A title recommendation method and device, computer equipment and a storage medium

Abstract

The present disclosure provides a question recommendation method and device, computer equipment and a storage medium, wherein the method comprises: obtaining an examination score of a target user, and determining an initial level corresponding to the target user based on the examination score; determining a target level of the target user; wherein the target level is higher than the initial level; different levels correspond to different question difficulties; determining a recommended question corresponding to the target user based on reference questions corresponding to each intermediate level between the initial level and the target level; wherein the intermediate level includes the target level.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a topic recommendation method, apparatus, computer equipment, and storage medium. Background Art

[0002] After students take an exam, teachers usually recommend questions to students based on their answers in order to strengthen their knowledge base.

[0003] In the prior art, for any given student, the system typically identifies reference students who have made similar mistakes based on the questions they failed to answer. These students then use their mistakes as recommended questions and recommend them to the student. However, different students have different learning goals. For example, if Student 1 wants to improve their score by 10 points, while Student 2 wants to improve their score by 5 points, the difficulty of the questions they need to master may vary. This method cannot recommend appropriate questions for students with different learning goals. Therefore, how to provide personalized question recommendations for each student has become an urgent problem. Summary of the Invention

[0004] The embodiments of the present disclosure at least provide a topic recommendation method, apparatus, computer device, and storage medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a topic recommendation method, comprising:

[0006] Obtaining an assessment score of a target user, and determining an initial level corresponding to the target user based on the assessment score;

[0007] Determining a target level for the target user; wherein the target level is higher than the initial level; different levels correspond to different levels of difficulty;

[0008] A recommended topic corresponding to the target user is determined based on reference topics corresponding to respective intermediate levels between the initial level and the target level, wherein the intermediate levels include the target level.

[0009] In one possible implementation, determining the initial level corresponding to the target user based on the assessment score includes:

[0010] Determine a target location area where the target user is located;

[0011] Based on the assessment score, an initial level of the target user in the target location area is determined.

[0012] In a possible implementation, determining the target level of the target user includes:

[0013] Determine the target level based on a preset target level determination rule and the initial level; or

[0014] In response to an instruction to set a target level, the target level is determined.

[0015] In one possible implementation, determining a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level includes:

[0016] For any reference topic of any intermediate level, determine the target similarity between each pre-stored candidate topic and the reference topic;

[0017] Based on the target similarity, a recommendation topic corresponding to the target user is determined.

[0018] In a possible implementation, for any reference topic of any intermediate level, determining the target similarity between each pre-stored candidate topic and the reference topic includes:

[0019] For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method;

[0020] The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

[0021] In a possible implementation, determining a recommended topic corresponding to the target user based on the target similarity includes:

[0022] Determine the weight of the recommended questions corresponding to each intermediate level; wherein the recommended question weight is used to represent the importance of the questions based on each intermediate level;

[0023] For any candidate topic, a recommendation score for the candidate topic is determined based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic;

[0024] Based on the recommendation score, a recommendation topic corresponding to the target user is determined.

[0025] In one possible implementation, determining the weight of the recommended questions corresponding to each intermediate level includes:

[0026] Determining the initial weight corresponding to each intermediate level, and determining a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted;

[0027] determining a first level among the intermediate levels that is lower than the reference level;

[0028] Based on the initial weight corresponding to the first level and the initial weight of the reference level, the weight of the reference level is determined; and the initial weights of other levels except the reference level are used as the weights of the other levels.

[0029] In a possible implementation, determining the reference level in the intermediate level includes:

[0030] Determine the historical initial level corresponding to the historical assessment score of the target user;

[0031] Determining a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user;

[0032] The reference level is determined based on the initial level corresponding to the target user and the progress vector.

[0033] In a possible implementation, determining the reference level based on the initial level corresponding to the target user and the progress vector includes:

[0034] Determining an average promotion speed of the target user based on each value in the progress vector; the average promotion speed is used to represent the difference between the initial level corresponding to the target user's next assessment score and the initial level corresponding to the current assessment score;

[0035] The reference level is determined based on the initial level corresponding to the target user and the average promotion speed.

[0036] In a second aspect, an embodiment of the present disclosure further provides a topic recommendation device, comprising:

[0037] An acquisition module, configured to acquire an assessment score of a target user and determine an initial level corresponding to the target user based on the assessment score;

[0038] A first determining module is configured to determine a target level for the target user; wherein the target level is higher than the initial level; and different levels correspond to different levels of difficulty of questions;

[0039] The second determining module is configured to determine a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level, wherein the intermediate levels include the target level.

[0040] In a possible implementation, when determining the initial level corresponding to the target user based on the assessment score, the acquisition module is configured to:

[0041] Determine a target location area where the target user is located;

[0042] Based on the assessment score, an initial level of the target user in the target location area is determined.

[0043] In a possible implementation, the first determining module, when determining the target level of the target user, is configured to:

[0044] Determine the target level based on a preset target level determination rule and the initial level; or

[0045] In response to an instruction to set a target level, the target level is determined.

[0046] In one possible implementation, the second determining module, when determining the recommended topic corresponding to the target user based on the reference topics corresponding to the intermediate levels between the initial level and the target level, is configured to:

[0047] For any reference topic of any intermediate level, determine the target similarity between each pre-stored candidate topic and the reference topic;

[0048] Based on the target similarity, a recommendation topic corresponding to the target user is determined.

[0049] In one possible implementation, for any reference topic of any intermediate level, the second determination module, when determining the target similarity between each pre-stored candidate topic and the reference topic, is configured to:

[0050] For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method;

[0051] The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

[0052] In a possible implementation, the second determination module, when determining the recommended topic corresponding to the target user based on the target similarity, is configured to:

[0053] Determine the weight of the recommended questions corresponding to each intermediate level; wherein the recommended question weight is used to represent the importance of the questions based on each intermediate level;

[0054] For any candidate topic, a recommendation score for the candidate topic is determined based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic;

[0055] Based on the recommendation score, a recommendation topic corresponding to the target user is determined.

[0056] In a possible implementation, the second determination module, when determining the weight of the recommended questions corresponding to each intermediate level, is configured to:

[0057] Determining the initial weight corresponding to each intermediate level, and determining a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted;

[0058] determining a first level among the intermediate levels that is lower than the reference level;

[0059] Based on the initial weight corresponding to the first level and the initial weight of the reference level, the weight of the reference level is determined; and the initial weights of other levels except the reference level are used as the weights of the other levels.

[0060] In a possible implementation manner, the second determining module, when determining the reference level in the intermediate level, is configured to:

[0061] Determine the historical initial level corresponding to the historical assessment score of the target user;

[0062] Determining a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user;

[0063] The reference level is determined based on the initial level corresponding to the target user and the progress vector.

[0064] In a possible implementation, the second determining module, when determining the reference level based on the initial level corresponding to the target user and the progress vector, is configured to:

[0065] Determining an average promotion speed of the target user based on each value in the progress vector; the average promotion speed is used to represent the difference between the initial level corresponding to the target user's next assessment score and the initial level corresponding to the current assessment score;

[0066] The reference level is determined based on the initial level corresponding to the target user and the average promotion speed.

[0067] In a third aspect, an embodiment of the present disclosure further provides a computer device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned first aspect or any possible implementation of the first aspect are performed.

[0068] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned first aspect or any possible implementation of the first aspect are executed.

[0069] The topic recommendation method, apparatus, computer equipment and storage medium provided by the embodiments of the present disclosure can determine the initial level and target level of the target user, wherein the initial level can represent the current level of the target user, and the target level can represent the goal that the target user expects to achieve. Furthermore, the recommended topics corresponding to the target user can be determined based on the reference topics corresponding to the intermediate levels between the initial level and the target level, that is, the recommended topics can be determined according to the current level and learning goals of the target user. Since the target levels of different users are different, that is, the learning goals corresponding to different users are different, this method can match the learning goals of different users to achieve the purpose of recommending personalized topics for users. In addition, when determining the recommended topics, the reference topics corresponding to the various levels between the initial level and the target level are combined, thereby ensuring that the determined recommended topics can cover various levels from the initial level to the target level, thereby achieving step-by-step learning and improving learning effects.

[0070] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings required for use in the embodiments. The drawings herein are incorporated into and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and, together with the specification, are used to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without inventive effort.

[0072] Figure 1 A flowchart of a topic recommendation method provided by an embodiment of the present disclosure is shown;

[0073] Figure 2 A flowchart of a method for determining recommended topics provided by an embodiment of the present disclosure is shown;

[0074] Figure 3 A flowchart showing a specific method for determining recommended topics provided by an embodiment of the present disclosure is shown;

[0075] Figure 4 A flow chart of a method for determining a weight of a topic recommendation provided by an embodiment of the present disclosure is shown;

[0076] Figure 5 A flowchart of a method for determining a reference level provided by an embodiment of the present disclosure is shown;

[0077] Figure 6 A schematic diagram of the architecture of a topic recommendation device provided by an embodiment of the present disclosure is shown;

[0078] Figure 7 A schematic structural diagram of a computer device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0079] 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 drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the disclosure for which protection is sought, 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 work are within the scope of protection of the present disclosure.

[0080] After students take an exam, teachers usually recommend questions to students based on their answers in order to strengthen their training on their weak points in knowledge. For example, teachers can find similar questions based on the questions that students got wrong and use them to train students.

[0081] In related art, for any given student, the system typically identifies reference students who have made similar mistakes based on the student's mistakes. These students then use their other mistakes as recommended questions and recommend them to the student. However, the level of a reference student, determined based on only one mistake, varies widely, potentially exceeding or exceeding the student's level. A reference student's mistakes aren't necessarily the student's weaknesses. Even if the reference student's level is comparable to the student's, simply recommending questions from students of the same level won't significantly improve the student's academic performance.

[0082] Furthermore, different students have different learning goals. For example, Student 1 wants to improve by 10 points, while Student 2 wants to improve by 5 points. Therefore, the difficulty of the questions they need to master will vary. Student 1 will require more difficult questions than Student 2. This method cannot recommend appropriate questions for students with different learning goals. Therefore, for students with different learning goals, it is necessary to recommend questions of appropriate difficulty based on their respective learning goals to better improve learning outcomes.

[0083] Based on the above research, the present disclosure provides a topic recommendation method, device, computer equipment and storage medium, which can determine the initial level and target level of the target user, the initial level can represent the current level of the target user, and the target level can represent the goal that the target user expects to achieve. Furthermore, based on the reference topics corresponding to each intermediate level between the initial level and the target level, the recommended topics corresponding to the target user can be determined, that is, the recommended topics can be determined according to the current level and learning goals of the target user. Since the target levels of different users are different, that is, the learning goals corresponding to different users are different, this method can match the learning goals of different users to achieve the purpose of recommending personalized topics for users; in addition, when determining the recommended topics, the reference topics corresponding to each level between the initial level and the target level are combined, thereby ensuring that the determined recommended topics can cover each level from the initial level to the target level, thereby achieving step-by-step learning and improving learning effects.

[0084] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0085] The term "and / or" herein simply describes an association relationship, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, and the existence of B alone. In addition, the term "at least one" herein refers to any combination of at least two of any one or more of a plurality of items. For example, "at least one of A, B, and C" can represent any one or more elements selected from the set consisting of A, B, and C.

[0086] It is understandable that before using the technical solutions disclosed in the various embodiments of this disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved in this disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0087] For example, in response to a user's active request, a prompt message is sent to the user to clearly inform the user that the operation requested will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the electronic device, application, server, storage medium, or other software or hardware that performs the operations of the disclosed technical solution based on the prompt message.

[0088] As an optional but non-limiting implementation, in response to receiving a user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. Furthermore, the pop-up window may also contain a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0089] It is understandable that the above notification and user authorization process are merely illustrative and do not limit the implementation of the present disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0090] To facilitate understanding of this embodiment, a detailed description of a topic recommendation method disclosed in this embodiment is first provided. The topic recommendation method provided in this embodiment is generally executed by a computer device with certain computing capabilities, such as a client or server. Exemplary client devices include personal computers, smartphones, and tablet computers. In some possible implementations, the topic recommendation method can be implemented by a processor invoking computer-readable instructions stored in a memory.

[0091] See also Figure 1 FIG. 1 is a flowchart of a method for recommending a topic according to an embodiment of the present disclosure, wherein the method includes steps 101 to 103, wherein:

[0092] Step 101: Obtain an assessment score of a target user, and determine an initial level corresponding to the target user based on the assessment score;

[0093] Step 102: Determine the target level of the target user; wherein the target level is higher than the initial level; different levels correspond to different levels of difficulty;

[0094] Step 103: Determine a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level; wherein the intermediate levels include the target level.

[0095] The following are detailed instructions for the above steps:

[0096] For step 101,

[0097] Among them, the assessment score is the assessment score of the current examination of the target user (also referred to as the current assessment score). If the assessment score is the assessment score of the current examination of any subject, then the initial level is the initial level of the target user in the subject.

[0098] In a possible implementation, when determining the initial level corresponding to the target user based on the assessment score, the levels corresponding to different score intervals can be determined in advance, and then the level corresponding to the target score interval in which the target user's assessment score is located can be used as the initial level corresponding to the target user.

[0099] For example, it can be pre-set that 0 to 10 points correspond to level 1, 11 to 20 points correspond to level 2, 21 to 30 points correspond to level 3... and so on. If the assessment score of the target user is 25 points, the target score range of the assessment score is 21 to 30, and the level corresponding to the target score range is level 3, then level 3 will be used as the initial level corresponding to the target user.

[0100] In another possible implementation, when determining the initial level corresponding to the target user based on the assessment score, the assessment scores of each user can be first sorted (from large to small) to obtain the score ranking corresponding to each user, thereby determining the target score ranking of the target user; then, the correspondence between the predetermined score ranking interval and the level is obtained, and based on the correspondence, the target score ranking interval in which the score ranking of the target user is located is determined; finally, the level corresponding to the target score ranking interval is used as the initial level corresponding to the target user.

[0101] For example, it can be preset that the ranking between 1% and 10% is level 1, the ranking between 11% and 20% is level 2, and the ranking between 21% and 30% is level 3. If the target user's assessment score is ranked in the range of 11% to 20%, the target user's initial level is level 3.

[0102] Because different regions have varying competitive pressures and requirements for grades and question difficulty, recommending questions by region can better personalize the recommendations for users. Therefore, in one possible implementation, when determining the initial level corresponding to the target user based on the assessment score, the target location region of the target user can be first determined, and then, based on the assessment score, the initial level of the target user in that target location region can be determined.

[0103] Specifically, different location areas (such as different provinces and cities) can set their own initial level division rules (such as the above-mentioned method of dividing the initial level according to the score interval, or the above-mentioned method of dividing the initial level according to the score ranking interval). When determining the target location area where the target user is located, the target location area can be selected or input by the user, or the target location area where the user is located can be determined by a positioning device with the user's permission. The positioning device can exemplarily be a global positioning system (GPS).

[0104] Using this method, users can be divided according to their respective location areas, and the initial level of each user in their location area can be determined, so that topics can be recommended to target users based on users of various levels in the same location area as the target users, thereby improving the topic recommendation effect and realizing regional personalized topic recommendations.

[0105] For step 102,

[0106] Among them, the target level can represent the level that the user is expected to reach. Therefore, the target level should be greater than the initial level. Since higher levels correspond to users with higher assessment scores, the proportion of more difficult questions in the questions corresponding to higher levels is often higher than that of lower levels.

[0107] The following are three specific methods for determining the target level of the target user:

[0108] Method 1: The target level may be determined based on a preset target level determination rule and the initial level.

[0109] Specifically, the target level rule may be to add the initial level to the preset level to obtain the target level. Exemplarily, the target level determination rule may be: the initial level + 5 is the target level. If the initial level of the target user is level 3, then the target level is 3+5=level 8.

[0110] Method 2: The target level may be determined in response to an instruction to set the target level.

[0111] Specifically, the input target level may be determined by a user input operation, or a plurality of target levels to be selected may be displayed, and the triggered target level may be determined in response to a triggering operation on any target level to be selected.

[0112] Method three: the target level may be determined in response to an instruction for setting the target level, and the target level may be determined based on a preset target level determination rule and the initial level if there is no response (or no generation) to an instruction for setting the target level.

[0113] With this method, the target level can be manually set by the user and / or automatically determined based on the initial level, so as to subsequently recommend topics to the user for different learning goals (ie, the target levels).

[0114] For step 103,

[0115] Here, the intermediate levels between the initial level and the target level include the initial level and the target level. For example, if the initial level is level 1 and the target level is level 3, the intermediate levels are level 1, level 2, and level 3. The candidate questions may be questions in a pre-established question bank.

[0116] The reference topics may include historical incorrect questions answered by users of each intermediate level, and the historical incorrect questions may include incorrect questions answered by the users of the intermediate level in at least one previous exam (including the current exam). The users of the intermediate level include the target user and other users other than the target user. Since the levels of the other users are higher than or equal to the target user, it indicates that the current learning level of the other users is likely higher than or equal to that of the target user. Therefore, the incorrect questions answered by the other users are likely to be the target user's weak points in knowledge. Therefore, the recommended topics can be determined with reference to the incorrect questions answered by the other users. In addition, determining the recommended topics based on the incorrect questions answered by the other users also expands the recommended scope of the recommended topics, thereby determining a richer range of recommended topics (covering more knowledge points).

[0117] In a possible implementation manner, when determining a recommended topic corresponding to the target user from a plurality of pre-stored candidate topics based on reference topics corresponding to various intermediate levels between the initial level and the target level, as follows: Figure 2 As shown, the following steps 201 to 202 may be performed, specifically as follows:

[0118] Step 201: For any reference topic of any intermediate level, determine the target similarity between each pre-stored candidate topic and the reference topic.

[0119] In one possible implementation, the initial similarity between any candidate topic and the reference topic can be determined based on at least one similarity determination method; then the initial similarities are weighted and summed according to preset weights to obtain the target similarity between the candidate topic and the reference topic, or any initial similarity can be used as the target similarity between the candidate topic and the reference topic.

[0120] The calculation method of the initial similarity can be any one of the following methods:

[0121] Method 1: Determine the first knowledge point corresponding to the candidate question and the second knowledge point corresponding to the reference question, and determine the initial similarity between the candidate question and the reference question based on the first knowledge point and the second knowledge point.

[0122] Specifically, any candidate question can correspond to at least one first knowledge point, and any reference question can correspond to at least one second knowledge point. The knowledge points corresponding to each question (including the candidate question and the reference question) can be manually set, or can be determined based on a pre-trained neural network (such as a natural language understanding model) after analyzing the stem and / or problem-solving steps of each question, which is not limited here. After determining the first knowledge point and the second knowledge point, the number of each of the first knowledge point and the second knowledge point can be counted.

[0123] In a possible implementation, when determining the initial similarity between the candidate question and the reference question based on the first knowledge point and the second knowledge point, the identical knowledge points between the first knowledge point and the second knowledge point can be first determined, and the number of the first knowledge points can be compared with the number of the second knowledge points to determine the target knowledge points with a larger number (when the number of the first knowledge points is the same as the number of the second knowledge points, the first knowledge point or the second knowledge point can be used as the target knowledge point), and then the number of the identical knowledge points can be divided by the number of the target knowledge points to obtain the initial similarity. The calculation process is shown in the following formula:

[0124]

[0125] Among them, SamePointsSimilarity represents the initial similarity, pointList i Indicates the first knowledge point corresponding to the candidate question, pointList j Indicates the second knowledge point corresponding to the reference question.

[0126] For example, if the first knowledge points corresponding to the candidate question include: knowledge point 1, knowledge point 2, knowledge point 3, and the second knowledge points corresponding to the reference question include: knowledge point 2, knowledge point 3, knowledge point 4, and knowledge point 5, then the identical knowledge points between the first knowledge point and the second knowledge point are: knowledge point 2 and knowledge point 3, the number of identical knowledge points is 2, and the target knowledge point with the larger number between the first knowledge point (number is 3) and the second knowledge point (number is 4) is the second knowledge point, then the number of identical knowledge points 2 is divided by the number of target knowledge points 4, and the initial similarity is 0.5.

[0127] Method 2: Determine the first keyword corresponding to the candidate title and the second keyword corresponding to the reference title, and determine the initial similarity between the candidate title and the reference title based on the first keyword and the second keyword.

[0128] Specifically, the candidate title and the reference title may be segmented first, and then the segmentation results may be subjected to keyword recognition to determine the first keyword and the second keyword. Here, when performing keyword recognition on the segmentation results, multiple pre-stored target keywords may be matched with the candidate title and the reference title respectively to determine the first keyword and the second keyword. The target keyword is determined by counting the words in the stored multiple candidate titles. For example, a word whose number of occurrences exceeds a preset number may be used as the target keyword. After determining the word whose number of occurrences exceeds the preset number, it may be further manually screened, and the screened-out words may be used as the target keyword.

[0129] Then, when determining the initial similarity between the candidate title and the reference title based on the first keyword and the second keyword, you can first determine the number of identical keywords in the first keyword and the second keyword, and determine the number of non-repeating keywords in the first keyword and the second keyword (that is, the union of the first keyword and the second keyword); then divide the number of identical keywords by the number of non-repeating keywords to obtain the initial similarity.

[0130] For example, if the first keyword corresponding to the candidate title includes: keyword 1, keyword 2, keyword 3, and the second keyword corresponding to the reference title includes: keyword 3, keyword 4, then the identical keyword between the first keyword and the second keyword is: keyword 3, and the number of identical keywords is 1. The non-repeated keywords between the first keyword and the second keyword are: keyword 1, keyword 2, keyword 3, keyword 4, and the number of non-repeated keywords is 4. Then, divide the number of identical keywords 1 by the number of non-repeated keywords 4, and the initial similarity is 0.25.

[0131] Method three: determining the first text information of the candidate topic and the second text information of the reference topic, and determining the initial similarity between the candidate topic and the reference topic based on the first text information and the second text information.

[0132] Specifically, the longest identical text information in the first text information and the second text information can be determined. The longest identical text information is: the longest identical and continuous text information in the first text information and the second text information, such as: the first text information is oabcdefg, the second text information is abbbefghi, then the longest identical text information is efg.

[0133] Furthermore, the length of the first text message and the length of the second text message are compared to determine the first target text message that is shorter between the first text message and the second text message, and then the length of the longest identical text message is divided by the length of the first target text message. The calculation process is shown in the following formula:

[0134]

[0135] Wherein, PartialSimilarity represents the initial similarity, str1 represents the first text information of the candidate question, and str2 represents the second text information of the reference question.

[0136] Continuing with the above example, the length of the longest identical text message efg is 3, the length of the first text message is 8, and the length of the second text message is 9. The first target text message with a shorter length is the first text message. Then, the length of the longest identical text message 3 is divided by the length of the first text message 8 to obtain the initial similarity 0.375.

[0137] Method 4: First determine the first text information of the candidate question, the second text information of the reference question, and the edit distance between the first text information and the second text information; then determine the initial similarity between the candidate question and the reference question based on the first text information, the second text information, and the edit distance.

[0138] Specifically, the edit distance is used to represent the minimum number of edits required to convert the first text information into the second text information. Exemplarily, the edit distance algorithm (Levenshtein Distance algorithm, also known as EditDistance algorithm) can be used for calculation. Among them, the editing operations corresponding to the edit distance may include character replacement, character insertion, and character deletion. Then, the length of the first text information and the length of the second text information are compared to determine the second target text information with the longer length between the first text information and the second text information. Finally, the ratio of the edit distance to the second target text information is subtracted from 1 to obtain the initial similarity. The calculation process is shown in the following formula:

[0139]

[0140] Wherein, Similarity represents the initial similarity, str1 represents the first text information of the candidate question, and str2 represents the second text information of the reference question.

[0141] For example, if the edit distance between the first text information and the second text information is 5, the length of the first text information is 10, and the length of the second text information is 8, the first text information with a longer length is used as the second target text information, and the edit distance 5 is divided by the length of the first text information 10 to obtain the initial similarity 0.5.

[0142] Here, other similarity determination methods may also be applied to this embodiment and are not limited here.

[0143] Then, after calculating at least one initial similarity, the initial similarities are weighted and summed according to preset weights to obtain the target similarity between the candidate question and the reference question. Taking the four initial similarities as an example, the following formula can be used for calculation:

[0144] score=w1·Score SamePoints +w2·Score JaccardSimilarity +w3·Score Similarity +w4·Score partialSimilarity

[0145] Among them, score represents the target similarity, Score SamePoints 、Score JaccardSimilarity Score Similarity 、Score partialSimilarity They represent four initial similarities respectively, and w1, w2, w3, and w4 represent the preset weights corresponding to the initial similarities respectively.

[0146] Step 202: Determine a recommendation topic corresponding to the target user based on the target similarity.

[0147] Here, by determining the recommended topics based on the target similarity, similar topics to topics that are easy to make mistakes for users of each intermediate level can be screened out from the candidate topics, so that the similar topics can be recommended to the target user to improve the learning effect of the target user.

[0148] In a possible implementation, when determining the recommended topics corresponding to the target user based on the target similarities, the target similarities can be sorted according to numerical values, and the candidate topics corresponding to the first preset number of target similarities with the largest numerical values ​​are used as the recommended topics corresponding to the target user.

[0149] Exemplarily, the target similarities of candidate topics 1 to 5 are 5, 3, 4, 2, and 1 respectively. After sorting the target similarities of candidate topics 1 to 5 from large to small, they are 5, 4, 3, 2, and 1 respectively. If the first preset number is 2, candidate topic 1 corresponding to target similarity 5 and candidate topic 3 corresponding to target similarity 4 are used as recommended topics corresponding to the target user.

[0150] However, since different users have different learning abilities and progress rates, the learning ability of the target user can also be estimated based on the user's historical assessment scores, and personalized topic recommendations can be made for the target user. Specifically, in another possible implementation, when determining the recommended topic corresponding to the target user based on the target similarity, steps 301 to 303 can be performed as follows:

[0151] Step 301: Determine the weight of recommended questions for each intermediate level. The weight represents the importance of questions at each intermediate level. By adjusting the weight of recommended questions for each intermediate level, the proportion of questions of each difficulty level can be adjusted, thereby achieving personalized recommendation of questions for the target user.

[0152] In one possible implementation, the weight of the recommended questions corresponding to each intermediate level can be set manually. Exemplarily, in response to the instruction to set the weight of the recommended questions of the target user terminal, the weight of the recommended questions corresponding to each intermediate level can be determined. The target user terminal can include the user terminal of the teacher of the target user, the user terminal of the parent of the target user, and the user terminal of the target user. Using this method, the weight of the recommended questions can be set manually for the target user based on the actual learning situation of the target user, so as to achieve personalized recommendation of questions for each user. Here, in the case where the weight of the recommended questions corresponding to each intermediate level is not manually set, the weight of the recommended questions corresponding to each intermediate level can be defaulted to the default weight, such as the weight corresponding to each intermediate level can be defaulted to 1.

[0153] In another possible implementation, the weight of the recommended questions corresponding to each intermediate level may be determined based on the historical assessment scores of the target user. Specifically, the following steps 401 to 403 may be performed:

[0154] Step 401: Determine the initial weight corresponding to each intermediate level, and determine a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted.

[0155] In one possible implementation, when determining the initial weight corresponding to each intermediate level, the initial weight may be a manually set weight, or the initial weight may be a weight calculated according to a preset algorithm. For example, the initial weight corresponding to each intermediate level may be 1 divided by the number of intermediate levels. For example, if the number of intermediate levels is 5, the initial weight corresponding to each intermediate level is 1 ÷ 5 = 0.2.

[0156] In actual applications, the initial weights can be manually and dynamically adjusted according to the actual situation of the target user, so as to adjust the proportion of similar topics in the recommended topics to topics of each level.

[0157] In a possible implementation, when determining the reference level in the intermediate level, the following steps 501 to 503 may be performed:

[0158] Step 501: Determine the historical initial level corresponding to the historical assessment score of the target user.

[0159] The historical assessment scores are the assessment scores of the target user's most recent N (N is a positive integer) examinations, wherein the value of N can be manually changed. In one possible implementation, the value of N can be changed in response to a change instruction from the target user. Here, if the value of N is not set, the value of N can be determined to be a default value, such as a default value of 3.

[0160] Here, the method used to determine the historical initial level corresponding to the historical assessment score of the target user is the same as the method used to determine the initial level corresponding to the target user based on the assessment score in step 101, and will not be repeated here.

[0161] Step 502: Determine a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user.

[0162] In a possible implementation, the progress vector may be used to represent the difference in historical initial grades corresponding to two adjacent historical assessment scores in the target user's most recent N exams (hereinafter referred to as grade difference).

[0163] Specifically, when determining the target user's progress vector, for any historical assessment score, the level difference between the historical assessment score and the previous historical assessment score can be calculated first. Similarly, the level difference between each two adjacent historical assessment scores can be calculated, and then the target user's progress vector can be determined based on each calculated level difference. It should be noted that if the historical initial level corresponding to the historical assessment score of the next exam is higher than the historical initial level of the historical assessment score of the previous exam, the level difference between the historical assessment scores of the next exam and the previous exam is determined to be 0.

[0164] For example, the historical initial levels corresponding to the historical assessment scores of the target user's most recent five exams are 5, 10, 12, 11, and 13, respectively. The level differences corresponding to each two adjacent historical assessment scores are 5, 2, 0, and 2, respectively. The progress vector of the target user is (5, 2, 0, 2).

[0165] Step 503: Determine the reference level based on the initial level corresponding to the target user and the progress vector.

[0166] Here, since the reference level is the level to which the user may be promoted, the reference level can be estimated based on the level that the user improves in each exam.

[0167] Therefore, in one possible implementation, when determining the reference level based on the target user's initial level and the progress vector, the target user's average promotion speed can be first determined based on the values ​​in the progress vector (i.e., the aforementioned level differences); and then the reference level can be determined based on the target user's initial level and the average promotion speed. The average promotion speed represents the difference between the target user's initial level corresponding to the next assessment score and the initial level corresponding to the current assessment score.

[0168] Specifically, the average promotion speed can be obtained by accumulating the values ​​in the progress vector and then dividing the accumulated result by the number of values ​​in the progress vector. The average promotion speed calculated in this way represents the average level of improvement of the user in multiple historical exams. Therefore, the average promotion speed can reflect the user's learning progress ability. The calculation formula of the average promotion speed can be as follows:

[0169]

[0170] Wherein, candidateLevel represents the reference level, i represents the number of the level difference, S i represents the level difference, and v represents the number of the level differences.

[0171] The average advancement rate is then added to the current assessment score to obtain the reference level. Since the reference level is calculated based on the average advancement rate, i.e., the reference level is the most suitable level for the user to improve based on their learning and progress abilities, more questions of the reference level can be recommended to improve the learning outcomes of the target user.

[0172] Step 402: Determine a first level among the intermediate levels that is lower than the reference level.

[0173] Specifically, the first level may be levels from a preset level to the upper level of the reference level. For example, if the preset level is 2 and the reference level is 5, the first level is 2, 3, and 4.

[0174] Step 403: Determine the weight of the reference level based on the initial weight corresponding to the first level and the initial weight of the reference level; and use the initial weights of other levels except the reference level as the weights of the other levels.

[0175] Specifically, when determining the weight of the reference level, the initial weight corresponding to each first level can be added to the initial weight of the reference level to obtain the weight of the reference level. For example, if the initial weights corresponding to the first levels are 1, 2, and 3, respectively, then the weight of the reference level is 1+2+3=6.

[0176] Alternatively, the adjustment coefficient corresponding to each level (including the first level and the reference level) can be set in advance, the initial weight corresponding to each first level can be multiplied by the adjustment coefficient corresponding to each first level, and the initial weight of the reference level can be multiplied by the adjustment coefficient of the reference level. Then, the multiplication results corresponding to each first level and the multiplication results corresponding to the reference level can be added to obtain the inference weight of the reference level.

[0177] For example, taking the preset level as 2 as an example, the calculation formula of the recommended question weight is as follows:

[0178]

[0179] Among them, w candidateLevel represents the weight of the recommended question, candidateLevel represents the reference level, i represents the number of each level (including the first level and the reference level), represents the adjustment coefficient, w i represents the initial weight.

[0180] By adopting this method, the weight of the recommended topics of the reference level can be increased, so that when determining the recommended topics subsequently, more topics of the reference level can be recommended to the target user.

[0181] When determining the weights of the other levels of inference, the other levels can be determined first. The other levels are the levels in the intermediate levels except the reference level. For example, if the initial level is level 1 and the target level is level 5, then the intermediate levels are level 1, level 2, level 3, level 4, and level 5. If the reference level is level 3, then the other levels are level 1, level 2, level 4, and level 5.

[0182] Then, for any other level, the initial weight of the other level can be used as the weight of the recommended question for the other level, or, when the adjustment coefficient corresponding to each level is set, the initial weight of the other level can be multiplied by the adjustment coefficient corresponding to the other level to obtain the weight of the recommended question for the other level.

[0183] Here, since the weights for the reference level have been redefined compared to the initial weights for the reference level, the weights for the intermediate levels need to be normalized. Specifically, for any intermediate level, the weight for that level can be divided by the sum of the weights for all intermediate levels to obtain the normalized weight for that level.

[0184] Step 302: For any candidate topic, determine the recommendation score of the candidate topic based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic.

[0185] Specifically, for any candidate topic, the target similarity between the candidate topic and any reference topic and the topic weight corresponding to the intermediate level of the reference topic can be multiplied to obtain at least one recommendation score for the candidate topic. Similarly, multiple recommendation scores for each candidate topic can be obtained. For example, if there are 5 reference topics, there are 5 target similarities between any candidate topic and any reference topic. After multiplying these 5 target similarities with the corresponding topic weights, 5 recommendation scores corresponding to the candidate topic can be obtained. The formula for the recommendation score can be as follows:

[0186] w i s i =w i [score i,1 ,score i,2 ......score i,n ]

[0187] Among them, w i Indicates the weight of the inference question corresponding to each intermediate level, i indicates the number of each intermediate level, s i score i,1 ,score i,2 ......score i,n Represents the similarity of each target.

[0188] Here, since the recommendation score is determined by combining the similarity with the easy-to-make mistakes of users at each intermediate level (i.e., the target similarity) and the learning ability of the target user (i.e., the weight of the recommended questions), the recommended questions subsequently determined based on the recommendation score are more suitable for the user's learning and training.

[0189] Step 303: Determine a recommended topic corresponding to the target user based on the recommendation score.

[0190] In a possible implementation, the recommendation scores may be sorted according to numerical values, and the target candidate topics corresponding to the second preset number of recommendation scores with the highest numerical values ​​are used as the recommended topics.

[0191] Here, since any candidate topic corresponds to multiple recommendation scores, there may be multiple target candidate topics. Therefore, after determining the target candidate topics, it is possible to determine whether there are repeated topics among the target candidate topics. If the target candidate topics include repeated topics, continue to obtain the target candidate topics with the highest values ​​according to the sorted recommendation scores until the number of non-duplicate target candidate topics reaches the second preset number.

[0192] For example, if the recommendation scores of the candidate questions are sorted from largest to smallest as follows: Candidate question 1, 99 points, Candidate question 2, 98 points, Candidate question 2, 97 points, Candidate question 3, 96 points, Candidate question 4, 95 points. If the second preset number is 3, then the three highest recommendation scores are Candidate question 1, 99 points, Candidate question 2, 98 points, and Candidate question 2, 97 points. Among them, 98 points and 97 points are both the recommendation scores of Candidate question 2. Therefore, according to this sorting, Candidate question 3 with the highest recommendation score can be obtained as the recommended question, and finally Candidate question 1, Candidate question 2, and Candidate question 3 can be used as the recommended question.

[0193] The topic recommendation method provided by the embodiment of the present disclosure can determine the initial level and target level of the target user. The initial level can represent the current level of the target user, and the target level can represent the goal that the target user expects to achieve. Furthermore, the recommended topic corresponding to the target user can be determined based on the reference topics corresponding to each intermediate level between the initial level and the target level. That is, the recommended topic can be determined according to the current level and learning goal of the target user. Since the target levels of different users are different, that is, the learning goals corresponding to different users are different, this method can match the learning goals of different users to achieve the purpose of recommending personalized topics for users. In addition, when determining the recommended topic, the reference topics corresponding to each level between the initial level and the target level are combined, thereby ensuring that the determined recommended topic can cover each level from the initial level to the target level, thereby achieving step-by-step learning and improving learning effects.

[0194] Those skilled in the art will understand that in the above-mentioned method of the specific implementation method, the writing order of each step does not mean a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of each step should be determined by its function and possible internal logic.

[0195] 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 principle of solving the problem by the device in the embodiment of the present disclosure is similar to the above-mentioned topic recommendation method in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0196] Reference Figure 6 FIG. 1 is a schematic diagram of the architecture of a topic recommendation device provided by an embodiment of the present disclosure, wherein the device includes: an acquisition module 601, a first determination module 602, and a second determination module 603; wherein,

[0197] An acquisition module 601 is configured to acquire an assessment score of a target user and determine an initial level corresponding to the target user based on the assessment score;

[0198] The first determination module 602 is configured to determine the target level of the target user; wherein the target level is higher than the initial level; different levels correspond to different levels of difficulty;

[0199] The second determining module 603 is configured to determine a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level, wherein the intermediate levels include the target level.

[0200] In a possible implementation, when determining the initial level corresponding to the target user based on the assessment score, the acquisition module 601 is configured to:

[0201] Determine a target location area where the target user is located;

[0202] Based on the assessment score, an initial level of the target user in the target location area is determined.

[0203] In a possible implementation, the first determining module 602, when determining the target level of the target user, is configured to:

[0204] Determine the target level based on a preset target level determination rule and the initial level; or

[0205] In response to an instruction to set a target level, the target level is determined.

[0206] In one possible implementation, the second determining module 603, when determining the recommended topic corresponding to the target user based on the reference topics corresponding to the intermediate levels between the initial level and the target level, is configured to:

[0207] For any reference topic of any intermediate level, determine the target similarity between each pre-stored candidate topic and the reference topic;

[0208] Based on the target similarity, a recommendation topic corresponding to the target user is determined.

[0209] In one possible implementation, for any reference topic of any intermediate level, the second determination module 603, when determining the target similarity between each pre-stored candidate topic and the reference topic, is configured to:

[0210] For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method;

[0211] The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

[0212] In a possible implementation, the second determining module 603, when determining the recommended topic corresponding to the target user based on the target similarity, is configured to:

[0213] Determine the weight of the recommended questions corresponding to each intermediate level; wherein the recommended question weight is used to represent the importance of the questions based on each intermediate level;

[0214] For any candidate topic, a recommendation score for the candidate topic is determined based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic;

[0215] Based on the recommendation score, a recommendation topic corresponding to the target user is determined.

[0216] In a possible implementation, the second determining module 603, when determining the weight of the recommended questions corresponding to each intermediate level, is configured to:

[0217] Determining the initial weight corresponding to each intermediate level, and determining a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted;

[0218] determining a first level among the intermediate levels that is lower than the reference level;

[0219] Based on the initial weight corresponding to the first level and the initial weight of the reference level, the weight of the reference level is determined; and the initial weights of other levels except the reference level are used as the weights of the other levels.

[0220] In a possible implementation, the second determining module 603, when determining the reference level in the intermediate level, is configured to:

[0221] Determine the historical initial level corresponding to the historical assessment score of the target user;

[0222] Determining a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user;

[0223] The reference level is determined based on the initial level corresponding to the target user and the progress vector.

[0224] In a possible implementation, the second determining module 603, when determining the reference level based on the initial level corresponding to the target user and the progress vector, is configured to:

[0225] Determining an average promotion speed of the target user based on each value in the progress vector; the average promotion speed is used to represent the difference between the initial level corresponding to the target user's next assessment score and the initial level corresponding to the current assessment score;

[0226] The reference level is determined based on the initial level corresponding to the target user and the average promotion speed.

[0227] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference can be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0228] Based on the same technical concept, the embodiment of the present disclosure also provides a computer device. Figure 7 FIG. 7 is a schematic diagram of the structure of a computer device 700 provided in an embodiment of the present disclosure, comprising a processor 701, a memory 702, and a bus 703. The memory 702 is used to store execution instructions and includes a memory 7021 and an external memory 7022. The memory 7021 is also referred to as internal memory and is used to temporarily store operation data in the processor 701 and data exchanged with an external memory 7022 such as a hard disk. The processor 701 exchanges data with the external memory 7022 via the memory 7021. When the computer device 700 is running, the processor 701 communicates with the memory 702 via the bus 703, so that the processor 701 executes the following instructions:

[0229] Obtaining an assessment score of a target user, and determining an initial level corresponding to the target user based on the assessment score;

[0230] Determining a target level for the target user; wherein the target level is higher than the initial level; different levels correspond to different levels of difficulty;

[0231] A recommended topic corresponding to the target user is determined based on reference topics corresponding to respective intermediate levels between the initial level and the target level, wherein the intermediate levels include the target level.

[0232] In a possible implementation, the instructions executed by the processor 701 include determining the initial level corresponding to the target user based on the assessment score, including:

[0233] Determine a target location area where the target user is located;

[0234] Based on the assessment score, an initial level of the target user in the target location area is determined.

[0235] In a possible implementation, in the instructions executed by the processor 701, determining the target level of the target user includes:

[0236] Determine the target level based on a preset target level determination rule and the initial level; or

[0237] In response to an instruction to set a target level, the target level is determined.

[0238] In one possible implementation, the instructions executed by the processor 701, wherein determining a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level, includes:

[0239] For any reference topic of any intermediate level, determine the target similarity between each pre-stored candidate topic and the reference topic;

[0240] Based on the target similarity, a recommendation topic corresponding to the target user is determined.

[0241] In one possible implementation, in the instructions executed by the processor 701, for any reference topic of any intermediate level, determining the target similarity between each pre-stored candidate topic and the reference topic includes:

[0242] For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method;

[0243] The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

[0244] In one possible implementation, in the instructions executed by the processor 701, determining a recommended topic corresponding to the target user based on the target similarity includes:

[0245] Determine the weight of the recommended questions corresponding to each intermediate level; wherein the recommended question weight is used to represent the importance of the questions based on each intermediate level;

[0246] For any candidate topic, a recommendation score for the candidate topic is determined based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic;

[0247] Based on the recommendation score, a recommendation topic corresponding to the target user is determined.

[0248] In one possible implementation, in the instructions executed by the processor 701, determining the weight of the recommended questions corresponding to each intermediate level includes:

[0249] Determining the initial weight corresponding to each intermediate level, and determining a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted;

[0250] determining a first level among the intermediate levels that is lower than the reference level;

[0251] Based on the initial weight corresponding to the first level and the initial weight of the reference level, the weight of the reference level is determined; and the initial weights of other levels except the reference level are used as the weights of the other levels.

[0252] In a possible implementation, in the instructions executed by the processor 701, determining the reference level in the intermediate level includes:

[0253] Determine the historical initial level corresponding to the historical assessment score of the target user;

[0254] Determining a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user;

[0255] The reference level is determined based on the initial level corresponding to the target user and the progress vector.

[0256] In a possible implementation, in the instructions executed by the processor 701, determining the reference level based on the initial level corresponding to the target user and the progress vector includes:

[0257] Determining an average promotion speed of the target user based on each value in the progress vector; the average promotion speed is used to represent the difference between the initial level corresponding to the target user's next assessment score and the initial level corresponding to the current assessment score;

[0258] The reference level is determined based on the initial level corresponding to the target user and the average promotion speed.

[0259] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program executes the steps of the method for recommending topics described in the above method embodiment. The storage medium may be a volatile or non-volatile computer-readable storage medium.

[0260] The embodiments of the present disclosure also provide a computer program product, which carries program code. The instructions included in the program code can be used to execute the steps of the question recommendation method described in the above method embodiment. For details, please refer to the above method embodiment and will not be repeated here.

[0261] The computer program product may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0262] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems and devices described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0263] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0264] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0265] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, 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, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0266] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The scope of protection of the present disclosure is not limited thereto. Although the present disclosure has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed in the present disclosure, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure shall be subject to the scope of protection of the claims.

Claims

1. A topic recommendation method, characterized in that: include: Obtaining an assessment score of a target user, and determining an initial level corresponding to the target user based on the assessment score; Determining a target level for the target user; wherein the target level is higher than the initial level; different levels correspond to different levels of difficulty; Determine a recommended topic corresponding to the target user based on reference topics corresponding to various intermediate levels between the initial level and the target level; wherein the intermediate levels include the target level, The step of determining a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level includes: For any reference topic of any intermediate level, determining a target similarity between each pre-stored candidate topic and the reference topic; Based on the target similarity, determine the recommended topic corresponding to the target user, The step of determining the target similarity between each pre-stored candidate topic and any reference topic of any intermediate level includes: For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method; The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

2. The method according to claim 1, characterized in that The determining the initial level corresponding to the target user based on the assessment score includes: Determine a target location area where the target user is located; Based on the assessment score, an initial level of the target user in the target location area is determined.

3. The method according to claim 1, characterized in that Determining the target level of the target user includes: Determine the target level based on a preset target level determination rule and the initial level; or In response to an instruction to set a target level, the target level is determined.

4. The method according to claim 1, wherein The determining of a recommendation topic corresponding to the target user based on the target similarity includes: Determine the weight of the recommended questions corresponding to each intermediate level; wherein the recommended question weight is used to represent the importance of the questions based on each intermediate level; For any candidate topic, a recommendation score for the candidate topic is determined based on the target similarity between the candidate topic and any reference topic, and the recommendation weight corresponding to the intermediate level of the reference topic; wherein the recommendation score is used to represent the probability of the candidate topic being the recommended topic; Based on the recommendation score, a recommendation topic corresponding to the target user is determined.

5. The method according to claim 4, characterized in that Determining the weight of the recommended questions corresponding to each intermediate level includes: Determining the initial weight corresponding to each intermediate level, and determining a reference level among the intermediate levels; wherein the reference level is the level to which the target user may be promoted; determining a first level among the intermediate levels that is lower than the reference level; Based on the initial weight corresponding to the first level and the initial weight of the reference level, the weight of the reference level is determined; and the initial weights of other levels except the reference level are used as the weights of the other levels.

6. The method according to claim 5, characterized in that The determining of the reference level in the intermediate level includes: Determine the historical initial level corresponding to the historical assessment score of the target user; Determining a progress vector of the target user based on the historical initial level; wherein the progress vector is used to represent a learning rate of the target user; The reference level is determined based on the initial level corresponding to the target user and the progress vector.

7. The method according to claim 6, characterized in that The determining the reference level based on the initial level corresponding to the target user and the progress vector includes: Determining an average promotion speed of the target user based on each value in the progress vector; the average promotion speed is used to represent the difference between the initial level corresponding to the target user's next assessment score and the initial level corresponding to the current assessment score; The reference level is determined based on the initial level corresponding to the target user and the average promotion speed.

8. A topic recommendation device, characterized in that: include: An acquisition module, configured to acquire an assessment score of a target user and determine an initial level corresponding to the target user based on the assessment score; A first determining module is configured to determine a target level for the target user; wherein the target level is higher than the initial level; and different levels correspond to different levels of difficulty of questions; The second determining module is configured to determine a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level; wherein the intermediate levels include the target level, The step of determining a recommended topic corresponding to the target user based on reference topics corresponding to intermediate levels between the initial level and the target level includes: For any reference topic of any intermediate level, determining a target similarity between each pre-stored candidate topic and the reference topic; Based on the target similarity, determine the recommended topic corresponding to the target user, The step of determining the target similarity between each pre-stored candidate topic and any reference topic of any intermediate level includes: For any candidate topic, determining an initial similarity between the candidate topic and the reference topic based on at least one similarity determination method; The initial similarities are weighted and summed according to preset weights to obtain a target similarity between the candidate topic and the reference topic.

9. A computer device, characterized in that: include: A processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via 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 performed.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the topic recommendation method according to any one of claims 1 to 7 are executed.

Images