Cognitive analysis method and device for learning object and electronic device thereof

Abstract

The present invention discloses a cognitive analysis method, device and electronic device for a learning object. The cognitive analysis method includes: obtaining a set of scores of all learning objects answering exercises in the historical process and a set of knowledge points examined by each exercise, wherein each exercise corresponds to multiple knowledge points, and then based on the score set and the knowledge point set, analyzing the learning object's cognitive probability parameter set for each knowledge point and the set of influence factors of each knowledge point on related questions, and then using the cognitive probability parameter set and the set of influence factors of each knowledge point on related questions to construct a cognitive model, and finally using the cognitive model to analyze the probability of the target learning object answering new exercises. The present invention solves the technical problem in the related art that it is impossible to handle multiple knowledge points examined simultaneously in one exercise, which affects the accuracy of the cognitive evaluation of the learning object.

Description

Technical Field

[0001] The present invention relates to the field of cognitive analysis technology, and in particular to a cognitive analysis method and device for a learning object and electronic equipment thereof. Background Art

[0002] In recent years, online education systems have been booming, providing students with a wealth of practice exercises and promoting the sharing of learning resources. Knowledge Tracing (KT) defines a task that tracks the changes in students' cognitive states based on their historical answers to exercises, and then predicts the answers to future questions.

[0003] Among the related technologies, the Bayesian Knowledge Tracing (BKT) model is mainly used in single knowledge point scenarios. As the number of knowledge points increases, the parameters of the BKT model will explode exponentially, and the complexity of the model algorithm will also increase exponentially. This not only limits the application scope of the BKT model (that is, it cannot be applied to scenarios with multiple knowledge points), but also affects the accuracy of the assessment of students' cognition during the learning process (because in actual scenarios a question often tests multiple knowledge points).

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method, device and electronic device for cognitive analysis of a learning object, so as to at least solve the technical problem in the related art that it is impossible to simultaneously examine multiple knowledge points in one exercise, thus affecting the accuracy of cognitive evaluation of the learning object.

[0006] According to one aspect of an embodiment of the present invention, a cognitive analysis method for a learning object is provided, comprising: obtaining a set of scores of all learning objects answering exercises in a historical process and a set of knowledge points examined by each exercise, wherein each exercise corresponds to a plurality of knowledge points; based on the set of scores and the set of knowledge points, analyzing a set of cognitive probability parameters of the learning object for each knowledge point and a set of influence factors of each knowledge point on associated questions; constructing a cognitive model using the set of cognitive probability parameters and the set of influence factors of each knowledge point on associated questions; and analyzing the probability of a target learning object answering a new exercise correctly using the cognitive model.

[0007] Optionally, based on the score set and the knowledge point set, the step of analyzing the learning object's cognitive probability parameter set for each knowledge point and the set of influence factors of each knowledge point on associated questions includes: based on the score set and the knowledge point set, analyzing the learning object's initial mastery probability, learning probability and forgetting probability for each knowledge point; based on the learning object's initial mastery probability, learning probability and forgetting probability for each knowledge point, determining the learning object's cognitive probability parameter set for each knowledge point.

[0008] Optionally, based on the score set and the knowledge point set, the step of analyzing the learning object's cognitive probability parameter set for each knowledge point and the set of influence factors of each knowledge point on associated questions also includes: confirming the first influence factor of each knowledge point on the error rate of associated questions and the second influence factor of each knowledge point on the guessing rate of associated questions; and constructing the influence factor set with the first influence factor and the second influence factor.

[0009] Optionally, the step of constructing a cognitive model using the cognitive probability parameter set and the influence factor set of each knowledge point on the associated questions includes: based on the score set and the knowledge point set, using a first formula to calculate a first probability value of the learning object answering the exercise correctly in the current cognitive state; based on the first influence factor of each knowledge point on the error rate of the associated questions, using a second formula to calculate the error rate of the learning object in answering the exercises; based on the second influence factor of each knowledge point on the guess rate of the associated questions, using a third formula to calculate the guess rate of the learning object in answering the exercises; based on the first probability value of the learning object answering the target exercise correctly in the current cognitive state, the error rate of the learning object in answering the target exercise, and the guess rate, using a fourth formula to calculate the second probability value of the learning object answering the exercise correctly; based on the first probability value and the second probability value, analyzing the mastery status of each knowledge point at the current moment by the learning object; and constructing a cognitive model corresponding to the learning object based on the mastery status of each knowledge point at the current moment by the learning object.

[0010] Optionally, after constructing the cognitive model, the cognitive analysis method also includes: if the question answered by the learning object at the previous moment did not involve the target knowledge point, determining the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; based on the learning object's mastery of each knowledge point at the previous moment and the probability of forgetting the target knowledge point from the previous moment to the current moment, using the fifth formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model.

[0011] Optionally, after constructing the cognitive model, the cognitive analysis method also includes: if the question answered by the learning object at the previous moment involves the target knowledge point, determining the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; determining the first learning probability of the learning object answering the question corresponding to the target knowledge point correctly at the previous moment and the second learning probability of not answering the question corresponding to the target knowledge point correctly; based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the first learning probability of answering the question corresponding to the target knowledge point correctly at the previous moment, using the sixth formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model; or, based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the second learning probability of not answering the question corresponding to the target knowledge point correctly at the previous moment, using the seventh formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model.

[0012] Optionally, after using the cognitive model to analyze the probability of the target learning object answering new exercises correctly, the cognitive analysis method also includes: training the cognitive model to update the learning object's mastery of each knowledge point; based on the updated mastery of each knowledge point by the learning object, adjusting the cognitive probability parameters and influencing factors of the cognitive model.

[0013] According to another aspect of an embodiment of the present invention, a cognitive analysis device for a learning object is also provided, comprising: an acquisition unit, used to acquire a set of scores of all learning objects answering exercises in a historical process and a set of knowledge points examined by each exercise, wherein each exercise corresponds to multiple knowledge points; a first analysis unit, used to analyze the learning object's cognitive probability parameter set for each knowledge point and a set of influence factors of each knowledge point on associated questions based on the score set and the knowledge point set; a construction unit, used to construct a cognitive model using the cognitive probability parameter set and the set of influence factors of each knowledge point on associated questions; and a second analysis unit, used to analyze the probability of the target learning object answering new exercises correctly using the cognitive model.

[0014] Optionally, the first analysis unit includes: a first analysis module, used to analyze the initial mastery probability, learning probability and forgetting probability of each knowledge point of the learning object based on the score set and the knowledge point set; a first determination module, used to determine the cognitive probability parameter set of the learning object for each knowledge point based on the initial mastery probability, learning probability and forgetting probability of each knowledge point of the learning object.

[0015] Optionally, the first analysis unit also includes: a second determination module, used to confirm the first influencing factor of each knowledge point on the error rate of the associated question and the second influencing factor of each knowledge point on the guessing rate of the associated question; a first construction module, used to construct the influencing factor set with the first influencing factor and the second influencing factor.

[0016] Optionally, the construction unit includes: a first calculation module, which is used to calculate the first probability value of the learning object answering the exercises correctly in the current cognitive state based on the score set and the knowledge point set by using a first formula; a second calculation module, which is used to calculate the error rate of the learning object answering the exercises by using a second formula based on the first influence factor of each knowledge point on the error rate of the associated questions; a third calculation module, which is used to calculate the guess rate of the learning object answering the exercises by using a third formula based on the second influence factor of each knowledge point on the guess rate of the associated questions; a fourth calculation module, which is used to calculate the second probability value of the learning object answering the exercises correctly in the current cognitive state based on the first probability value of the learning object answering the target exercises, the error rate of the learning object answering the target exercises, and the guess rate, by using a fourth formula; a second analysis module, which is used to analyze the mastery status of each knowledge point of the learning object at the current moment based on the first probability value and the second probability value; and a second construction module, which is used to construct a cognitive model corresponding to the learning object based on the mastery status of each knowledge point of the learning object at the current moment.

[0017] Optionally, the cognitive analysis device also includes: a third determination module, which is used to determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment after constructing the cognitive model, if the question answered by the learning object at the previous moment did not involve the target knowledge point; a first updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment and the probability of forgetting the target knowledge point from the previous moment to the current moment, using the fifth formula to update the cognitive model.

[0018] Optionally, the cognitive analysis device also includes: a fourth determination module, which is used to determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment after constructing the cognitive model, if the question answered by the learning object at the previous moment involves the target knowledge point; a fifth determination module, which is used to determine the first learning probability of the learning object answering the question corresponding to the target knowledge point correctly at the previous moment and the second learning probability of not answering the question corresponding to the target knowledge point correctly; a second updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the first learning probability of answering the question corresponding to the target knowledge point at the previous moment, using the sixth formula to update the cognitive model; or a third updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the second learning probability of not answering the question corresponding to the target knowledge point correctly at the previous moment, using the seventh formula to update the cognitive model.

[0019] Optionally, the cognitive analysis device also includes: a fourth updating module, which is used to train the cognitive model and update the learning object's mastery of each knowledge point after using the cognitive model to analyze the target learning object's probability of answering new exercises correctly; and a first adjustment module, which is used to adjust the cognitive probability parameters and influencing factors of the cognitive model based on the updated mastery of each knowledge point by the learning object.

[0020] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned cognitive analysis methods of a learning object by executing the executable instructions.

[0021] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the cognitive analysis methods for learning objects described above.

[0022] In an embodiment of the present invention, the score set of all learning objects answering exercises in the historical process and the knowledge point set examined by each exercise are first obtained, wherein each exercise corresponds to multiple knowledge points, and then based on the score set and the knowledge point set, the cognitive probability parameter set of the learning object for each knowledge point and the influence factor set of each knowledge point on the associated questions are analyzed, and then the cognitive probability parameter set and the influence factor set of each knowledge point on the associated questions are used to construct a cognitive model, and finally the cognitive model is used to analyze the probability of the target learning object answering new exercises correctly. In this embodiment, a cognitive analysis technology that can be oriented to multiple knowledge points is used, which can solve the scenario where traditional BKT cannot handle a question that examines multiple knowledge points at the same time. The accuracy of using a cognitive model to analyze the probability of the learning object answering new exercises is higher, thereby solving the technical problem that the related technology cannot handle the simultaneous examination of multiple knowledge points in a question, affecting the accuracy of the cognitive evaluation of the learning object. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0024] Figure 1 is a flow chart of an optional cognitive analysis method for a learning object according to an embodiment of the present invention;

[0025] Figure 2 is a schematic diagram of a cognitive analysis device for a learning object according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] To facilitate those skilled in the art to understand the present invention, some terms or nouns involved in the embodiments of the present invention are explained below:

[0029] Knowledge Tracing (KT) is a technology that accurately tracks students' dynamically changing cognitive states based on their past practice performance.

[0030] AUC (Area Under Curve), defined as the area under the ROC curve and the coordinate axis, is a performance indicator to measure the quality of the model.

[0031] The embodiment of the present invention can be applied to cognitive analysis systems / cognitive analysis software for various learning objects, for example, various online education software and education platforms. This embodiment obtains a data set on an education platform, and based on a cognitive tracking model (or simply referred to as a cognitive model) for multiple knowledge points, can analyze the results of learning objects (for example, students in school, learning objects using online education software, etc.) answering exercises in the future to better serve the learning objects. The cognitive analysis method proposed in the embodiment of the present invention not only broadens the application scenarios of the existing BKT model, so that it can not only handle the situation where multiple knowledge points are tested in the questions, but also can track the changes of each knowledge point separately, and is still effective when new questions with different knowledge point combinations appear in the later time steps. At the same time, it tracks problems with multiple knowledge points, but can effectively avoid the problem of exponential growth of algorithm complexity in the original BKT as the number of knowledge points increases, and uses heuristic optimization methods to tune parameters and use group intelligence to avoid local optimality.

[0032] According to an embodiment of the present invention, an embodiment of a cognitive analysis method for a learning object is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Embodiment 1

[0034] Figure 1 is a flowchart of an optional cognitive analysis method for a learning object according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0035] Step S102, obtaining a set of scores of all learning objects answering questions in the history process and a set of knowledge points examined by each question, wherein each question corresponds to multiple knowledge points.

[0036] Step S104: Based on the score set and the knowledge point set, analyze the learning object's cognitive probability parameter set for each knowledge point and the influence factor set of each knowledge point on the associated questions.

[0037] Step S106, constructing a cognitive model using a set of cognitive probability parameters and a set of influence factors of each knowledge point on the associated questions.

[0038] Step S108, using the cognitive model to analyze the probability of the target learning object answering the new exercise correctly.

[0039] Through the above steps, it is possible to first obtain the score set of all learning objects answering exercises in the historical process and the knowledge point set examined by each exercise, wherein each exercise corresponds to multiple knowledge points, and then analyze the cognitive probability parameter set of each knowledge point and the influence factor set of each knowledge point on the associated questions based on the score set and the knowledge point set, and then use the cognitive probability parameter set and the influence factor set of each knowledge point on the associated questions to construct a cognitive model, and finally use the cognitive model to analyze the probability of the target learning object answering new exercises. In this embodiment, a cognitive analysis technology that can be oriented to multiple knowledge points is used, which can solve the scenario where traditional BKT cannot handle a question that examines multiple knowledge points at the same time. The accuracy of using a cognitive model to analyze the probability of the learning object answering new exercises is higher, thereby solving the technical problem that the related technology cannot handle the simultaneous examination of multiple knowledge points in a question, affecting the accuracy of the cognitive evaluation of the learning object.

[0040] The embodiment of the present invention is described in detail below in conjunction with each step.

[0041] Step S102, obtaining a set of scores of all learning objects answering questions in the history process and a set of knowledge points examined by each question, wherein each question corresponds to multiple knowledge points.

[0042] In the embodiment of the present invention, the learning objects include but are not limited to: students in school, teachers who need to learn, parents who need to learn, and other people of any age who need to learn (they may also be robots, computers, servers). The data set required by the embodiment of the present invention can be obtained from any online education platform. The score set of the learning objects answering the exercises is expressed as Y={y}, y∈{0,1}, and the set of knowledge points examined in each exercise is expressed as KC.

[0043] Step S104: Based on the score set and the knowledge point set, analyze the learning object's cognitive probability parameter set for each knowledge point and the influence factor set of each knowledge point on the associated questions.

[0044] In the embodiment of the present invention, the number of knowledge points is set to I, the number of learning objects is set to S, the evolutionary generation is set to Gens, and the individual mutation rate and individual crossover rate are set, and the initial mastery probability of the learning object for each knowledge point is set to (PL 1 , PL 2 , ..., PL i , ..., PL I ), (i∈{1, 2, ..., I}), where PL i It represents the initial mastery of the i-th knowledge point by the learning object. The learning probability of each knowledge point by the learning object is (PT 1 , PT 2 , ..., PT i , ..., PT I ), (i∈{1, 2, ..., I}), where PT i represents the learning probability of the learning object for the i-th knowledge point, and the forgetting probability of the learning object for each knowledge point is (PF 1 , PF 2 , ..., PF i , ..., PF I ), (i∈{1, 2, ..., I}), where PF i It represents the probability that students forget the i-th knowledge point.

[0045] In the embodiment of the present invention, each question may test multiple knowledge points, so the error rate and guessing rate of this question will be jointly determined by these tested knowledge points, so the influence factor of each knowledge point on the error rate of the question associated with it is set to (S 1 , S 2 , ..., S i , ..., SI ), (i∈{1, 2, ..., I}), where S i represents the influence factor of the i-th knowledge point on the error rate of the question associated with it. The error rate represents the probability that the learning object can answer the question but answers it incorrectly. The influence factor of each knowledge point on the guessing rate of the question associated with it is set to (G 1 , G 2 , ..., G i , ..., G I ), (i∈{1, 2, ..., I}), where G i It represents the influence factor of the i-th knowledge point on the guessing rate of the question associated with it. The guessing rate represents the probability that the learning object can answer the question without having learned the question.

[0046] Optionally, based on the score set and the knowledge point set, the step of analyzing the learning object's cognitive probability parameter set for each knowledge point and the set of influence factors of each knowledge point on associated questions includes: based on the score set and the knowledge point set, analyzing the learning object's initial mastery probability, learning probability and forgetting probability for each knowledge point; based on the learning object's initial mastery probability, learning probability and forgetting probability for each knowledge point, determining the learning object's cognitive probability parameter set for each knowledge point.

[0047] In the embodiment of the present invention, by analyzing the initial mastering probability, learning probability and forgetting probability of each knowledge point of the learning object, the mastering status of each knowledge point of the learning object at the current moment can be obtained. Among them, the mastering state at time t is represented by P (t) The mastery state is also called a cognitive probability parameter set. For example, the mastery state can be roughly divided into {unmastered, partially mastered, mostly mastered, and completely mastered}, unmastered can mean that the probability of the learning object answering the exercise correctly is 0-0.2, partially mastered can mean that the probability of the learning object answering the exercise correctly is 0.3-0.5, mostly mastered can mean that the probability of the learning object answering the exercise correctly is 0.6-0.8, and completely mastered can mean that the probability of the learning object answering the exercise correctly is 0.9-1, and the embodiment of the present invention does not limit this.

[0048] Optionally, based on the score set and the knowledge point set, the step of analyzing the learning object's cognitive probability parameter set for each knowledge point and the influence factor set of each knowledge point on the associated questions also includes: confirming the first influence factor of each knowledge point on the error rate of the associated questions and the second influence factor of each knowledge point on the guessing rate of the associated questions; and constructing the influence factor set with the first influence factor and the second influence factor.

[0049] In the embodiment of the present invention, the first influencing factor is the influencing factor of each knowledge point on the error rate of the question associated with it, expressed as (S 1 , S 2 , ..., S i , ..., S I ), (i∈{1, 2, ..., I}), the second impact factor is the impact factor of each knowledge point on the guessing rate of the question associated with it, expressed as (G 1 , G 2 , ..., G i , ..., G I ), (i∈{1, 2, ..., I}), the first impact factor and the second impact factor are constructed as an impact factor set.

[0050] Step S106, constructing a cognitive model using a set of cognitive probability parameters and a set of influence factors of each knowledge point on the associated questions.

[0051] In this embodiment, a cognitive model is constructed by using a set of cognitive probability parameters and a set of influence factors of each knowledge point on related questions, that is, the current cognitive state of the learning object, including the initial mastery probability, learning probability, forgetting probability of each knowledge point and the influence factors of each knowledge point on the error rate and guessing rate of related questions of the current learning object.

[0052] Optionally, the steps of constructing a cognitive model using a set of cognitive probability parameters and a set of influence factors of each knowledge point on associated questions include: based on the score set and the knowledge point set, using a first formula to calculate a first probability value of the learning object answering the exercises correctly in the current cognitive state; based on the first influence factor of each knowledge point on the error rate of associated questions, using a second formula to calculate the error rate of the learning object in answering the exercises; based on the second influence factor of each knowledge point on the guess rate of associated questions, using a third formula to calculate the guess rate of the learning object in answering the exercises; based on the first probability value of the learning object answering the target exercises correctly in the current cognitive state, the error rate of the learning object in answering the target exercises, and the guess rate, using a fourth formula to calculate the second probability value of the learning object answering the exercises correctly; based on the first probability value and the second probability value, analyzing the learning object's mastery of each knowledge point at the current moment; and constructing a cognitive model corresponding to the learning object based on the learning object's mastery of each knowledge point at the current moment.

[0053] In this embodiment, suppose the target exercise at time t is E (t) , Target Exercise E (t) The set of knowledge points examined is The learning object's mastery of each knowledge point at the current time t is That is, the current cognitive state is P (t) .

[0054] The first formula in the embodiment of the present invention is: Where n∈KC t ={kc 1 , kc 2 ..., kc k}, P (t) represents the current cognitive state, and calculates the probability η that the learning object will answer the question in the current cognitive state without considering the guessing and error factors of the learning object, that is, the first probability value. The second formula is: Where n∈KC t ={kc 1 , kc 2 ..., kc k},S n The third formula is: Where n∈KC t ={kc 1 , kc 2 ..., kc k}, G n The fourth formula is: P represents the influence factor of each knowledge point on the guessing rate of the question associated with it, and the guessing rate of the current learning object to the question P(G) is calculated. corr =η(1-P(S))+(1-η)*P(G), where η represents the first probability value, P(S) represents the error rate of the current learning object to the question, and P(G) represents the guess rate of the current learning object to the question. Taking into account the cognitive state of the students, the guessing factors of the learning object to the question and the error factors, the fourth formula is used to finally obtain the probability P of the learning object answering the current question correctly. corr , which is the second probability value. Then the obtained probability value is used to analyze the mastery status of each knowledge point of the learning object at the current moment, which is used to build a cognitive model.

[0055] Optionally, after constructing the cognitive model, the cognitive analysis method also includes: if the question answered by the learning object at the previous moment did not involve the target knowledge point, determining the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; based on the learning object's mastery of each knowledge point at the previous moment and the probability of forgetting the target knowledge point from the previous moment to the current moment, using the fifth formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model.

[0056] In this embodiment, the cognitive state of the learning object, that is, the mastery of each knowledge point by the learning object, changes over time. In this embodiment, the fifth formula is: in, PFn It indicates the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment. Indicates the mastery status of the learning object at time t. For the knowledge points that were not tested in the questions at the previous time t, because the learning object did not practice at the previous time t, only the forgetting effect of the learning object on the knowledge point is considered, and the fifth formula is used to calculate the mastery status of the target knowledge point at the current time to update the cognitive model.

[0057] Alternatively, after constructing the cognitive model, the cognitive analysis method further includes: if the question answered by the learning object at the previous moment involves the target knowledge point, determining the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; determining the first learning probability of the learning object answering the question corresponding to the target knowledge point correctly at the previous moment and the second learning probability of not answering the question corresponding to the target knowledge point correctly; based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the first learning probability of answering the question corresponding to the target knowledge point correctly at the previous moment, using the sixth formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model; or, based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the second learning probability of not answering the question corresponding to the target knowledge point correctly at the previous moment, using the seventh formula to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model.

[0058] In this embodiment, the sixth formula is: Where n∈KC t ={kc 1 , kc 2 ..., kc k}, PF n It indicates the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment. represents the mastery status of the learning object at the previous time t, S n represents the influence factor of each knowledge point on the error rate of the question it is associated with, G n It represents the influence factor of each knowledge point on the guessing rate of the question associated with it. The seventh formula is: Where n∈KC t ={kc 1 , kc 2 ..., kc k}, PF n It indicates the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment. represents the mastery status of the learning object at the previous time t, S n represents the influence factor of each knowledge point on the error rate of the question it is associated with, G n Represents the influence factor of each knowledge point on the guessing rate of the question associated with it.

[0059] In this embodiment, for the knowledge point examined by the question at time t, because the learning object is practiced at time t, it is necessary to consider the learning effect and forgetting effect of the student on the knowledge point. However, at the previous time t, the learning object answered the question correctly and answered the question incorrectly, and the learning effect obtained was different. If the learning object answered the question correctly, then based on the mastery state of each knowledge point of the learning object at the previous moment, the forgetting probability of the target knowledge point of the learning object from the previous moment to the current moment, and the first learning probability of answering the question corresponding to the target knowledge point at the previous moment (i.e., the probability that the learning object answered the question correctly), the sixth formula is used to calculate the mastery state of the target knowledge point of the learning object at the current moment to update the cognitive model. If the learning object answered the question incorrectly, based on the mastery state of each knowledge point of the learning object at the previous moment, the forgetting probability of the target knowledge point of the learning object from the previous moment to the current moment, and the second learning probability of answering the question corresponding to the target knowledge point at the previous moment (i.e., the probability that the learning object answered the question incorrectly), the seventh formula is used to calculate the mastery state of the target knowledge point of the learning object at the current moment to update the cognitive model.

[0060] Step S108, using the cognitive model to analyze the probability of the target learning object answering the new exercise correctly.

[0061] Optionally, after using the cognitive model to analyze the probability of the target learning object answering new exercises correctly, the cognitive analysis method further includes: training the cognitive model to update the mastery status of each knowledge point of the learning object; and adjusting the cognitive probability parameters and influencing factors of the cognitive model based on the updated mastery status of each knowledge point of the learning object. In this embodiment, the step of training the cognitive model includes:

[0062] Step 1: Initialize the population;

[0063] Step 2: Set the loop iterator gen=1. When gen≤Gens, loop through steps 3-8.

[0064] Step 3: Evaluate individual fitness, evolve the population, and execute steps 4-8;

[0065] Step 4: Select a learning object s=1, (s∈{1, 2, ..., S}), when s≤S, loop through steps 5-8;

[0066] Step 5: The learning object answers the question at time step t = 1. When t ≤ T, steps 6-8 are executed in a loop;

[0067] Step 6: According to the cognitive tracking model, calculate the probability that the learner answers the question correctly at time t, save the predicted value to the variable Pred, and update the cognitive state of the learner;

[0068] Step 7: Calculate the variable Pred and the evaluation index in the data set, and use the evaluation index as the fitness value of the individual;

[0069] Step 8: According to the fitness values ​​of individuals in the population, use the evolutionary algorithm to perform crossover and mutation operations on the individuals to generate the next generation of population;

[0070] Step 9: Get the DNA value of the best individual in the last generation of the population.

[0071] In this embodiment, the cognitive model can be iteratively trained by comparing the predicted answer results with the actual answer results of the learning subjects in the data set, and finally the cognitive probability parameters and influencing factors can be obtained, that is, the learning subject's mastery probability, learning probability, forgetting probability of each knowledge point, and the influence factor of each knowledge point on the error rate and guessing rate of related questions.

[0072] Embodiment 2

[0073] In this embodiment, a Bayesian cognitive tracking method for mixed knowledge points is used to solve the problem that the existing KT model cannot handle mixed knowledge points and that simply expanding the original BKT model will cause the algorithm complexity to grow exponentially with the increase in the number of knowledge points. The Bayesian cognitive tracking method for mixed knowledge points in this embodiment has the following specific steps:

[0074] Step 1, Initialization:

[0075] The optimal parameters are solved by using the evolutionary algorithm in heuristic optimization, and the parameters to be optimized are regarded as the individual DNA in the population.

[0076] 1.1, set the number of knowledge points to I; set the number of students to S; set the evolutionary algorithm evolutionary generation to Gens; set the individual mutation rate and crossover rate of the evolutionary algorithm population;

[0077] 1.2, set the student's initial mastery probability of each knowledge point (PL 1 , PL 2 , ..., PL i , ..., PL I ), (i∈{1, 2, ..., I}), where PL i represents the student's initial mastery of the i-th knowledge point; the student's learning probability for each knowledge point (PT1 , PT 2 , ..., PT i , ..., PT I ), (i∈{1, 2, ..., I}), where PT i represents the probability of students learning the i-th knowledge point; the probability of students forgetting each knowledge point (PF 1 , PF 2 , ..., PF i , ..., PF I ), (i∈{1, 2, ..., I}), where PF i represents the probability of students forgetting the i-th knowledge point. Each question may test multiple knowledge points, so the error rate and guessing rate of this question will be determined by these tested knowledge points. The influence factor of each knowledge point on the error rate of the question it is associated with is set as (S 1 , S 2 , ..., S i , ..., S I ), (i∈{1, 2, ..., I}), where S i represents the influence factor of the i-th knowledge point on the error rate of the question associated with it; the influence factor of each knowledge point on the guessing rate of the question associated with it is set to (G 1 , G 2 , ..., G i , ..., G I ), (i∈{1, 2, ..., I}), where G i Represents the influence factor of the i-th knowledge point on the guessing rate of the question associated with it;

[0078] 1.3. Input the score set Y={y}, y∈{0,1} of all students answering the exercises and the knowledge point set KC examined by all students answering the exercises.

[0079] Step 2, Modeling:

[0080] At a certain time t, the question the student answered is E (t) , the set of knowledge points examined in this question is The student's mastery of each knowledge point at the current moment is

[0081] 2.1, according to formula (1), calculate the probability η that the student will answer the question in the current cognitive state without considering the student's guesses and mistakes;

[0082] Where n∈KC t ={kc 1 , kc2 ..., kc k} (1);

[0083] 2.2, according to formula (2) (3), calculate the error rate and guessing rate of the current students on the questions respectively;

[0084] Where n∈KC t ={kc 1 , kc 2 ..., kc k} (2);

[0085] Where n∈KC t ={kc 1 , kc 2 ..., kc k} (3);

[0086] 2.3, taking into account the students' cognitive state, the guessing factors of the students following the questions, and the error factors, the probability P of the students answering the current question correctly is obtained using formula (4): corr ;

[0087] P corr =η*(1-P(S))+(1-η)*P(G) (4);

[0088] 2.4, Update time t+1, the change of students' cognitive state, that is, the students' mastery of each knowledge point P (t+1) .

[0089] 2.4.1, for question E at time t t For the knowledge points that were not examined, since the students did not practice at time t, only the forgetting effect of the students on the knowledge points was considered and calculated according to formula (5);

[0090] in,

[0091] 2.4.2, for question E at time t t For the knowledge points examined, since students get to practice at time t, it is necessary to consider the students’ learning effect and forgetting effect on the knowledge points. However, at time t, the learning effects obtained by students are different if they answer the questions correctly or incorrectly, and they are calculated according to formulas (6) and (7) respectively.

[0092]

[0093] Where n∈KC t ={kc 1 , kc 2..., kc k} (6);

[0094]

[0095] Where n∈KC t ={kc 1 , kc 2 ..., kc k} (7);

[0096] Step 3: Model training

[0097] 3.1, initialize the population;

[0098] 3.2, set loop iterator gen = 1, when gen ≤ Gens, loop through steps 3.3-3.8;

[0099] 3.3, evaluate individual fitness, evolve the population, and execute steps 3.4-3.8;

[0100] 3.4, select student s=1, (s∈{1, 2, ..., S}), when s≤S, loop through steps 3.5-3.8;

[0101] 3.5, student answering time step t = 1, when t ≤ T, loop through steps 3.6-3.8;

[0102] 3.6, according to step 2, calculate the probability of the student answering the question correctly at time t, and perform binarization to obtain the predicted value of whether the student can answer the current question correctly and save it in the variable Pred, and update the student's cognitive state, that is, the student's mastery of each knowledge point;

[0103] 3.7, calculate the AUC, accuracy and other evaluation indicators of the predicted value Pred and the actual answer results of the students in the data set, and use the evaluation indicators as the fitness value of the individual;

[0104] 3.8, according to the fitness values ​​of individuals in the population, use the evolutionary algorithm to perform crossover and mutation operations on individuals to generate the next generation of population;

[0105] 3.9, the DNA value of the best individual in the last generation of the population is obtained, which is the parameters of the model PL, PT, PF, S, and G, where PL represents the mastery probability, PT represents the learning probability, PF represents the past probability, S represents the influence factor of each knowledge point on the error rate of the associated question, and G represents the influence factor of each knowledge point on the guessing rate of the associated question.

[0106] The multi-knowledge point Bayesian cognitive tracking method proposed in the embodiment of the present invention has the following beneficial effects:

[0107] (1) It broadens the application scenarios of the existing BKT model so that it can handle situations where multiple knowledge points are tested in the questions;

[0108] (2) Track the changes of each knowledge point individually, which is still effective when new questions with different combinations of knowledge points appear in the future time steps;

[0109] (3) It can track multiple knowledge points and solve the problem in the original BKT where the algorithm complexity grows exponentially with the number of knowledge points.

[0110] (4) Use heuristic optimization methods to tune parameters and exploit swarm intelligence to avoid local optimality.

[0111] The present invention is described below by means of another optional embodiment.

[0112] Embodiment 3

[0113] A cognitive analysis device for a learning object provided in this embodiment includes multiple implementation units, each of which corresponds to each implementation step in the above-mentioned embodiment 1.

[0114] Figure 2 is a schematic diagram of a cognitive analysis device for a learning object according to an embodiment of the present invention. Figure 2 As shown, the analysis device may include: an acquisition unit 20, a first analysis unit 22, a construction unit 24, and a second analysis unit 26, wherein:

[0115] An acquisition unit 20 is used to acquire a set of scores of all learning objects answering questions in the history process and a set of knowledge points examined by each question, wherein each question corresponds to a plurality of knowledge points;

[0116] A first analysis unit 22, configured to analyze, based on the score set and the knowledge point set, a set of cognitive probability parameters of the learning object for each knowledge point and a set of influence factors of each knowledge point on the associated questions;

[0117] A construction unit 24, configured to construct a cognitive model using a set of cognitive probability parameters and a set of influence factors of each knowledge point on an associated question;

[0118] The second analyzing unit 26 is used to analyze the probability of the target learning object answering the new exercise correctly by using the cognitive model.

[0119] The cognitive analysis device for the learning object can obtain the score set of all the learning objects answering the exercises in the historical process and the knowledge point set examined by each exercise through the acquisition unit 20, wherein each exercise corresponds to multiple knowledge points, and then analyze the cognitive probability parameter set of the learning object for each knowledge point and the influence factor set of each knowledge point on the related questions through the first analysis unit 22 based on the score set and the knowledge point set, and then use the cognitive probability parameter set and the influence factor set of each knowledge point on the related questions to construct a cognitive model through the construction unit 24, and finally use the cognitive model to analyze the correct answer probability of the target learning object answering new exercises through the second analysis unit 2. In this embodiment, a cognitive analysis technology that can face multiple knowledge points is adopted, which can solve the scenario where the traditional BKT cannot handle the scenario where multiple knowledge points are examined at the same time in one question, and the accuracy of the correct answer probability of the learning object answering new exercises by using the cognitive model is higher, thereby solving the technical problem that the related technology cannot handle the simultaneous examination of multiple knowledge points in one exercise, which affects the accuracy of the cognitive evaluation of the learning object.

[0120] Optionally, the first analysis unit includes: a first analysis module, used to analyze the learning object's initial mastery probability, learning probability and forgetting probability of each knowledge point based on the score set and the knowledge point set; a first determination module, used to determine the learning object's cognitive probability parameter set for each knowledge point based on the learning object's initial mastery probability, learning probability and forgetting probability of each knowledge point.

[0121] Optionally, the first analysis unit also includes: a second determination module, used to confirm the first influencing factor of each knowledge point on the error rate of the associated question and the second influencing factor of each knowledge point on the guessing rate of the associated question; a first construction module, used to construct an influencing factor set with the first influencing factor and the second influencing factor.

[0122] Optionally, the construction unit includes: a first calculation module, which is used to calculate the first probability value of the learning object answering the exercises correctly in the current cognitive state based on the score set and the knowledge point set by using the first formula; a second calculation module, which is used to calculate the error rate of the learning object answering the exercises by using the second formula based on the first influence factor of each knowledge point on the error rate of the associated questions; a third calculation module, which is used to calculate the guess rate of the learning object answering the exercises by using the third formula based on the second influence factor of each knowledge point on the guess rate of the associated questions; a fourth calculation module, which is used to calculate the second probability value of the learning object answering the exercises correctly based on the first probability value of the learning object answering the target exercises correctly in the current cognitive state, the error rate of the learning object answering the target exercises and the guess rate, by using the fourth formula; a second analysis module, which is used to analyze the mastery status of each knowledge point of the learning object at the current moment based on the first probability value and the second probability value; a second construction module, which is used to construct a cognitive model corresponding to the learning object based on the mastery status of each knowledge point of the learning object at the current moment.

[0123] Optionally, the cognitive analysis device also includes: a third determination module, which is used to determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment after constructing the cognitive model, if the question answered by the learning object at the previous moment did not involve the target knowledge point; a first updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment and the probability of forgetting the target knowledge point from the previous moment to the current moment, using the fifth formula to update the cognitive model.

[0124] Optionally, the cognitive analysis device also includes: a fourth determination module, which is used to determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment after constructing the cognitive model, if the question answered by the learning object at the previous moment involves the target knowledge point; a fifth determination module, which is used to determine the first learning probability of the learning object answering the question corresponding to the target knowledge point correctly at the previous moment and the second learning probability of not answering the question corresponding to the target knowledge point correctly; a second updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the first learning probability of answering the question corresponding to the target knowledge point at the previous moment, using the sixth formula to update the cognitive model; or, a third updating module, which is used to calculate the learning object's mastery of the target knowledge point at the current moment based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the second learning probability of not answering the question corresponding to the target knowledge point at the previous moment, using the seventh formula to update the cognitive model.

[0125] Optionally, the cognitive analysis device also includes: a fourth updating module, which is used to train the cognitive model and update the learning object's mastery of each knowledge point after using the cognitive model to analyze the target learning object's probability of answering new exercises correctly; and a first adjustment module, which is used to adjust the cognitive probability parameters and influencing factors of the cognitive model based on the updated learning object's mastery of each knowledge point.

[0126] The cognitive analysis device of the above-mentioned learning object may also include a processor and a memory. The above-mentioned acquisition unit 20, first analysis unit 22, construction unit 24, second analysis unit 26, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0127] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the probability of the target learning object answering a new exercise correctly can be analyzed by adjusting kernel parameters.

[0128] The above-mentioned memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one storage chip.

[0129] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned cognitive analysis methods for a learning object by executing the executable instructions.

[0130] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned cognitive analysis methods for learning objects.

[0131] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0132] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be 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 interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0134] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0135] In addition, each functional unit in each embodiment of the present invention 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0136] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0137] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A cognitive analysis method for a learning object, characterized in that: Applied in cognitive analysis systems for learning objects, including: Obtaining a set of scores of all learning objects answering questions in the history process and a set of knowledge points examined by each question, wherein each question corresponds to a plurality of knowledge points, and the learning objects include robots and computers; Based on the score set and the knowledge point set, analyzing the learning object's cognitive probability parameter set for each knowledge point and the influence factor set of each knowledge point on the associated questions, including: based on the score set and the knowledge point set, analyzing the learning object's initial mastering probability, learning probability and forgetting probability of each knowledge point; based on the learning object's initial mastering probability, learning probability and forgetting probability of each knowledge point, determining the learning object's cognitive probability parameter set for each knowledge point; confirming the first influence factor of each knowledge point on the error rate of the associated questions and the second influence factor of each knowledge point on the guessing rate of the associated questions; constructing the influence factor set with the first influence factor and the second influence factor; The cognitive model is constructed by using the cognitive probability parameter set and the influence factor set of each knowledge point on the associated question, including: based on the score set and the knowledge point set, using a first formula to calculate a first probability value of the learning object answering the exercise correctly in the current cognitive state, wherein the first formula is: Where n∈KC t ={kc1,kc2…,kc k }, P (t) represents the current cognitive state, t is the current moment, kc t Indicates the knowledge point set at the current moment; based on the first influencing factor of each knowledge point on the error rate of the associated question, the error rate of the learning object answering the exercise is calculated using the second formula, the second formula is: Where n∈KC t ={kc1,kc2…,kc k },S n Indicates the influence factor of each knowledge point on the error rate of the question associated with it; based on the second influence factor of each knowledge point on the guessing rate of the associated question, the guessing rate of the learning object answering the exercise is calculated using the third formula, the third formula is: Where n∈KC t ={kc1,kc2…,kc k }, G n represents the influence factor of each knowledge point on the guessing rate of the question associated with it; based on the first probability value of the learning object answering the target question correctly in the current cognitive state, the error rate and guessing rate of the learning object answering the target question, the fourth formula is used to calculate the second probability value of the learning object answering the question correctly, and the fourth formula is: P corr =η*(1-P(S))+(1-η)*P(G), where η represents the first probability value, P(S) represents the error rate of the learning object to the question, and P(G) represents the guessing rate of the learning object to the question; based on the first probability value and the second probability value, analyzing the mastery status of each knowledge point of the learning object at the current moment; based on the mastery status of each knowledge point of the learning object at the current moment, constructing a cognitive model corresponding to the learning object; The cognitive model is used to analyze the probability of the target learning subject answering new exercises correctly.

2. The cognitive analysis method according to claim 1, characterized in that: After constructing the cognitive model, the cognitive analysis method further includes: If the question answered by the learning object at the previous moment does not involve the target knowledge point, determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; Based on the mastery status of each knowledge point by the learning object at the previous moment and the probability of forgetting the target knowledge point from the previous moment to the current moment, the fifth formula is used to calculate the mastery status of the target knowledge point by the learning object at the current moment to update the cognitive model. The fifth formula is: in, PF n It indicates the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment. Represents the mastery status of the learning object at time t.

3. The cognitive analysis method according to claim 2, characterized in that: After constructing the cognitive model, the cognitive analysis method further includes: If the question answered by the learning object at the previous moment involves the target knowledge point, determine the probability of the learning object forgetting the target knowledge point from the previous moment to the current moment; Determine a first learning probability that the learning subject correctly answered the question corresponding to the target knowledge point at the previous moment and a second learning probability that the learning subject did not correctly answer the question corresponding to the target knowledge point; Based on the mastery status of each knowledge point by the learning object at the previous moment, the forgetting probability of the target knowledge point by the learning object from the previous moment to the current moment, and the first learning probability of correctly answering the question corresponding to the target knowledge point at the previous moment, the sixth formula is used to calculate the mastery status of the target knowledge point by the learning object at the current moment to update the cognitive model; or, Based on the learning object's mastery of each knowledge point at the previous moment, the learning object's forgetting probability of the target knowledge point from the previous moment to the current moment, and the second learning probability of not answering the question corresponding to the target knowledge point at the previous moment, the seventh formula is used to calculate the learning object's mastery of the target knowledge point at the current moment to update the cognitive model.

4. The cognitive analysis method according to claim 1, characterized in that: After adopting the cognitive model to analyze the probability of the target learning object answering the new exercise correctly, the cognitive analysis method further includes: Training the cognitive model to update the learning subject's mastery of each knowledge point; Based on the updated mastery status of each knowledge point by the learning object, the cognitive probability parameters and influencing factors of the cognitive model are adjusted.

5. A cognitive analysis device for a learning object, characterized in that: Applied in cognitive analysis systems for learning objects, including: An acquisition unit, used to acquire a set of scores of all learning objects answering questions in a historical process and a set of knowledge points examined by each question, wherein each question corresponds to a plurality of knowledge points, and the learning objects include robots and computers; The first analysis unit is used to analyze the learning object's cognitive probability parameter set for each knowledge point and the influence factor set of each knowledge point on the associated questions based on the score set and the knowledge point set, including: analyzing the learning object's initial mastering probability, learning probability and forgetting probability for each knowledge point based on the score set and the knowledge point set; determining the learning object's cognitive probability parameter set for each knowledge point based on the learning object's initial mastering probability, learning probability and forgetting probability for each knowledge point; confirming the first influence factor of each knowledge point on the error rate of the associated questions and the second influence factor of each knowledge point on the guessing rate of the associated questions; constructing the influence factor set with the first influence factor and the second influence factor; A construction unit is used to construct a cognitive model using the cognitive probability parameter set and the influence factor set of each knowledge point on the associated question, including: based on the score set and the knowledge point set, using a first formula to calculate a first probability value of the learning object answering the exercise correctly in the current cognitive state, wherein the first formula is: Where n∈KC t ={kc1,kc2…,kc k }, P (t) represents the current cognitive state, t is the current moment, kc t Indicates the knowledge point set at the current moment; based on the first influencing factor of each knowledge point on the error rate of the associated question, the error rate of the learning object answering the exercise is calculated using the second formula, the second formula is: Where n∈KC t ={kc1,kc2…,kc k },S n Indicates the influence factor of each knowledge point on the error rate of the question associated with it; based on the second influence factor of each knowledge point on the guessing rate of the associated question, the guessing rate of the learning object answering the exercise is calculated using the third formula, the third formula is: Where n∈KC t ={kc1,kc2…,kc k }, G n represents the influence factor of each knowledge point on the guessing rate of the question associated with it; based on the first probability value of the learning object answering the target question correctly in the current cognitive state, the error rate and guessing rate of the learning object answering the target question, the fourth formula is used to calculate the second probability value of the learning object answering the question correctly, and the fourth formula is: P corr =η*(1-P(S))+(1-η)*P(G), where η represents the first probability value, P(S) represents the error rate of the learning object to the question, and P(G) represents the guessing rate of the learning object to the question; based on the first probability value and the second probability value, analyzing the mastery status of each knowledge point of the learning object at the current moment; based on the mastery status of each knowledge point of the learning object at the current moment, constructing a cognitive model corresponding to the learning object; The second analyzing unit is used to analyze the probability of the target learning object answering the new exercise correctly by using the cognitive model.

6. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the cognitive analysis method of a learning object as described in any one of claims 1 to 4 by executing the executable instructions.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the cognitive analysis method for a learning object according to any one of claims 1 to 4.

Images