Cognitive evaluation method based on attribute-oriented concept reduction

Abstract

The invention discloses a cognitive evaluation method based on attribute-oriented concept reduction. The method comprises the following steps: mapping tau to construct a formal background based on knowledge points; generating an object intuition graph HQ about the question based on the formal background; all object concepts of the attribute-oriented concept lattice are obtained according to the object intuition graph HQ, an attribute-oriented concept coordination set is obtained, redundant concepts in the attribute-oriented concept coordination set are deleted, and an attribute-oriented concept reduction F is obtained; and according to the student answering data set and the attribute-oriented concept reduction, performing cognitive assessment on the student to obtain an assessment result. According to the method, attribute-oriented concept reduction is obtained through the object intuitive graph according to the binary relation between the problem and the knowledge point, the core structure is extracted from the complex relation between the problem and the knowledge point, the cognitive evaluation process is simplified, meanwhile, redundant information is reduced, key education information is reserved, the operation amount of a computer is greatly reduced, and the cognitive evaluation efficiency is improved. And the operation speed is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of educational information analysis, and in particular relates to a cognitive evaluation method based on attribute-oriented concept reduction. Background Art

[0002] Over the past few decades, with the rapid development of educational informatization and Internet technology, the data collection and analysis capabilities in educational scenarios have been greatly improved. The popularity of various intelligent terminal devices and online learning platforms has enabled students' learning behavior data, answer records and other information to be collected and processed in real time. These data can not only reflect students' learning progress and performance, but also deeply reveal their mastery of knowledge points and cognitive characteristics. These massive amounts of student behavior data provide favorable conditions for cognitive assessment research.

[0003] The Journal of Nanjing University (Natural Science) Vol. 59 No. 4 published "Knowledge Assessment and Learning Path Selection under Knowledge Point Network". This method uses formal concept analysis to reveal the intrinsic connection between the problem set and the knowledge point set. The corresponding relationship between the problem and the knowledge point is identified through the attribute-oriented concept lattice, thereby revealing the underlying cognitive structure. However, this method has some shortcomings in dealing with large-scale data in cognitive assessment. First, with the continuous increase in the number of exercises and knowledge point data, the scale and complexity of the formal background also increase. The attribute concept lattice needs to comprehensively model the combination of all attributes, resulting in an increase in the computational complexity of constructing the attribute concept lattice, which will cause a huge computational burden on the computer. Secondly, the attribute-oriented concept lattice of this method is prone to contain redundant information. Since the attribute-oriented concept lattice describes the data relationship through all possible attribute combinations, it often generates some irrelevant or redundant attributes. This redundant information may interfere with the accuracy of the evaluation results, increase the complexity of the analysis, and reduce the efficiency of cognitive assessment. In addition, the attribute-oriented concept lattice of this method is usually static and lacks the ability to adapt to dynamic changes. When the data or attributes change, the concept lattice needs to be reconstructed. Therefore, the real-time and flexibility of this method are poor. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a cognitive assessment method based on attribute-oriented concept reduction that has a reasonable design and realizes more precise teaching intervention and personalized learning support.

[0005] The technical solution adopted to solve the above technical problems is: a cognitive evaluation method based on attribute concept reduction, comprising the following steps:

[0006] Step 1. Construct the formal background (Q, D, R) based on the knowledge point mapping τ, where Q is the problem set, D is the set of knowledge points associated with the problem, and R is the binary relationship between the problem and the knowledge point;

[0007] Step 2. Formal background (Q, D, R) Generate the labeled question set G of the question set Q according to the formal concept analysis method, construct the corresponding hierarchical structure according to the partial order relationship between the labeled question classes in the labeled question set G, and obtain the object intuitive graph H about the problem Q , object intuitive diagram H Q Each node in represents a labeled question class, and the edges between nodes represent partial order relations;

[0008] Step 3. According to the object intuitive map H Q Obtain all object concepts of the attribute-oriented concept lattice, obtain the attribute-oriented concept coordination set, delete the redundant concepts in the attribute-oriented concept coordination set, and obtain the attribute-oriented concept reduction F;

[0009] The redundant concept is an object concept that does not affect the integrity of the complementary binary relationship between the attribute-oriented concept coordination set reconstruction problem and the knowledge points;

[0010] Step 4. Test students with question set Q, collect students’ test results, remove the results of “careless wrong answers” ​​and “guessing the right answers”, obtain the students’ answer data set, conduct cognitive assessment of students based on the students’ answer data set and attribute-oriented concept reduction, and obtain the assessment results.

[0011] As a preferred technical solution, the method for generating a labeled question set G of the question set Q from the formal background (Q, D, R) according to the formal concept analysis method is:

[0012] Step 2.1. For each question q∈Q, find all d that satisfy (q,d)∈R, where d is the knowledge point related to question q, d∈D;

[0013] Step 2.2. Build a labeled question class

[0014] Each labeled question class is a pair (A, B), where A is a subset of the question set Q and B is a subset of the knowledge point set D, satisfying the following conditions:

[0015] Condition 1: The set of knowledge points common to all questions in A is B;

[0016] Condition 2: All question sets with knowledge point B are A;

[0017] Step 2.3. Systematically traverse the formal background and use the Galois connection "*" operation in the formal concept analysis method to generate labeled question classes;

[0018] The Galois connection "*" operation is: for any problem set Calculate its knowledge point closure X′; for any knowledge point set Calculate the problem closure Y′;

[0019] The method for generating the labeled question class is: by traversing all question sets X and knowledge point sets Y, and applying the "*" operation, all pairs (A, B) satisfying X'=B and Y'=A are generated, and (A, B) is the labeled question class;

[0020] Step 2.4. Collect all generated labeled question classes (A, B) to form a labeled question class set G.

[0021] As a preferred technical solution, the object intuitive map H Q The method to obtain all object concepts of the attribute-oriented concept lattice and obtain the attribute-oriented concept coordination set is:

[0022] Step 3.1. From the object intuitive graph H Q In , for each labeled question class, all the labeled question classes related to it are found downward according to the partial order structure to form a related set;

[0023] Step 3.2. For each question set in the marked question class, combine it with all question sets in the related set as the question set of the new question class. For each knowledge point set in the marked question class, directly use it as the knowledge point set of the new question class to obtain the new question class as the object concept in the attribute-oriented concept lattice.

[0024] Step 3.3. Collect all new problem classes to form a set F Q , the set F Q That is, it is an attribute-oriented concept coordination set.

[0025] As a preferred technical solution, the method for cognitively evaluating students based on student answer data sets and attribute-oriented concept simplification is: matching the student answer data with the object concepts obtained in the attribute-oriented concept simplification, analyzing the student's performance on each knowledge point, and determining which knowledge points the students perform well on and which knowledge points they have difficulties on.

[0026] The beneficial effects of the present invention are as follows:

[0027] Based on the binary relationship between questions and knowledge points, the present invention obtains attribute-oriented concept simplification through object intuitive diagrams, extracts the core structure from the relationship between complex questions and knowledge points, simplifies the cognitive assessment process, reduces redundant information, retains key educational information, greatly reduces the amount of computer operation, improves the operating speed, and improves data processing efficiency, especially for large-scale educational scenarios. At the same time, removing redundant information can make the assessment process more focused, reduce information interference, and thus improve the accuracy and precision of the assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of the cognitive evaluation method based on attribute concept reduction of the present invention.

[0029] Figure 2 It is a schematic diagram of constructing a formal background based on knowledge point mapping τ in an embodiment of the present invention.

[0030] Figure 3 It is an intuitive diagram of the object H of the problem in the embodiment of the present invention. Q .

[0031] Figure 4 It is a schematic diagram of the attribute-oriented concept reduction process in an embodiment of the present invention.

[0032] Figure 5 It is a schematic diagram of cognitive diagnosis through concept simplification and student answer data set of the present invention. DETAILED DESCRIPTION

[0033] The present invention will be further described in detail below in conjunction with the accompanying drawings and examples, but the present invention is not limited to the following embodiments.

[0034] Example

[0035] exist Figure 1 In the embodiment, the cognitive evaluation method based on attribute concept reduction includes the following steps:

[0036] Step 1. Construct the formal background (Q, D, R) based on the knowledge point mapping τ, where Q is the problem set, D is the set of knowledge points associated with the problem, and R is the relationship between the problem and the knowledge point. Q = {problem 1, problem 2, problem 3, problem 4, problem 5} is abbreviated as Q = {1, 2, 3, 4, 5}, D = {knowledge point a, knowledge point b, knowledge point c, knowledge point d, knowledge point e} is abbreviated as D = {a, b, c, d, e}. In the formal background, "×" indicates that the problem and the knowledge point are related, and a blank indicates that the problem and the knowledge point are not related, such as Figure 2 ;

[0037] Step 2. Formal background (Q, D, R) Generate the labeled question set G of the question set Q according to the formal concept analysis method, construct the corresponding hierarchical structure according to the partial order relationship between the labeled question classes in the labeled question set G, and obtain the object intuitive graph H about the problem Q , object intuitive diagram H Q Each node in represents a labeled question class, and the edges between nodes represent partial order relations, such as Figure 3 ;

[0038] Among them, the method of generating the labeled question set G of the question set Q according to the formal concept analysis method from the formal background (Q, D, R) is as follows:

[0039] Step 2.1. For each question q∈Q, find all d that satisfy (q,d)∈R, where d is the knowledge point related to question q, d∈D;

[0040] Step 2.2. Build a class of labeled questions

[0041] Each labeled question class is a pair (A, B), where A is a subset of the question set Q and B is a subset of the knowledge point set D, satisfying the following conditions:

[0042] Condition 1: The set of knowledge points common to all questions in A is B;

[0043] Condition 2: All question sets with knowledge point B are A;

[0044] Step 2.3. Systematically traverse the formal background and use the Galois connection "*" operation in the formal concept analysis method to generate labeled question classes;

[0045] The Galois connection "*" operation is: for any problem set Calculate its knowledge point closure X′; for any knowledge point set Calculate the problem closure Y′;

[0046] By traversing all question sets X and knowledge point sets Y and applying the “*” operation, all pairs (A, B) that satisfy X′=B and Y′=A are generated, where (A, B) is a labeled question class;

[0047] Step 2.4. Collect all generated labeled question classes (A, B) to form a labeled question class set G. The labeled question class set G in this embodiment = {({1},{a,c,d,e}),({2},{a,c}),({3},{b,e}),({4},{a}),({5},{a,b,e})}.

[0048] Step 3. Figure 4 , according to the object intuitive diagram H Q Obtain all object concepts of the attribute-oriented concept lattice to obtain the attribute-oriented concept coordination set, delete the redundant concepts in the attribute-oriented concept coordination set, and obtain the attribute-oriented concept reduction, wherein the redundant concepts are object concepts that do not affect the integrity of the complementary binary relationship between the attribute-oriented concept coordination set reconstruction problem and the knowledge points;

[0049] Among them, according to the object intuitive map H Q The method to obtain all object concepts of the attribute-oriented concept lattice and obtain the attribute-oriented concept coordination set is:

[0050] Step 3.1. From the object intuitive graph H Q In , for each labeled question class (A, B), all the labeled question classes related to it are found downward according to the partial order structure to form a related set;

[0051] In this embodiment, the relevant set of the labeled question class ({1},{a,c,d,e}) is {({2},{a,c}),({4},{a})};

[0052] The relevant set of the labeled question class ({2},{a,c}) is {({4},{a})};

[0053] The relevant set of the labeled question class ({3},{b,e}) is an empty set;

[0054] The relevant set with the labeled question class ({4},{a}) is an empty set;

[0055] The relevant set for the labeled question class ({5},{a,b,e}) is {({3},{b,e}),({4},{a})}.

[0056] Step 3.2. For each question set in the marked question class, combine it with all question sets in the related set as the question set of the new question class. For each knowledge point set in the marked question class, directly use it as the knowledge point set of the new question class to obtain the new question class as the object concept in the attribute-oriented concept lattice.

[0057] The new question classes obtained in this embodiment are ({1,2,4},{a,c,d,e}), ({2,4},{a,c}), ({3},{b,e}), ({4},{a}), and ({3,4,5},{a,b,e}).

[0058] Step 3.3. Collect all new problem classes to form a set F Q ={({1,2,4},{a,c,d,e}),({2,4},{a,c}),({3},{b,e}),({4},{a}),({3,4,5},{a,b,e})}, the set F Q That is, it is an attribute-oriented concept coordination set.

[0059] Among them, ({4},{a}) is deleted as a redundant concept, and the resulting attribute-oriented concept reduction F = {({1,2,4},{a,c,d,e}), ({2,4},{a,c}), ({3},{b,e}), ({3,4,5},{a,b,e})}.

[0060] Step 4. Test students with question set Q, collect students’ test results, remove students’ “careless wrong answers” ​​and “guessing right answers” ​​results, obtain students’ answer data set, conduct cognitive assessment on students based on students’ answer data set and attribute-oriented concept reduction, and obtain assessment results, specifically: match students’ answer data with the question set in attribute-oriented concept reduction, determine the knowledge point set corresponding to the questions answered by students; analyze students’ performance on each knowledge point, determine which knowledge points students perform well on and which knowledge points they have difficulties on, such as Figure 5 .

Claims

1. A cognitive assessment method based on attribute-oriented concept reduction, characterized in that: The following steps are involved: Step 1. Construct the formal background (Q, D, R) based on the knowledge point mapping τ, where Q is the problem set, D is the set of knowledge points associated with the problem, and R is the binary relationship between the problem and the knowledge point; Step 2. Formal background (Q, D, R) Generate the labeled question set G of the question set Q according to the formal concept analysis method, construct the corresponding hierarchical structure according to the partial order relationship between the labeled question classes in the labeled question set G, and obtain the object intuitive graph H about the problem Q , object intuitive diagram H Q Each node in represents a labeled question class, and the edges between nodes represent partial order relations; Step 3. According to the object intuitive map H Q Obtain all object concepts of the attribute-oriented concept lattice, obtain the attribute-oriented concept coordination set, delete the redundant concepts in the attribute-oriented concept coordination set, and obtain the attribute-oriented concept reduction F; The redundant concept is an object concept that does not affect the integrity of the complementary binary relationship between the attribute-oriented concept coordination set reconstruction problem and the knowledge points; Step 4. Test students with question set Q, collect students' test results, remove the results of "careless wrong answers" and "guessing the right answers", obtain the student answer data set, conduct cognitive assessment of students based on the student answer data set and attribute-oriented concept reduction, and obtain the assessment results.

2. The cognitive assessment method based on attribute concept reduction according to claim 1 is characterized in that: The method for generating the labeled question set G of the question set Q according to the formal concept analysis method is as follows: Step 2.

1. For each question q∈Q, find all d that satisfy (q,d)∈R, where d is the knowledge point related to question q, d∈D; Step 2.

2. Build a labeled question class Each labeled question class is a pair (A, B), where A is a subset of the question set Q and B is a subset of the knowledge point set D, satisfying the following conditions: Condition 1: The set of knowledge points common to all questions in A is B; Condition 2: All question sets with knowledge point B are A; Step 2.

3. Systematically traverse the formal background and use the Galois connection "*" operation in the formal concept analysis method to generate labeled question classes; The Galois connection "*" operation is: for any problem set Calculate its knowledge point closure X′; for any knowledge point set Calculate the problem closure Y′; The method for generating the labeled question class is: by traversing all question sets X and knowledge point sets Y, and applying the "*" operation, all pairs (A, B) satisfying X'=B and Y'=A are generated, and (A, B) is the labeled question class; Step 2.

4. Collect all generated labeled question classes (A, B) to form a labeled question class set G.

3. The cognitive assessment method based on attribute concept reduction according to claim 1 is characterized in that: The object intuitive map H Q The method to obtain all object concepts of the attribute-oriented concept lattice and obtain the attribute-oriented concept coordination set is: Step 3.

1. From the object intuitive graph H Q In , for each labeled question class, all the labeled question classes related to it are found downward according to the partial order structure to form a related set; Step 3.

2. For each question set in the marked question class, combine it with all question sets in the related set as the question set of the new question class. For each knowledge point set in the marked question class, directly use it as the knowledge point set of the new question class to obtain the new question class as the object concept in the attribute-oriented concept lattice. Step 3.

3. Collect all new problem classes to form a set F Q , the set F Q That is, it is an attribute-oriented concept coordination set.

4. The cognitive assessment method based on attribute concept reduction according to claim 1 is characterized in that: The method for cognitively evaluating students based on student answer data sets and attribute-oriented concept simplification is as follows: matching the student answer data with the object concepts obtained in the attribute-oriented concept simplification, analyzing the student's performance on each knowledge point, and determining which knowledge points the student performs well on and which knowledge points he has difficulties on.

Images