Scene learning recommendation method and system based on knowledge points in target range

Abstract

The invention discloses a scene learning recommendation method and system based on knowledge points in a target range, and the method comprises the steps: obtaining a to-be-recommended knowledge point set based on a learned knowledge point set of a student and a knowledge graph; traversing all training scenes, for each scene, sorting learned knowledge points in consideration of enhanced review time and current time, and obtaining a to-be-recommended knowledge point set corresponding to a learned knowledge point set after scene sorting; calculating a recommendation value of the scene according to the learning mode, only considering learned knowledge points for a review mode, and considering the learned knowledge points and recommended knowledge points simultaneously for a new knowledge learning mode; and in all the training scenes, selecting the scene with the highest recommendation value as a recommendation learning scene. The target range scene needing to be learned subsequently can be recommended in a personalized manner according to the current learning ability of the student, the learning cost of the student is reduced, and the learning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to a scene learning recommendation method system based on knowledge points in a shooting range, and belongs to the technical field of computer software and network security. Background Art

[0002] The cyber range is a platform that uses virtualization technology to simulate the real cyberspace attack and defense combat environment and can support various experiments such as combat capability research and teaching training. The range contains a large amount of teaching content. Students can search for keywords to obtain the content they are interested in for learning; at the same time, the platform will also recommend the current popular courses as the direction for students to further study. The current system's recommendation method has the following problems: 1. The system randomly recommends courses or scenarios, and cannot make personalized learning recommendations based on the knowledge that students have already mastered. Systematic learning can reduce the cost of learning new knowledge, while the cost of blind learning will be very high; 2. The recommended learning content is all new content, and it cannot be comprehensively recommended based on the students' already learned abilities to achieve the purpose of learning new and old knowledge at the same time. Summary of the invention

[0003] Purpose of the invention: In view of the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a method and system for recommending scene learning based on knowledge points in a shooting range, which can make personalized recommendations on shooting range scenes that need to be learned later according to the current learning ability of the students.

[0004] Technical solution: To achieve the above-mentioned invention object, the present invention adopts the following technical solution: a scene learning recommendation method based on knowledge points in a shooting range, comprising the following steps:

[0005] Based on the set of knowledge points that students have learned and the knowledge graph, a set of knowledge points to be recommended is obtained;

[0006] Traverse all training scenarios, and for each scenario, sort the learned knowledge points by considering the intensive review time and the current time, and obtain the set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario sorting; calculate the recommendation value of the scenario according to the learning mode, only consider the learned knowledge points for the review mode, and consider both the learned knowledge points and the recommended knowledge points for the new knowledge learning mode;

[0007] Among all training scenarios, the scenario with the highest recommendation value is selected as the recommended learning scenario.

[0008] Preferably, the intersection of the set of knowledge points learned by the student and the set of knowledge points involved in the scene is taken as the set of knowledge points learned by the student in the corresponding scene; the intersection of the set of knowledge points to be recommended and the set of knowledge points involved in the scene is taken as the set of knowledge points to be recommended for the student in the corresponding scene.

[0009] As a preferred method, in the process of sorting the learned knowledge points by considering the intensive review time and the current time, the distance F is calculated for each knowledge point. i And according to the distance F i Sort in descending order; if the knowledge point has completed the preset number of intensive reviews, then the distance F i is 0; if the next intensive review time of the knowledge point is later than the current time, the distance F i is 0; if the next intensive review time of the knowledge point is not later than the current time, then the distance F i It is inversely proportional to the difference between the two. The smaller the difference, the higher the ranking.

[0010] As a preference, in review mode, scene S j Recommended value where F i ′ is the scene S j The sorted set of learned knowledge points D is the distance value of the i-th knowledge point in the set D′ sorted by distance; k is the upper limit of the number of recommended knowledge points.

[0011] As a preferred option, in the new knowledge learning mode, scenario S j Recommended value Among them, H i ′ is the identification value of whether the i-th knowledge point exists in the set of recommended knowledge points E′ corresponding to the set D′, which is 1 if it exists and 0 if it does not exist; W1 and W2 are weights.

[0012] Preferably, after the user has learned the recommended scenario, the user's learned knowledge points and the corresponding knowledge point review period are updated.

[0013] Preferably, for each knowledge point in the set of knowledge points that the student has learned, a best matching recommended knowledge point is selected from the knowledge graph.

[0014] A scene learning recommendation system based on knowledge points in a shooting range, comprising:

[0015] The recommended knowledge point acquisition module is used to obtain the set of knowledge points to be recommended based on the set of knowledge points that the students have learned and the knowledge graph;

[0016] The scenario recommendation value calculation module is used to traverse all training scenarios. For each scenario, the learned knowledge points are sorted by considering the intensive review time and the current time, and the set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario sorting is obtained; the recommendation value of the scenario is calculated according to the learning mode. For the review mode, only the learned knowledge points are considered, and for the new knowledge learning mode, both the learned knowledge points and the recommended knowledge points are considered;

[0017] And the scene recommendation module is used to select the scene with the highest recommendation value among all training scenes as the recommended learning scene.

[0018] A computer system includes a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor. When the computer program / instruction is executed by the processor, the steps of the scene learning recommendation method based on knowledge points in a shooting range are implemented.

[0019] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of the scene learning recommendation method based on knowledge points in a shooting range.

[0020] Beneficial effects: Compared with the prior art, the present invention has the following advantages: 1. The present invention can recommend new knowledge that needs to be learned in the future according to the current ability of the students, thereby reducing the learning cost of the students and improving the learning efficiency; 2. When recommending new knowledge points, the present invention matches the forgetting curve based on the students' historical learning situation. Priority is given to recommending comprehensive scenarios that include the knowledge that the students need to review most at present, as well as the new knowledge points that have the highest matching degree with the knowledge that needs to be reviewed most at present. Students can consolidate the historical knowledge points they have learned while learning new knowledge. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments.

[0023] A method for recommending scenario learning based on knowledge points in a shooting range disclosed in an embodiment of the present invention comprises the following steps: first, based on a set of knowledge points already learned by trainees and a knowledge graph, a set of knowledge points to be recommended is obtained; then, all training scenarios are traversed, and for each scenario, the learned knowledge points are sorted by considering the intensive review time and the current time, and a set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario is sorted is obtained; the recommendation value of the scenario is calculated according to the learning mode, and for the review mode, only the learned knowledge points are considered, and for the new knowledge learning mode, both the learned knowledge points and the recommended knowledge points are considered; finally, among all training scenarios, the scenario with the highest recommendation value is selected as the recommended learning scenario.

[0024] Each course or practice scenario in the shooting range involves multiple different knowledge points. For example, in a confrontation scenario, it will include knowledge such as network reconnaissance and lateral movement. In this embodiment, the association relationship between knowledge points is formed into a knowledge graph in advance, and the best subsequent recommended knowledge point Ki_next=maxargs(Ki) of the current knowledge point Ki can be found from the knowledge graph. If there is more than one best matching value for the subsequent recommended knowledge point of a certain knowledge point, one is randomly selected as the best subsequent recommended knowledge point.

[0025] In this embodiment, according to the forgetting curve of long-term memory, new knowledge is reviewed a total of X times after learning, and the review cycle is recorded as T = [T1, T2, ... T X ], that is, the knowledge point is T1, T2, ...T after the new learning X Review every day.

[0026] In some embodiments, the distance F is calculated for each knowledge point in order to sort the learned knowledge points by considering the intensive review time and the current time. i And according to the distance F i Sort in descending order; if the knowledge point has completed the preset number of intensive reviews, then the distance F i is 0; if the next intensive review time of the knowledge point is later than the current time, the distance F i is 0; if the next intensive review time of the knowledge point is not later than the current time, then the distance F i It is inversely proportional to the difference between the two. The smaller the difference, the higher the ranking.

[0027] like Figure 1 As shown in the figure, the detailed steps of the scene learning recommendation method based on knowledge points in the target range are as follows:

[0028] Step 1: Get the set of knowledge points that students have learned A = [A1, A2, ..., A n ], n is the number of knowledge points that the students have learned.

[0029] Step 2: Based on the knowledge point A that students have already learned i , find A in the knowledge graph i The corresponding next knowledge point B i The set of knowledge points to be recommended based on A is recorded as B = [B1, B2, ..., B n ]. If a certain A i Can't find the corresponding B in the knowledge graph i , then we think that B i Is empty.

[0030] Step 3: Traverse all training scenarios S = [S1, S2, ... S C ], C is the number of scenes, calculate each scene S iThe recommended value based on the knowledge point is calculated in the same way as steps 4 to 7.

[0031] Step 4: For each scene S j , the knowledge points involved are recorded as C sj . Calculate the knowledge points that the students have learned in this scenario D = A ∩ C sj =[D1,D2,...,D p ], and the recommended knowledge points E=B∩C associated with the knowledge graph sj =[E1,E2,...,E q ]; where p and q are the corresponding numbers of knowledge points.

[0032] Step 5: Calculate the distance F between the p learned knowledge points in D and the current time and sort them. i The calculation method is as follows:

[0033] (a) Knowledge point D i The next intensive review time is T di Later than the current time, distance F i Recorded as 0;

[0034] (b) For knowledge point Di, X reviews have been completed, and the distance from F i Recorded as 0;

[0035] (c)D i The next intensive review time is T di No later than the current time,

[0036] Step 6: Calculate the F i Arrange in descending order, and the set of knowledge points corresponding to the sorted results in D is denoted as D′; the set corresponding to each knowledge point in D′ is denoted as E′.

[0037] Step 7: Calculate scenario S according to business requirements j Recommended value of G j :

[0038] In step 5, F i Corresponding to a knowledge point that has been learned D i , and the D i There is a corresponding best recommended knowledge point E i . E i The following assignment is used, and the result is recorded as H i :If E i The corresponding knowledge point exists, then the corresponding value H i =1; otherwise H i =0.

[0039] For review mode, It means that only the knowledge points that have been learned will be selected, and k means that at most k knowledge points will be recommended each time to ensure the effectiveness of the review.

[0040] For new knowledge learning mode, k means that at most k knowledge points are recommended each time. W1 and W2 are weights, W1+W2=1, and the specific value can be adjusted as needed, for example, W1 can be set to 0.8.

[0041] Step 8: Repeat steps 4 to 7 to calculate the recommended values ​​for all scenarios. The resulting recommended value list is recorded as G = [G1, G2, ..., G C ].

[0042] Step 9: Take the scene corresponding to the maximum recommendation value in G as the scene for this recommendation learning;

[0043] Step 10: After the user has learned the recommended scenario, the system updates the user's list of learned knowledge points and the corresponding knowledge point review period.

[0044] In this step, the review period of a knowledge point is assumed to be [T1, T2, …T X ], this review is T k cycle, the initial learning time is recorded as T0. k After reviewing, the remaining review cycles need to be updated. The update method is as follows: i =T i +(now-T k ). For example, suppose a knowledge point needs to be reviewed in three cycles [1, 4, 10] after learning is completed, and the initial learning time is T0 (such as 2024 / 8 / 8), the review cycle = [T0+1, T0+4, T0+10]. Assuming that the review of cycle 2 is actually completed on the 7th day, which is 3 days later than the originally planned 4th day, then the corresponding review plan for the 10th day should also be postponed (7-4) days, that is, the review on the 13th day.

[0045] Based on the same inventive concept, an embodiment of the present invention discloses a scenario learning recommendation system based on knowledge points in a shooting range, including: a recommended knowledge point acquisition module, used to obtain a set of knowledge points to be recommended based on a set of knowledge points already learned by trainees and a knowledge graph; a scenario recommendation value calculation module, used to traverse all training scenarios, and for each scenario, sort the learned knowledge points considering the intensive review time and the current time, and obtain a set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario is sorted; the recommendation value of the scenario is calculated according to the learning mode, only the learned knowledge points are considered for the review mode, and both the learned knowledge points and the recommended knowledge points are considered for the new knowledge learning mode; and a scenario recommendation module, used to select the scene with the highest recommendation value among all training scenarios as the recommended learning scene.

[0046] Based on the same inventive concept, an embodiment of the present invention discloses a computer system, including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, the steps of the scene learning recommendation method based on knowledge points in the target range are implemented.

[0047] Based on the same inventive concept, an embodiment of the present invention discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the scene learning recommendation method based on knowledge points in a shooting range.

Claims

1. A scene learning recommendation method based on knowledge points in a shooting range, characterized in that: The steps include: Based on the set of knowledge points that students have learned and the knowledge graph, a set of knowledge points to be recommended is obtained; Traverse all training scenarios, and for each scenario, sort the learned knowledge points by considering the intensive review time and the current time, and obtain the set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario sorting; The recommended value of the scenario is calculated based on the learning mode. For the review mode, only the learned knowledge points are considered. For the new knowledge learning mode, both the learned knowledge points and the recommended knowledge points are considered. Among all training scenarios, the scenario with the highest recommendation value is selected as the recommended learning scenario.

2. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: The intersection of the set of knowledge points that the student has learned and the set of knowledge points involved in the scene is taken as the set of knowledge points that the student has learned in the corresponding scene; the intersection of the set of knowledge points to be recommended and the set of knowledge points involved in the scene is taken as the set of knowledge points to be recommended for the student in the corresponding scene.

3. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: In the above-mentioned sorting of the learned knowledge points by considering the intensive review time and the current time, the distance F is calculated for each knowledge point. i And according to the distance F i Sort in descending order; if the knowledge point has completed the preset number of intensive reviews, then the distance F i is 0; if the next intensive review time of the knowledge point is later than the current time, the distance F i is 0; if the next intensive review time of the knowledge point is not later than the current time, then the distance F i It is inversely proportional to the difference between the two. The smaller the difference, the higher the ranking.

4. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: In review mode, scene S j Recommended value where F i ′ is the scene S j The sorted set of learned knowledge points D is the distance value of the i-th knowledge point in the set D′ sorted by distance; k is the upper limit of the number of recommended knowledge points.

5. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: In the new knowledge learning mode, scenario S j Recommended value where F i ′ is the scene S j The sorted set of learned knowledge points D is the distance value of the i-th knowledge point in the set D′ sorted by distance; H i ′ is the identification value of whether the ith knowledge point exists in the set of recommended knowledge points E′ corresponding to the set D′, which is 1 if it exists and 0 if it does not; W1 and W2 are weights; k is the upper limit of the number of recommended knowledge points.

6. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: After the user has learned the recommended scenario, the user's learned knowledge points and the corresponding knowledge point review cycle are updated.

7. The scene learning recommendation method based on knowledge points in a shooting range according to claim 1, characterized in that: For each knowledge point in the set of knowledge points that the student has learned, select a best matching recommended knowledge point from the knowledge graph.

8. A scene learning recommendation system based on knowledge points in a shooting range, characterized in that: include: The recommended knowledge point acquisition module is used to obtain the set of knowledge points to be recommended based on the set of knowledge points that the students have learned and the knowledge graph; The scenario recommendation value calculation module is used to traverse all training scenarios, and for each scenario, sort the learned knowledge points by considering the intensive review time and the current time, and obtain the set of knowledge points to be recommended corresponding to the set of learned knowledge points after the scenario sorting; The recommended value of the scenario is calculated based on the learning mode. For the review mode, only the learned knowledge points are considered. For the new knowledge learning mode, both the learned knowledge points and the recommended knowledge points are considered. And the scene recommendation module is used to select the scene with the highest recommendation value among all training scenes as the recommended learning scene.

9. A computer system comprising a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, characterized in that: When the computer program / instructions are executed by a processor, the steps of the scene learning recommendation method based on knowledge points in a shooting range are implemented according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the scene learning recommendation method based on knowledge points in a shooting range are implemented according to any one of claims 1 to 7.

Images