WO2023071505A1

WO2023071505A1 - Question recommendation method and apparatus, and computer device and storage medium

Info

Publication number: WO2023071505A1
Application number: PCT/CN2022/116076
Authority: WO
Inventors: 李梦瑶; 邓澍军; 阚砚馨; 刘思涛
Original assignee: 北京有竹居网络技术有限公司
Priority date: 2021-10-27
Filing date: 2022-08-31
Publication date: 2023-05-04
Also published as: CN116028602A

Abstract

Provided are a question recommendation method and apparatus, and a computer device and a storage medium. The method comprises: in response to a question recommendation request, initiated by a user for the first time, for a target knowledge point, obtaining current learning ability information of the user, the learning ability information being used for representing the grasp ability of the user on the target knowledge point (S101); calling a rudiments recommendation strategy in a strategy engine, and according to the learning ability information and answer correct rate information of each candidate question associated with the target knowledge point in a question library, determining an initial recommended question recommended to the user in the current recommendation process (S102); and in response to completion of the initial recommended question and the current recommendation process being not ended, enabling a single question recommendation strategy in the strategy engine, and determining, according to an answer condition of the user for the recommended question in the current recommendation process, whether a question level adjustment condition is satisfied; and selecting, according to a determination result, from the question library the next question to be recommended to the user (S103). Therefore, more appropriate questions are recommended to the user, thereby improving learning efficiency.

Description

A topic recommendation method, device, computer equipment and storage medium

This disclosure claims the priority of the Chinese patent application filed with the State Intellectual Property Office of China on October 27, 2021, with the application number 202111254777.7, and the title of the invention is "a method, device, computer equipment and storage medium for topic recommendation", all of which The contents are incorporated by reference in this disclosure.

technical field

The present disclosure relates to the technical field of the Internet, in particular, to a topic recommendation method, device, computer equipment and storage medium.

Background technique

Nowadays, the method of recommending topics for students is relatively fixed, that is, a fixed number of topics are selected from the pre-set question bank and recommended to students.

Due to the different learning progress of different students, even the same student, in the process of answering and learning, the learning progress will change. The above-mentioned fixed recommendation method has poor flexibility, which will lead to a waste of learning time and unsatisfactory learning effects.

Contents of the invention

Embodiments of the present disclosure at least provide a topic recommendation method, device, computer equipment, and storage medium to recommend more suitable learning topics for users and improve learning efficiency.

In the first aspect, the embodiment of the present disclosure provides a topic recommendation method, including:

In response to the topic recommendation request for the target knowledge point initiated by the user for the first time, the current learning ability information of the user is obtained; the learning ability information is used to represent the user's ability to grasp the target knowledge point;

Invoke the entry-level recommendation strategy in the strategy engine, and determine the initial recommendation topic recommended for the user in the current recommendation process according to the learning ability information and the correct answer information of each candidate topic associated with the target knowledge point in the question bank;

In response to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the policy engine, and judge whether the user meets the topic adjustment according to the user's answers to the recommended topics in the current recommendation process. Level conditions; according to the judgment result, select the next question to be recommended for the user from the question bank.

In the second aspect, an embodiment of the present disclosure further provides a topic recommendation device, including:

An information acquisition module, configured to acquire current learning ability information of the user in response to a topic recommendation request for a target knowledge point initiated by the user for the first time; the learning ability information is used to represent the user's ability to grasp the target knowledge point;

The first topic recommendation module is used to call the entry-level recommendation strategy in the strategy engine, and determine the correct answer rate information of each candidate topic associated with the target knowledge point in the question bank according to the learning ability information and determine the current recommendation process for all Describe the initial recommendation topic recommended by the user;

The second topic recommendation module is used to respond to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the strategy engine, and according to the user's answer to the recommended topic in the current recommendation process, Judging whether the user satisfies the question leveling condition; according to the judgment result, selecting the next question to be recommended for the user from the question bank.

In a third aspect, an embodiment of the present disclosure further provides a computer device, including: a processor, a memory, and a bus, the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processing The processor communicates with the memory through a bus, and when the machine-readable instructions are executed by the processor, the steps of the topic recommendation method of the above-mentioned first aspect are executed.

In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the topic recommendation method in the first aspect above are executed.

For the effect description of the above-mentioned topic recommendation apparatus, computer equipment and storage medium, please refer to the description of the above-mentioned topic recommendation method, which will not be repeated here.

A topic recommendation method, device, computer equipment, and storage medium provided by the embodiments of the present disclosure can obtain the current learning ability information of the user by responding to the topic recommendation request for the target knowledge point initiated by the user for the first time. Here, the learning ability information It is used to represent the user's ability to master the target knowledge points. After that, call the entry-level recommendation strategy in the strategy engine, and determine the initial recommendation topic for the user in the current recommendation process based on the learning ability information and the correct answer information of each candidate topic associated with the target knowledge point in the question bank. Here, the correct answer rate of the candidate topic itself (predicted based on the historical answering situation of each user) can reflect the difficulty of the topic. Therefore, combined with the user's learning ability and the correct answer rate of the candidate topic, it can be more reasonable for the user's initial recommendation and learning ability. Matching initial recommended topics. Afterwards, in response to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the policy engine, and judge whether the target user meets the upgrade conditions according to the user's answers to the recommended topics in the current process. Here, since the user can update the user's mastery of the target knowledge points based on the user's answers to the recommended questions, it is possible to formulate a next-step answer plan for the user. For example, users who are not good at answering questions can downgrade to practice, that is, practice relatively simple questions first, and gradually master the corresponding target knowledge points from easy to difficult. Another example is a user who has a very good answer to the questions, indicating that the current difficult questions are relatively easy for the user. At this time, upgrade exercises can be carried out for the user to improve learning efficiency.

In this way, the embodiments of the present disclosure can not only initially recommend introductory practice questions for users based on the user's learning ability, but also continuously adjust the practice questions adaptively according to the user's answering situation during the user's learning process, through this personalized recommendation and practice. The flexible adjustment can ensure the rationality of the topic recommendation in the whole practice process, which is beneficial to reduce the user's invalid answering time and improve the learning efficiency.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present disclosure more clearly, the following will briefly introduce the accompanying drawings used in the embodiments. The accompanying drawings here are incorporated into the specification and constitute a part of the specification. The drawings show embodiments consistent with the present disclosure, and are used together with the specification to illustrate the technical solutions of the embodiments of the present disclosure. It should be understood that the following drawings only show some embodiments of the present disclosure, and therefore should not be regarded as limiting the scope. For those skilled in the art, they can also make From these drawings other related drawings are obtained.

FIG. 1 shows a flowchart of a topic recommendation method provided by an embodiment of the present disclosure;

FIG. 2 shows a specific flow chart of topic recommendation provided by an embodiment of the present disclosure;

FIG. 3 shows a diagram of a user interaction interface provided by an embodiment of the present disclosure;

Fig. 4 shows a schematic diagram of a topic recommendation device provided by an embodiment of the present disclosure;

Fig. 5 shows a schematic structural diagram of a computer device provided by an embodiment of the present disclosure.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only It is a part of the embodiments of the present disclosure, but not all of them. The components of the disclosed embodiments generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the present disclosure provided in the accompanying drawings is not intended to limit the scope of the claimed embodiments of the present disclosure, but merely represents selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the embodiments of the present disclosure.

In addition, the terms "first", "second" and the like in the description and claims in the embodiments of the present disclosure and the above drawings are used to distinguish similar objects, and not necessarily used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments described herein can be practiced in sequences other than those illustrated or described herein.

"Plural or several" mentioned herein means two or more. "And/or" describes the association relationship of associated objects, indicating that there may be three types of relationships, for example, A and/or B may indicate: A exists alone, A and B exist simultaneously, and B exists independently. The character "/" generally indicates that the contextual objects are an "or" relationship.

After research, it is found that the current method of recommending topics for students is relatively fixed, that is, using a preset fixed number of topics to recommend to students. For example, recommend two topics for top students and three topics for middle students. In this way, in the process of answering questions for middle school students, because the middle school students have mastered the knowledge points of this study well, they want to go to the next process after only answering one question. , which will cause poor experience for students. Therefore, the above-mentioned fixed recommendation method has poor flexibility, resulting in poor user experience during the answering process.

Based on the above research, the embodiment of the present disclosure provides a topic recommendation method. Since the correct answer rate information of each candidate topic can accurately reflect the correct answer rate of most users on the candidate topic, according to the learning ability information, and The correct answer information of each candidate question associated with the target knowledge point in the question bank can recommend initial recommended questions for users, and can more accurately recommend initial recommended questions for users that match their current learning ability. Since the answers to the recommended questions can further represent the state of the user's mastery of the target knowledge points, making full use of the answers to the recommended questions can accurately determine whether the user meets the adjustment conditions, such as having mastered the target knowledge points, judging that the user meets upgrade conditions. According to the user's level, the difficulty of subsequent recommended questions can be adjusted individually, so that users can truly feel that they have mastered the target knowledge points they have learned. At the same time, it can also reduce the practice of the same type of questions, save practice time and improve the quality Practice efficiency and improve user experience.

The defects in the above solutions are all the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions to the above problems proposed by the embodiments of the present disclosure below should be Contributions made by the inventors to embodiments of the disclosure.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

To facilitate the understanding of this embodiment, a topic recommendation method disclosed in the embodiments of the present disclosure is firstly introduced in detail. The subject of the topic recommendation method provided in the embodiments of the present disclosure is generally a computer device with certain computing capabilities. In some possible implementation manners, the topic recommendation method may be implemented by a processor invoking computer-readable instructions stored in a memory.

The topic recommendation method provided by the embodiments of the present disclosure will be described below by taking the computer device as an example as the execution subject.

First of all, the application scenarios of a method for recommending topics disclosed in the embodiments of the present disclosure may include online education application scenarios, and specifically may be applied to scenarios of recommending practice topics for students (including in-class practice topics, after-class practice topics, etc.). Referring to FIG. 1 , it is a flowchart of a topic recommendation method provided by an embodiment of the present disclosure. The method includes steps S101 to S103, wherein:

S101: In response to a topic recommendation request for a target knowledge point initiated by the user for the first time, obtain the user's current learning ability information; the learning ability information is used to represent the user's ability to master the target knowledge point.

In this step, the topic recommendation request may be a request initiated by the user for the first time to obtain practice topics containing target knowledge points.

During specific implementation, the topic recommendation request may include the user's identity information, and in response to the user's topic recommendation request, the learning ability information of the user is obtained according to the user's identity information. Here, the identity information may include, for example, a registered identity document (id). The learning ability information may include the learning level of the user for the target knowledge point, such as the first level, the second level, . . . . Here, the learning ability corresponding to the first level is higher than that of the latter level, such as the learning ability corresponding to the second level.

Exemplarily, in the case of responding to a topic recommendation request for a target knowledge point initiated by a user for the first time, the learning level of the user at the current moment may be queried by using the user's identity id.

In some embodiments, the manner of determining the learning ability information, specifically, in the case that there is historical learning data of the user for the target knowledge point, can be based on the historical learning data of the user for the historical answer accuracy of the target knowledge point, and the user's Determine the user's target learning ability level based on the proportion of downgraded learning time for the explanation content of the target knowledge point. The target learning ability level is used as learning ability information. However, if there is no historical learning data of the user, the default learning ability level is used as the learning ability information of the user.

The historical learning data may include the data of the user historically practicing topics containing the target knowledge points and/or learning the explanation content of the target knowledge points. The correctness rate of historical answers may include the correctness rate of each question answered by the user at the historical moment corresponding to the target knowledge point.

The explanation content for the target knowledge point may be an explanation content for a course including the target knowledge point. The display form of the explanation content may include, for example, a video explanation content, an audio explanation content, a presentation (such as PPT, etc.) explanation content, and the like. The degraded learning duration may include at least one of the user's fallback behavior duration, pause behavior duration, and deceleration behavior duration for the explained content. Exemplarily, the duration of the rollback behavior may be the duration recorded by the system for repeatedly playing the same explanation content when the user performs the rollback behavior while watching a video explaining the content. The duration of the pause behavior may be the pause duration and/or number of pauses recorded by the system when the user performs the pause behavior of the video while watching the video explaining the content. The duration of the deceleration behavior can be the slow playback duration recorded by the system when the user watches the video at a lower than normal speed during the process of watching the video explaining the content. Proportion of downgraded learning time = downgraded learning time/teaching time of explaining content, where the teaching time of explaining content is the playing time of video of explaining content.

To determine the user's target learning ability level, a possible implementation method is to determine the user's goal when the proportion of the downgraded learning time is less than the preset time proportion, and the correct rate of historical answers is greater than or equal to the preset correct rate. The level of learning ability is the first level. Conversely, when the proportion of downgraded learning time is greater than or equal to the proportion of preset time, or the correct rate of historical answers is less than the correct rate of preset answers, it is determined that the user's target learning ability level is the second level.

Another possible implementation manner is to determine that the user's target learning ability level is the first level when the downgraded learning duration is less than the preset duration. On the contrary, when the downgraded learning duration is greater than or equal to the preset duration, it is determined that the user's target learning ability level is the second level.

Here, in the case where the degraded learning duration includes the fallback learning duration of the same knowledge point in the explanation knowledge points for the candidate topics, if multiple candidate topics include multiple explanation knowledge points, and each explanation knowledge point corresponds to a fallback learning time, multiple fallback learning durations are counted, and there are multiple downgraded learning duration ratios determined by using the fallback learning duration. At this time, you can choose the largest downgraded learning duration ratio and compare it with the preset duration ratio .

Here, the preset duration ratio and the preset accumulative accuracy rate can be set according to the difficulty level of the question and historical experience data, which are not specifically limited in this embodiment of the present disclosure.

Exemplarily, in the case where the degraded learning duration includes the fallback learning duration, if the fallback learning duration is less than the preset duration and the correct answer rate of historical questions is greater than the preset answer correct rate, determine the learning ability of the target user Information is first class. Conversely, if the fallback learning duration is greater than or equal to the preset duration, or the correct answer rate of historical questions is less than or equal to the preset correct answer rate, it is determined that the learning ability information of the target user is at the second level.

Here, the preset answer duration and the preset answer accuracy rate may be set according to the difficulty level of the question and historical experience data, which are not specifically limited in this embodiment of the present disclosure.

Alternatively, the historical learning data may also include the behavior data of the target user during the process of learning the explanation content of the target knowledge point (the explanation for the knowledge point in the history topic). For example, the target user's classroom feedback after studying the course containing the target knowledge points (including the course content is simple, or the course content is complex). Here, the learning ability information of the target user can be judged more objectively by using the classroom feedback result. Exemplarily, when the class feedback is that the course content is simple, determine the learning ability information of the target user as the first level; when the class feedback is that the course content is complex, determine the target user learning ability as the second level.

S102: Invoke the introduction recommendation strategy in the strategy engine, and determine the initial recommended topic for the user in the current recommendation process according to the learning ability information and the correct answer information of each candidate topic associated with the target knowledge point in the question bank.

In this step, the question bank stores a plurality of candidate questions that are preset and associated with the target knowledge points. Candidate topics include topics corresponding to each knowledge point. Here, the knowledge points may be in a tree structure, and there may be multiple lower-level knowledge points under each upper-level knowledge point, and the embodiments of the present disclosure do not limit the level of the target knowledge points. Exemplarily, the target knowledge point may be addition and subtraction, or a lower level of addition and subtraction, such as addition of tens digits, or addition of hundreds of digits.

Each candidate topic corresponds to the correct answer rate information, and the correct answer rate information includes the correct answer rate and the estimated confidence of the correct answer rate. The correct answer rate information may be the information output by the model, which is determined by counting the historical learning data of the candidate topic. For example, count the answers of multiple students after answering the candidate question, and determine the correct answer rate and estimated confidence of some/all students for the candidate question. For example, the average of the correct answer rates of some/all students for the candidate topic may be counted, and the average value may be used as the correct answer rate of the candidate topic.

According to the learning ability information, the user's current learning ability level can be judged. According to the correct answer rate information of each candidate topic associated with the target knowledge point in the question bank, the learning ability level matching each candidate topic can be determined, and then the user's current learning ability level can be judged. Candidate topics that can be learned, and select one of the topics from the candidate topics that can be learned as the initial recommendation topic.

Here, the corresponding relationship between the correct answer rate information and the learning ability level can be set in advance; according to the corresponding relationship, when the correct answer rate information of the candidate topic is determined, the learning ability level matching the candidate topic can be determined .

Exemplarily, according to the correct answer rate information of each candidate topic, it is determined that the correct answer rate of candidate topic A is [P2, P1], and then, according to the corresponding relationship between the correct answer rate and the level of learning ability, determine that the correct answer rate of candidate topic A is [P2, P1]. The matching learning ability level is the first level. Determine that the correct answer rate of candidate topic B is [P3, P2], and according to the corresponding relationship between the correct answer rate and the learning ability level, determine that the learning ability level matching candidate topic B is the second level. Wherein, P1≥P2≥P3, and the value range of P1, P3 and P3 is 0-100%, including 0 and 100%. If the learning ability level of the user is the first level, any one of the candidate topics corresponding to [P2, P1] can be selected as the initial recommended topic. If the user's learning ability is at the second level, any one of the candidate topics corresponding to [P3, P2] can be selected as the initial recommended topic.

The policy engine in the system stores a pre-set introductory recommendation strategy, and the introductory recommendation strategy may be a strategy for recommending an initial recommended topic for the user for the first time. Introductory recommendation strategies can include the following strategies:

Strategy 1. If the learning ability level indicated by the learning ability information is the first level, and there are candidate questions whose correct answer rate is within the target correct rate range among the candidate questions, select the candidate questions whose answer correct rate is within the target correct rate range. , select any topic as the initial recommendation topic.

Wherein, the correct answer rate is the correct answer rate indicated by the correct answer rate information. The range of the target accuracy rate may be determined according to the statistics of at least some users' answer accuracy rates for the candidate questions, which is not specifically limited in this embodiment of the present disclosure.

Exemplarily, the range of the target accuracy rate is recorded as [Tmin, Tmax], and the correct answer rate information of each candidate topic indicates that the answer accuracy rate is recorded as T. If Tmin≤Tmax, the topic corresponding to T is used as the initial recommended topic, and recommended to users.

There may not be any candidate questions in the question bank whose correct answer rate is within the target correct rate range. At this time, the following strategies are implemented.

Strategy 2. If the learning ability level indicated by the learning ability information is the first level, and the correct answer rate of the candidate questions is less than the minimum value within the target correct rate range, select the question with the highest answer correct rate from the candidate questions as the initial recommendation topic.

If there is no candidate question in the question bank whose correct answer rate is within the target correct rate range, if the correct answer rates of the candidate questions in the question bank are all lower than the minimum value within the target correct rate range, that is, T<Tmin, then T can be determined The highest answer correct rate in the test, and the candidate topic corresponding to the highest answer correct rate is used as the initial recommendation topic, and recommended to the user.

Strategy 3. If the learning ability level indicated by the learning ability information is the first level, and the correct answer rate of the candidate questions is greater than the maximum value within the target correct rate range, select the question with the lowest answer correct rate from the candidate questions as the initial recommendation topic.

If the correct answer rate of the candidate questions in the question bank is greater than the maximum value within the target correct rate range, that is, T>Tmax, then the lowest answer correct rate in T can be determined, and the candidate question corresponding to the lowest answer correct rate can be used as the initial recommendation topics and recommend them to users.

Strategy 4. If the learning ability level indicated by the learning ability information is the second level, select the topic with the highest correct answer rate from each candidate topic as the initial recommendation topic.

As mentioned above, the level of the correct rate of answering questions can be determined by comparing the numerical values of the correct rate of answering questions.

In addition, the processing steps for the boundary conditions when selecting a topic are as follows:

S1021: If a plurality of candidate questions matching the learning ability level of the user are screened from the question bank, a question with the lowest difficulty may be selected from the selected candidate questions as an initial recommendation question.

Here, the level of difficulty of the topic may be set according to the importance of the target knowledge points involved in the candidate topic (for example, the importance of the required knowledge points is greater than the importance of the selected knowledge points); and/or, It may be set according to previous students' answering conditions (for example, the correct rate of answering questions); and/or set by the teacher based on the objective experience summary of candidate questions, which is not specifically limited in the embodiments of the present disclosure.

S1022: If multiple candidate questions that match the user's learning ability level are screened from the question bank, and the difficulty of each selected candidate question is the same, then the predicted answer information indicated by the correct answer rate information may be selected from the candidate questions with the same difficulty. The question with the highest estimated confidence is used as the initial recommendation question.

S1023: If multiple candidate topics matching the user's learning ability level are screened from the question bank, the difficulty of each selected candidate topic is the same, and the estimated confidence indicated by the correct answer information of each selected candidate topic If the degrees are the same, a topic can be randomly selected from the screened candidate topics as the initial recommendation topic.

S103: In response to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the strategy engine, and judge whether the user meets the topic adjustment condition according to the user's answer to the recommended topic in the current recommendation process; Judge the result and select the next question to be recommended by the user from the question bank.

After the user finishes answering the initial recommendation questions, for example, submitting the answer results of the initial recommendation questions, it can be determined that the initial recommendation questions are completed, that is, in response to the completion of the initial recommendation questions, it can be further judged whether the current recommendation process is over. In the case that the current recommendation process has not ended, the single-item recommendation strategy in the strategy engine can be further enabled.

The strategy engine in the system stores a pre-set single-question recommendation strategy. The single-question recommendation strategy can be a strategy for continuing to recommend topics for users after the user answers the initial recommendation topic and before the recommendation process is over.

In this step, the recommended topics in the current recommendation process may include the initial recommended topics that have been recommended in the above-mentioned introductory recommendation process. The recommended questions may include at least one candidate question that has been answered after the user initiates a question recommendation request. The answers to the recommended questions may include, but are not limited to, the results of the user's answers to the initially recommended questions, the actual answering time, and the user's feedback on the difficulty of the initially recommended questions. Wherein, the answer results of the recommended questions include correctness and error; the user's feedback information on the difficulty level of the recommended questions may include easier questions and more difficult questions.

The topic transfer condition may include an upgrade condition or a downgrade condition.

To determine whether the user meets the upgrade conditions, you can judge in the following two ways:

Method 1. If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is less than the preset ratio threshold, it is determined that the user meets the upgrade condition.

Method 2. If the first feedback information from the user on the recommended topic is received, and the first feedback information indicates that the recommended topic is relatively simple, then it is determined that the user meets the upgrade condition.

To judge whether the user meets the downgrade conditions, you can judge in the following three ways:

Method 1. If the user's answer to the recommended question is incorrect, it is determined that the user meets the downgrading conditions.

Method 2. If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is greater than or equal to the preset ratio threshold, it is determined that the user meets the downgrade condition;

Mode 3. If the second feedback information from the user on the recommended topic is received, and the second feedback information indicates that the recommended topic is relatively difficult, then it is determined that the user meets the downgrading condition.

As mentioned above, the preset ratio threshold may be set according to empirical values, which is not specifically limited in this embodiment of the present disclosure.

In some embodiments, before judging whether the user satisfies the question leveling condition, it may also be judged whether the user meets the question ending condition according to the user's answer to the recommended question.

Here, the answers to the recommended questions may also include, but are not limited to: the user’s practice time for the questions, the number of continuous errors in the user’s answers, the duration of the continuous errors in the user’s answers, the user’s mastery of the target knowledge points, etc.

Here, the practice duration of the topic may include the practice duration of the user on the recommended topic. The number of continuous errors may include the number of consecutive wrong answers in which the user answers multiple recommended questions. The duration of persistent errors may include the duration of the user's continuous incorrect answers in answering multiple recommended questions.

Here, the practice duration of the questions may include the cumulative duration of the practice durations of the recommended questions; the number of continuous errors in the user's answers may include the sum of the number of mistakes in the answers to the recommended questions. The continuous error duration of the user's answer is the cumulative duration of the continuous error duration of the recommended questions answered. The user's mastery of the target knowledge points may be the mastery of the knowledge points included in the recommended topic, specifically, it may be the estimated result output by the model.

According to at least one of the following information, determine whether the user has reached the end answering condition: whether the user's question practice time exceeds the preset question practice time; whether the number of persistent errors in the user's answer is greater than the set threshold; Set the duration; whether the user's mastery of the target knowledge points meets the conditions for the knowledge points to be mastered.

Among them, the conditions for ending answering include but are not limited to at least one of the following: the user’s practice time exceeds the preset question practice time; The mastery of the target knowledge points satisfies the condition of having mastered the knowledge points; the user directly triggers the process end button and does not continue to answer questions.

In some embodiments, if the user reaches the end answering condition, but the user requests to continue answering the question by triggering the next question button (such as the next question button 35 in FIG. Suggest the next topic for the user.

Here, the preset question practice duration, setting threshold, setting duration, and knowledge point mastered conditions can be set according to the specific information of the topic in the actual scene (such as the difficulty of the topic) and the experience value, which are not specifically limited in the embodiments of the present disclosure. .

Afterwards, if it is determined that the user has not met the condition for ending answering, it may continue to judge whether the user meets the condition for leveling up the question according to the answering conditions of the recommended questions. Conditions other than the above-mentioned conditions for ending the answer are not met. The process of judging whether the user satisfies the upgrade condition is the same as the above judging process, and the repeated parts will not be repeated here.

With respect to S103, according to the judgment result of the question leveling condition, the next question recommended for the user is selected from the question bank.

If the user satisfies the upgrade conditions, it can be determined that the user is an upgrade user, and a candidate question matching the upgrade user can be selected for the user from the question bank, and recommended to the upgrade user. If the user satisfies the downgrade condition, it can be determined that the user is a downgrade user, and a candidate question matching the downgrade user can be selected for the user from the question bank, and recommended to the downgrade user.

Recommend topics that match the upgraded user for the upgraded user, and the recommendation method is as follows:

Method 1. From each candidate question, select a candidate question whose answer accuracy rate is within the target accuracy rate range and whose difficulty is greater than that of the previous recommended question, as the next question recommended for the upgraded user. Exemplarily, the range of the target accuracy rate is recorded as [Tmin, Tmax], and the correct answer rate information of each candidate topic indicates that the answer accuracy rate is recorded as T, if Tmin≤T≤Tmax, and the candidate topic corresponding to T is more difficult than the recommended topic Difficulty, the candidate topic corresponding to T is used as the next topic recommended for the upgraded user.

Method 2. If there is no candidate question whose difficulty is greater than the difficulty of the previous recommended question among the candidate questions whose correct answer rate is within the target correct rate range, you can also choose a candidate question with the same difficulty as the recommended question as an upgrade user Recommended next topic.

Method 3. If none of the methods 1 and 2 mentioned above satisfy the conditions for upgrading user-recommended questions, you can also choose from various candidate questions to answer questions with a correct answer rate greater than or equal to that selected in S1021-S1023 The candidate topic with the highest correct rate will be the next topic recommended for upgraded users.

Recommend topics that match the downgraded user for the downgraded user. For example, determine the highest answer accuracy rate in T, and use the candidate topic corresponding to the highest answer accuracy rate as the topic that matches the downgraded user, and recommend it to the downgraded user. user.

After the user completes the next question, use the completed next question as the new recommended question, and continue to judge whether the current recommendation process is over. If the current recommendation process is not over, go back to S103 to enable the single question recommendation in the strategy engine strategy steps.

See Figure 2, which is a specific flow chart of topic recommendation. First, the user triggers the topic practice button to send a topic recommendation request to the strategy engine, requesting to obtain the initial recommended topic. Afterwards, the strategy engine recommends the initial recommendation topic for the user according to the entry recommendation strategy. Here, the entry recommendation strategy may refer to the entry recommendation strategy described in S102 above, and the repeated parts will not be repeated here. Afterwards, the policy engine judges whether the user has reached the end answering condition according to the end answering strategy, if yes, ends the topic recommendation process, if not, recommends the next question for the user according to the single question recommendation strategy. Here, the strategy for ending the answer may include the conditions for ending the answer. For details, refer to the detailed description of the conditions for ending the answer, and the repeated parts will not be repeated here. The single-item recommendation strategy can refer to the single-item recommendation strategy explained in S103, and the repeated parts will not be repeated here. Afterwards, it is judged whether the user has reached the end answering condition, if yes, the topic recommendation process is ended, if not, the next question is recommended for the user according to the single question recommendation strategy in a loop, until the end answering condition is met, and the topic recommendation process is ended.

Refer to Fig. 3, which is a user interaction interface diagram. Including, the topic practice button 31 recommends topics after being triggered. The process end button 32 is triggered to exit topic recommendation strategies (including entry-level recommendation strategies and single-item recommendation strategies). Recommended topic 33. The question answering area 34 is used for users to answer recommended questions. The next question button 35 is used to display the next recommended question after being triggered.

The above S101-S103 can obtain the user's current learning ability information by responding to the topic recommendation request for the target knowledge point initiated by the user for the first time. Here, the learning ability information is used to represent the user's ability to grasp the target knowledge point. After that, call the entry-level recommendation strategy in the strategy engine, and determine the initial recommendation topic for the user in the current recommendation process based on the learning ability information and the correct answer information of each candidate topic associated with the target knowledge point in the question bank. Here, the correct answer rate of the candidate topic itself (predicted based on each user's historical answering situation) can reflect the difficulty of the topic. Therefore, combining the user's learning ability and the correct answer rate of the candidate topic, it can be more reasonable for the user's initial recommendation and learning ability. Matching initial recommended topics. Afterwards, in response to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the policy engine, and judge whether the target user meets the upgrade conditions based on the user's answers to the recommended topics in the current process. Here, since the user can update the user's mastery of the target knowledge points based on the user's answers to the recommended questions, it is possible to formulate a next-step answer plan for the user. For example, users who are not good at answering questions can downgrade to practice, that is, practice relatively simple questions first, and gradually master the corresponding target knowledge points from easy to difficult. Another example is a user who has a very good answer to the questions, indicating that the current difficult questions are relatively easy for the user. At this time, upgrade exercises can be carried out for the user to improve learning efficiency.

Those skilled in the art can understand that in the above method of specific implementation, the writing order of each step does not mean a strict execution order and constitutes any limitation on the implementation process. The specific execution order of each step should be based on its function and possible The inner logic is OK.

Based on the same inventive concept, the embodiment of the present disclosure also provides a topic recommendation device corresponding to the topic recommendation method. Since the problem-solving principle of the device in the embodiment of the disclosure is similar to the above-mentioned topic recommendation method of the embodiment of the disclosure, the implementation of the device Reference can be made to the implementation of the method, and repeated descriptions will not be repeated.

Referring to FIG. 4 , it is a schematic diagram of a topic recommendation device provided by an embodiment of the present disclosure. The device includes: an information acquisition module 401 , a first topic recommendation module 402 and a second topic recommendation module 403 . in,

The information acquisition module 401 is configured to obtain the user's current learning ability information in response to the topic recommendation request for the target knowledge point initiated by the user for the first time; the learning ability information is used to represent the user's ability to grasp the target knowledge point ;

The first topic recommendation module 402 is used to call the entry-level recommendation strategy in the strategy engine, and determine the current recommendation process as The initial recommendation topic recommended by the user;

The second topic recommendation module 403 is used to respond to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the strategy engine, according to the user's answer to the recommended topic in the current recommendation process , judging whether the user satisfies the question leveling condition; according to the judging result, selecting the next question to be recommended for the user from the question bank.

In an optional implementation manner, the information acquisition module 401 is specifically configured to:

In the case of the user's historical learning data on the target knowledge point, according to the historical learning data of the user's historical answer accuracy rate on the target knowledge point, and the user's explanation content on the target knowledge point The proportion of the degraded learning duration of the user is determined, the target learning ability level of the user is determined, and the target learning ability level is used as the learning ability information; the degraded learning duration includes the user's fallback behavior duration for the explanation content, pause At least one of behavior duration and deceleration behavior duration;

If there is no historical learning data of the user, the default learning ability level is used as the learning ability information of the user.

In an optional implementation manner, the first topic recommendation module 402 is specifically used for:

If the learning ability level indicated by the learning ability information is the first level, and there is a candidate topic among the candidate topics whose correct answer rate is within the target correct rate range, then the correct answer rate is within the target correct rate range Among the candidate topics in , select any topic as the initial recommendation topic;

If the learning ability level indicated by the learning ability information is the first level, and the correct answer rates of the candidate questions are all less than the minimum value within the target correct rate range, then select the highest answer correct rate from the candidate questions The topic of is used as the initial recommendation topic;

If the learning ability level indicated by the learning ability information is the first level, and the correct answer rates of the candidate questions are all greater than the maximum value within the target correct rate range, then select the candidate questions with the lowest answer correct rate The topic of is used as the initial recommendation topic;

If the learning ability grade indicated by the learning ability information is the second grade, selecting the topic with the highest correct answer rate from each candidate topic as the initial recommendation topic;

Wherein, the learning ability corresponding to the first level is higher than the learning ability corresponding to the second level.

In an optional implementation manner, the second topic recommendation module 403 is specifically used for:

If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is less than the preset ratio threshold, determine said user satisfies the upgrade criteria; or,

If the user's first feedback information on the recommended topic is received, and the first feedback information indicates that the recommended topic is relatively simple, it is determined that the user meets the upgrade condition.

In an optional implementation manner, the second topic recommendation module 403 is also used to:

If the user's answer to the recommended question is incorrect, determine that the user meets the downgrading condition; or,

If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is greater than or equal to the preset ratio threshold, it is determined that the user satisfies the demotion condition; or,

If the second feedback information from the user on the recommended topic is received, and the second feedback information indicates that the recommended topic is relatively difficult, it is determined that the user meets the downgrade condition.

In an optional implementation manner, the topic recommendation device further includes an end recommendation judgment module 404, configured to:

According to the user's answers to the recommended questions, it is judged whether the user meets the conditions for ending the answer;

The second topic recommendation module is also used for:

When it is determined that the user has not met the condition for ending answering, it is determined whether the user meets the condition for leveling up the topic according to the user's answering condition for the recommended topic.

In an optional implementation manner, the end recommendation judging module 404 is configured to judge whether the user meets the end answering condition according to at least one of the following information:

Whether the practice duration of the user's topic exceeds the preset duration of practice;

Whether the number of continuous errors in the user's answers is greater than the set threshold;

Whether the continuous error duration of the user's answer is greater than the set duration;

Whether the user's mastery of the target knowledge point satisfies the condition that the knowledge point has been mastered.

For the description of the processing flow of each module in the topic recommendation apparatus and the interaction flow between the modules, reference may be made to the relevant description in the above-mentioned topic recommendation method embodiment, which will not be described in detail here.

Based on the same technical idea, the embodiment of the present disclosure also provides a computer device. Referring to FIG. 5, it is a schematic structural diagram of a computer device provided by an embodiment of the present disclosure, including:

Processor 51 , memory 52 and bus 53 . Wherein, the memory 52 stores machine-readable instructions executable by the processor 51, and the processor 51 is used to execute the machine-readable instructions stored in the memory 52. When the machine-readable instructions are executed by the processor 51, the processor 51 executes The following steps: S101: In response to the topic recommendation request for the target knowledge point initiated by the user for the first time, obtain the user's current learning ability information; the learning ability information is used to represent the user's ability to master the target knowledge point; S102: Call the strategy engine Introductory recommendation strategy, according to the learning ability information and the correct answer information of each candidate question associated with the target knowledge point in the question bank, determine the initial recommended topic recommended for the user in the current recommendation process; S103: respond to the completion of the initial recommended topic, and The current recommendation process is not over, enable the single-question recommendation strategy in the strategy engine, and judge whether the user meets the question leveling conditions according to the user's answers to the recommended questions in the current recommendation process; according to the judgment result, select from the question bank as the user's waiting list Recommended next topic.

Above-mentioned storer 52 comprises internal memory 521 and external memory 522; Internal memory 521 here is also called internal memory, is used for temporarily storing computing data in processor 51, and the data exchanged with external memory 522 such as hard disk, processor 51 communicates with external memory 521 through internal memory 521. The external memory 522 performs data exchange. When the computer device is running, the processor 51 communicates with the memory 52 through the bus 53, so that the processor 51 executes the execution instructions mentioned in the above method embodiments.

Embodiments of the present disclosure also provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the topic recommendation method described in the foregoing method embodiments are executed. Wherein, the storage medium may be a volatile or non-volatile computer-readable storage medium.

An embodiment of the present disclosure further provides a computer program product, including computer instructions, and when the computer instructions are executed by a processor, the steps of the above topic recommendation method are implemented. Among them, the computer program product can be any product that can realize the recommended method of the above topic, and some or all of the solutions in the computer program product that contribute to the existing technology can be implemented as a software product (such as a software development kit (Software Development Kit, SDK) ), the software product can be stored in a storage medium, and the computer instructions contained therein make relevant devices or processors execute some or all of the steps of the above topic recommendation method.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the device described above can refer to the corresponding process in the foregoing method embodiment, and details are not repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods may be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined. Or some features can be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces, and the indirect coupling or communication connection of devices or modules may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in each embodiment of the present disclosure may be integrated into one processing module, each module may exist separately physically, or two or more modules may be integrated into one module.

If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on this understanding, the technical solution of the embodiments of the present disclosure is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present disclosure. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disc and other media that can store program codes. .

Finally, it should be noted that: the above-mentioned embodiments are only the specific implementation modes of the present disclosure, and are used to illustrate the technical solutions of the embodiments of the present disclosure, rather than to limit them, and the protection scope of the present disclosure is not limited thereto, although The present disclosure has been described in detail with reference to the foregoing embodiments, and those of ordinary skill in the art should understand that: within the technical scope disclosed in the embodiments of the present disclosure, any person familiar with the technical field can still make reference to the foregoing embodiments. Modifications or changes can be easily imagined in the technical solutions, or equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the disclosure, and should be within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be defined by the protection scope of the claims.

Claims

A topic recommendation method, comprising:

In response to the topic recommendation request for the target knowledge point initiated by the user for the first time, the current learning ability information of the user is obtained; the learning ability information is used to represent the user's ability to grasp the target knowledge point;

Invoke the entry-level recommendation strategy in the strategy engine, and determine the initial recommendation topic recommended for the user in the current recommendation process according to the learning ability information and the correct answer information of each candidate topic associated with the target knowledge point in the question bank;

In response to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the policy engine, and judge whether the user meets the topic adjustment according to the user's answers to the recommended topics in the current recommendation process. Level conditions; according to the judgment result, select the next question to be recommended for the user from the question bank.
The method according to claim 1, wherein the acquisition of the user's current learning ability information in response to the topic recommendation request for the target knowledge point initiated by the user for the first time includes:

In the case of the user's historical learning data on the target knowledge point, according to the historical learning data of the user's historical answer accuracy rate on the target knowledge point, and the user's explanation content on the target knowledge point The proportion of the degraded learning duration of the user is determined, the target learning ability level of the user is determined, and the target learning ability level is used as the learning ability information; the degraded learning duration includes the user's fallback behavior duration for the explanation content, pause At least one of behavior duration and deceleration behavior duration;

If there is no historical learning data of the user, the default learning ability level is used as the learning ability information of the user.
The method according to claim 1, wherein, according to the learning ability information and the correct answer information of each candidate question associated with the target knowledge point in the question bank, determine the recommended one for the user in the current recommendation process Initial recommended topics, including:

If the learning ability level indicated by the learning ability information is the first level, and there is a candidate topic among the candidate topics whose correct answer rate is within the target correct rate range, then the correct answer rate is within the target correct rate range Among the candidate topics in , select any topic as the initial recommendation topic;

If the learning ability level indicated by the learning ability information is the first level, and the correct answer rates of the candidate questions are all less than the minimum value within the target correct rate range, then select the highest answer correct rate from the candidate questions The topic of is used as the initial recommendation topic;

If the learning ability level indicated by the learning ability information is the first level, and the correct answer rates of the candidate questions are all greater than the maximum value within the target correct rate range, then select the candidate questions with the lowest answer correct rate The topic of is used as the initial recommendation topic;

If the learning ability grade indicated by the learning ability information is the second grade, selecting the topic with the highest correct answer rate from each candidate topic as the initial recommendation topic;

Wherein, the learning ability corresponding to the first level is higher than the learning ability corresponding to the second level.
The method according to claim 1, wherein, according to the user's answers to the recommended topics in the current recommendation process, judging whether the user satisfies the topic adjustment condition includes:

If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is less than the preset ratio threshold, determine said user satisfies the upgrade criteria; or,

If the user's first feedback information on the recommended topic is received, and the first feedback information indicates that the recommended topic is relatively simple, it is determined that the user meets the upgrade condition.
The method according to claim 4, wherein, according to the user's answers to the recommended topics in the current recommendation process, judging whether the user satisfies the topic adjustment condition further includes:

If the user's answer to the recommended question is incorrect, determine that the user meets the downgrading condition; or,

If the user's answer to the recommended question is correct, and the ratio between the actual answering time of the recommended question and the maximum answering time corresponding to the recommended question is greater than or equal to the preset ratio threshold, it is determined that the user satisfies the demotion condition; or,

If the second feedback information from the user on the recommended topic is received, and the second feedback information indicates that the recommended topic is relatively difficult, it is determined that the user meets the downgrade condition.
The method according to claim 1, wherein, before judging whether the user satisfies the topic leveling condition according to the user's answer to the recommended topic in the current recommendation process, it also includes:

According to the user's answers to the recommended questions, it is judged whether the user meets the conditions for ending the answer;

According to the user's answers to the recommended topics in the current recommendation process, judging whether the user meets the topic adjustment conditions includes:

When it is determined that the user has not met the condition for ending answering, it is determined whether the user meets the condition for leveling up the topic according to the user's answering condition for the recommended topic.
The method according to claim 6, wherein, according to at least one of the following information, it is judged whether the user meets the condition of ending the answer:

Whether the practice duration of the user's topic exceeds the preset duration of practice;

Whether the number of continuous errors in the user's answers is greater than the set threshold;

Whether the continuous error duration of the user's answer is greater than the set duration;

Whether the user's mastery of the target knowledge point satisfies the condition that the knowledge point has been mastered.
A topic recommendation device, comprising:

An information acquisition module, configured to acquire current learning ability information of the user in response to a topic recommendation request for a target knowledge point initiated by the user for the first time; the learning ability information is used to represent the user's ability to grasp the target knowledge point;

The first topic recommendation module is used to call the entry-level recommendation strategy in the strategy engine, and determine the correct answer rate information of each candidate topic associated with the target knowledge point in the question bank according to the learning ability information and determine the current recommendation process for all Describe the initial recommendation topic recommended by the user;

The second topic recommendation module is used to respond to the completion of the initial recommended topic, and the current recommendation process is not over, enable the single-item recommendation strategy in the strategy engine, and according to the user's answer to the recommended topic in the current recommendation process, Judging whether the user satisfies the question leveling condition; according to the judgment result, selecting the next question to be recommended for the user from the question bank.
A computer device, comprising: a processor, a memory, and a bus, the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory through the bus , when the machine-readable instructions are executed by the processor, the steps of the topic recommendation method according to any one of claims 1 to 7 are executed.
A computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the steps of the topic recommendation method according to any one of claims 1 to 7 are executed.