question-easy, chinese-ancient poetry-merwrite-easy, … …, etc. Therefore, the basic topic set corresponding to each category of information can be obtained so as to achieve the purpose of classifying the basic topics.

In step S1030, the quality of each basic topic included in each basic topic set is scored based on the attribute information (e.g., year, true topic, composite topic, accuracy, etc.) of the basic topic included in each basic topic set.

In step S1040, the basic questions whose quality score is TopK before are selected from each basic question set as the top questions, and are fused according to the category information (subject-knowledge points) of the top questions, so as to obtain a top question set corresponding to the subject-knowledge points. Where K is a natural number different from zero, in an example, the value of K may be preset, for example, K may be 1000, 2000, 10000, or the like. In another example, the value K may be determined according to the number of basic questions included in each basic question set, for example, K may be 30% of the number of basic questions included in the basic question set, for example, if 1000 basic questions are included in the basic question set, then K is 1000 × 30%, which is 300, so that the basic questions whose quality scores in the basic question set account for the top 300 are taken as the refined questions.

After selecting the essence questions from each basic question set, the selected essence questions can be divided according to the subject-knowledge points to obtain the essence question set corresponding to the category information of each subject-knowledge point. Therefore, the number of the fine product set can be reduced, and the fine product problems belonging to the same subject and the same knowledge point can be integrated, so that the richness of the fine product problems of the same subject and the same knowledge point can be ensured.

In step S1050, all the essence question sets are deduplicated to form a essence question library.

And comparing the competitive products contained in each competitive product set, if repeated competitive products exist, deleting the repeated competitive products from the competitive product sets with the repeated competitive products, and only keeping the repeated competitive products in one competitive product set. For example, the extract question set a, the extract question set B and the extract question set C all contain extract questions D, two extract question sets can be arbitrarily selected from the extract question set a, the extract question set B and the extract question set C to delete the extract questions D contained therein, and the extract questions D are only retained in one extract question set, so that the problem that multiple repeated extract questions exist in the extract question library and occupy resources is avoided.

The following describes embodiments of the apparatus of the present application, which can be used to perform the recommendation method for topics in the above embodiments of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the subject recommendation method described above in the present application.

FIG. 11 shows a block diagram of a title recommendation device according to an embodiment of the present application.

Referring to fig. 11, an apparatus for recommending a title according to an embodiment of the present application includes:

a first obtaining module 1110, configured to obtain a question to be pushed and a correctness corresponding to the question to be pushed, where the correctness is used to describe a possibility that a target object answers the question to be pushed;

a second obtaining module 1120, configured to obtain a current answer behavior record of the target object, and determine, according to the current answer behavior record, category information of an answer of the target object and an answer result, where the answer result includes an answer right and an answer error;

an updating module 1130, configured to update a correctness of a to-be-pushed question corresponding to the category information in the to-be-pushed question according to the category information and the answer result;

the processing module 1140 is configured to determine a target topic in the to-be-pushed topic according to the updated accuracy of the to-be-pushed topic, so as to push the target topic to the target object.

In some embodiments of the present application, based on the foregoing, the first obtaining module 1110 is configured to: acquiring a plurality of to-be-processed titles and category information and attribute information corresponding to the plurality of to-be-processed titles; dividing according to the category information of the plurality of to-be-processed questions to obtain to-be-processed question sets corresponding to the category information; determining the quality score of each topic to be processed in the topic set to be processed according to the attribute information of each topic to be processed contained in the topic set to be processed; selecting a question to be pushed from the question set to be processed according to the quality score; and predicting the questions to be pushed by adopting a pre-trained prediction model to obtain the corresponding accuracy of the questions to be pushed.

In some embodiments of the present application, based on the foregoing, the first obtaining module 1110 is configured to: acquiring weights corresponding to all parameters in the attribute information; and calculating the quality scores of the to-be-processed questions contained in the to-be-processed question set according to the attribute information and the weight of each to-be-processed question contained in the to-be-processed question set.

In some embodiments of the present application, based on the foregoing solution, before the multiple to-be-processed titles and the category information and the attribute information corresponding to the multiple to-be-processed titles are obtained, the first obtaining module 1110 is further configured to: displaying a category information editing interface of the to-be-processed question according to the category information editing request of the to-be-processed question; and associating the category information with the corresponding to-be-processed title according to the category information of the to-be-processed title acquired by the category information editing interface.

In some embodiments of the present application, based on the foregoing, the processing module 1140 is configured to: dividing according to the category information of the questions to be pushed to obtain a set of the questions to be pushed corresponding to each category information; calculating the mastery degree of the target object to the knowledge points corresponding to the category information according to the accuracy of each to-be-pushed question contained in each to-be-pushed question set; and selecting a target topic from the topics to be pushed according to the mastery degree and a preset threshold value for determining the lower limit of the mastery degree of the target object.

In some embodiments of the present application, based on the foregoing, the processing module 1140 is configured to: acquiring a knowledge graph, wherein the knowledge graph comprises priority relations between knowledge points corresponding to all kinds of information; identifying the knowledge points with the mastery degree lower than the preset threshold value as knowledge points to be mastered; identifying the knowledge points with the priority higher than the knowledge points to be mastered as knowledge points to be pushed according to the knowledge graph; and determining the to-be-pushed question corresponding to the to-be-pushed knowledge point as the target question.

In some embodiments of the present application, based on the foregoing, the processing module 1140 is further configured; and calculating the average value of the mastery degrees according to the mastery degrees of the knowledge points corresponding to the various kinds of information, wherein the average value is used as the preset threshold value.

In some embodiments of the present application, based on the foregoing, the update module 1130 is configured to: if the answer result is correct, updating the correctness of the question to be pushed corresponding to the category information in the questions to be pushed to be 1; and if the answer result is wrong, updating the accuracy of the to-be-pushed question corresponding to the category information in the to-be-pushed question to be 0.

In some embodiments of the present application, based on the foregoing, the update module 1130 is configured to: acquiring a historical answer behavior record of a target object; training a prediction model according to the historical answer behavior record and the current answer behavior record; and predicting the to-be-pushed questions corresponding to the category information in the to-be-pushed questions by adopting the trained prediction model to obtain the accuracy of the to-be-pushed questions corresponding to the category information in the to-be-pushed questions.

It should be noted that the computer system of the electronic device shown in fig. 11 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.

As shown in fig. 12, the computer system includes a Central Processing Unit (CPU)1201, which can perform various appropriate actions and processes, such as performing the methods described in the above embodiments, according to a program stored in a Read-Only Memory (ROM) 1202 or a program loaded from a storage section 1208 into a Random Access Memory (RAM) 1203. In the RAM 1203, various programs and data necessary for system operation are also stored. The CPU 1201, ROM 1202, and RAM 1203 are connected to each other by a bus 1204. An Input/Output (I/O) interface 1205 is also connected to bus 1204.

The following components are connected to the I/O interface 1205: an input section 1206 including a keyboard, a mouse, and the like; an output section 1207 including a Display device such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and a speaker; a storage section 1208 including a hard disk and the like; and a communication section 1209 including a Network interface card such as a LAN (Local Area Network) card, a modem, or the like. The communication section 1209 performs communication processing via a network such as the internet. A driver 1210 is also connected to the I/O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 1210 as necessary, so that a computer program read out therefrom is mounted into the storage section 1208 as necessary.

In particular, according to embodiments of the application, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising a computer program for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 1209, and/or installed from the removable medium 1211. The computer program executes various functions defined in the system of the present application when executed by a Central Processing Unit (CPU) 1201.

It should be noted that the computer readable medium shown in the embodiments of the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a Read-Only Memory (ROM), an Erasable Programmable Read-Only Memory (EPROM), a flash Memory, an optical fiber, a portable Compact Disc Read-Only Memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with a computer program embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. Each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present application may be implemented by software, or may be implemented by hardware, and the described units may also be disposed in a processor. Wherein the names of the elements do not in some way constitute a limitation on the elements themselves.

As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method described in the above embodiments.

It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the application. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.

Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.

Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains.

It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A method for recommending a topic, comprising:

2. The method according to claim 1, wherein obtaining the topic to be pushed and the correctness corresponding to the topic to be pushed comprises:

acquiring a plurality of to-be-processed titles and category information and attribute information corresponding to the plurality of to-be-processed titles;

dividing according to the category information of the plurality of to-be-processed questions to obtain to-be-processed question sets corresponding to the category information;

determining the quality score of each topic to be processed in the topic set to be processed according to the attribute information of each topic to be processed contained in the topic set to be processed;

selecting a question to be pushed from the question set to be processed according to the quality score;

and predicting the questions to be pushed by adopting a pre-trained prediction model to obtain the corresponding accuracy of the questions to be pushed.

3. The method according to claim 2, wherein determining the quality score of each topic to be processed in the set of topics to be processed according to the attribute information of each topic to be processed contained in the set of topics to be processed comprises:

acquiring weights corresponding to all parameters in the attribute information;

and calculating the quality scores of the to-be-processed questions contained in the to-be-processed question set according to the attribute information and the weight of each to-be-processed question contained in the to-be-processed question set.

4. The method according to claim 2, wherein before obtaining a plurality of to-be-processed titles and category information and attribute information corresponding to the plurality of to-be-processed titles, the method further comprises:

displaying a category information editing interface of the to-be-processed question according to the category information editing request of the to-be-processed question;

and associating the category information with the corresponding to-be-processed title according to the category information of the to-be-processed title acquired by the category information editing interface.

5. The method according to claim 1, wherein determining a target topic in the topics to be pushed according to the updated accuracy rate of the topics to be pushed comprises:

dividing according to the category information of the questions to be pushed to obtain a set of the questions to be pushed corresponding to each category information;

calculating the mastery degree of the target object to the knowledge points corresponding to the category information according to the accuracy of each to-be-pushed question contained in each to-be-pushed question set;

and selecting a target topic from the topics to be pushed according to the mastery degree and a preset threshold value for determining the lower limit of the mastery degree of the target object.

6. The method according to claim 5, wherein selecting the target topic from the topics to be pushed according to the mastery degree and a predetermined threshold value for determining a lower limit of the mastery degree of the target object comprises:

acquiring a knowledge graph, wherein the knowledge graph comprises priority relations between knowledge points corresponding to all kinds of information;

identifying the knowledge points with the mastery degree lower than the preset threshold value as knowledge points to be mastered;

identifying the knowledge points with the priority higher than the knowledge points to be mastered as knowledge points to be pushed according to the knowledge graph;

and determining the to-be-pushed question corresponding to the to-be-pushed knowledge point as the target question.

7. The method of claim 5, further comprising:

8. The method according to claim 1, wherein updating a correct rate of the to-be-pushed question corresponding to the category information in the to-be-pushed question according to the category information and the answer result comprises:

9. The method according to claim 1, wherein updating a correct rate of the to-be-pushed question corresponding to the category information in the to-be-pushed question according to the category information and the answer result comprises:

acquiring a historical answer behavior record of a target object;

training a prediction model according to the historical answer behavior record and the current answer behavior record;

and predicting the to-be-pushed questions corresponding to the category information in the to-be-pushed questions by adopting the trained prediction model to obtain the accuracy of the to-be-pushed questions corresponding to the category information in the to-be-pushed questions.

10. An apparatus for recommending a topic, comprising: