CN109191345A - A kind of cognitive diagnosis method of Student oriented cognitive process - Google Patents

A kind of cognitive diagnosis method of Student oriented cognitive process Download PDF

Info

Publication number
CN109191345A
CN109191345A CN201811081743.0A CN201811081743A CN109191345A CN 109191345 A CN109191345 A CN 109191345A CN 201811081743 A CN201811081743 A CN 201811081743A CN 109191345 A CN109191345 A CN 109191345A
Authority
CN
China
Prior art keywords
node
exercise
student
knowledge point
knowledge
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201811081743.0A
Other languages
Chinese (zh)
Other versions
CN109191345B (en
Inventor
胡学钢
刘菲
卜晨阳
吴共庆
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN201811081743.0A priority Critical patent/CN109191345B/en
Publication of CN109191345A publication Critical patent/CN109191345A/en
Application granted granted Critical
Publication of CN109191345B publication Critical patent/CN109191345B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/20—Education
    • G06Q50/205—Education administration or guidance

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Educational Administration (AREA)
  • Educational Technology (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of cognitive diagnosis methods of Student oriented cognitive process, comprising the following steps: the cognitive diagnosis analysis of the representing multiple graininess model of building knowledge point and exercise, the school work state table representation model for constructing student's node, student's node.The school work state that the present invention can use the more granularity earth's surface dendrographies of the method for knowledge mapping raw corresponds to the Grasping level of knowledge point according to the answer situation analysis student of student so as to be in different cognitive processes for student.

Description

A kind of cognitive diagnosis method of Student oriented cognitive process
Technical field
The present invention relates to educational data excavation applications, in particular to more granularities cognition of a kind of Student oriented cognitive process is examined Disconnected method.
Background technique
With the development for admiring the open educational resources platform such as class, " internet+education " is promoted to obtain the height of State-level Degree concern and attention.That educational researcher passes through cognitive diagnosis (Cognitive Diagnosis) quantitative assessment student Property difference and human-subject test.There are the magnanimity learning datas of student on internet, and know involved by most of topics Know point and all shows different granularity levels.In the cognitive process of student, student knows grasp in the different study stages The granularity layers time requirement for knowing point is also different.
Therefore, by analyzing real-time learning data, real-time tracking is carried out to the learning state of student, and targetedly It carries out individual instructions and alarming system has great importance.Currently, the DINA model based on knowledge-ID incidence matrix It is the cognitive diagnosis model of mainstream, however, the existing method based on knowledge-ID incidence matrix causes model abundant Indicate the different grain size of knowledge point.
Based on the above circumstances, a kind of more granularity cognitive diagnosis methods for designing reasonable Student oriented cognitive process are especially heavy It wants.
Summary of the invention
The present invention is to propose a kind of reasonable Student oriented cognition to solve above-mentioned the shortcomings of the prior art place The cognitive diagnosis method of process, to the school work state that the more granularity earth's surface dendrographies of method of knowledge mapping can be used raw, so as to It is in different cognitive processes for student, the Grasping level of knowledge point is corresponded to according to the answer situation analysis student of student.
In order to achieve the above object, the technical scheme adopted by the invention is as follows:
A kind of the characteristics of cognitive diagnosis method of Student oriented cognitive process of the present invention, is, comprising the following steps:
(1), the representing multiple graininess model of knowledge point and exercise is constructed:
(1.1), setting knowledge point number is P, exercise number is Q, student's number is M；
(1.2), creation of knowledge point node K={ K1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…,Jq,…, JQ, student node I={ I1,I2,…,Im,…,IM}；Wherein, KpIndicate p-th of knowledge point node, JqIndicate q-th of exercise section Point, ImIndicate m-th of student's node, p=1,2 ..., P, q=1,2 ..., Q, m=1,2 ..., M；
(1.3), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, difficulty value Kp.difficuty；To define the attribute of P knowledge point node；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise is answered Case Jq.answer；To define the attribute of Q exercise node；
Define m-th of student's node ImAttribute be student name Im.name；To define the attribute of M student's node；
(1.4), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
(1.5) if, p-th of knowledge point node KpInclude v-th of knowledge point node Kv, then it represents that p-th of knowledge point node Kp With v-th of knowledge point node KvBetween there are sides, be denoted as L1(Kp,Kv), and L1(Kp,Kv)=1；
If p-th of knowledge point node KpNot comprising v-th of knowledge point node Kv, then L is enabled1(Kp,Kv)=0；V=1,2 ..., P, and v ≠ p；
The knowledge point node division that side is interconnected is a knowledge cluster, to be R by all knowledge point node divisions A knowledge cluster C={ C1,C2,…,Cr,…,CR}；CrIndicate r-th of knowledge cluster, r=1,2 ..., R；
(1.6) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, and be r-th of knowledge cluster CrLeaf section Point, and q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then it represents that q-th of exercise node JqIt is saved with p-th of knowledge point Point KpBetween there are sides, be denoted as L2(Jq,Kp), and L2(Jq,Kp)=1；Otherwise, L is enabled2(Jq,Kp)=0；
(1.7) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, but be not r-th of knowledge cluster CrLeaf section Point, and i-th of knowledge point node KiIt is r-th of knowledge cluster CrLeaf node, and p-th of knowledge point node KpKnow comprising i-th Know point node Ki, q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then q-th of exercise node JqWith i-th of leaf section Point KiBetween there are side L2(Jp,Ki), and L2(Jp,Ki)=1；Otherwise, L is enabled2(Jq,Kp)=0；
(1.8) if, m-th of student's node ImComplete q-th of exercise node JqAnswer, then it represents that m-th of student's node Im With q-th of exercise node JqBetween there are side L3(Im,Jq), and L3(Im,Jq)=1, otherwise L3(Im,Jq)=0；
Side L is set3(Im,Jq) attribute be L3(Im,Jq).ansm, indicate m-th of student's node ImIt answers q-th of exercise section Point JqAnswer；
(1.9), side L is set3(Im,Jq) attribute be L3(Im,Jq) .flag, indicate m-th of student's node ImAnswer q A exercise node JqAnswer it is whether correct；If L3(Im,Jq).ansm=Jq.answer, then L is enabled3(Im,Jq) .flag=1, it is no Then enable L3(Im,Jq) .flag=0；
(1.10), it calculates and q-th of exercise node JqThere are student's node number n on sideq, to obtain and all exercise J There are the student node number n={ n on side1,n2,…,nq,…,nQ}；
It calculates and q-th of exercise node JqThere are student's node number n on sideqIn, m-th of student's node ImIt is practised with q-th Inscribe node JqBetween there are side L3(Im,Jq) attribute L3(Im,Jq) .flag=1 student's node numberTo obtain and institute With the presence of exercise J while attribute be " 1 " student's node number while number
(2), m-th of student's node I is constructedmSchool work state table representation model:
(2.1), knowledge point node K={ K is re-created1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…, Jq,…,JQ, the side L between the node of knowledge point1, side L between exercise node and knowledge point2；
(2.2), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, m-th of student ImTo p-th of knowledge point node KpGrasping level Kp.cognitionm；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise is answered Case Jq.answer, m-th of student ImAnswer Jq.ansm, m-th of student ImReaction time Jq.timem；
(2.3), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
By m-th of student's node ImTo p-th of knowledge point node KpGrasping level Kp.cognition it is set as " -1 "； Thus by m-th of student's node Im" -1 " is disposed as to the Grasping level of all knowledge point nodes；
(3), m-th of student's node ImCognitive diagnosis analysis:
(3.1), q-th of exercise node J is setqInitial difficulty coefficient beTo which the initial difficulty of all exercise J be arranged Spending coefficient is
(3.2), setting iteration total degree is T, and current iteration number is t, and initializes t=1；
(3.3), q-th of exercise node J is calculated by formula (1)qRegulation coefficient wq', to obtain the tune of all exercise J Integral coefficient w '={ w1′,w2′,…,wq′,…,wQ' }:
(3.4), q-th of exercise node J of the t times iteration is updated by formula (2)qDegree-of-difficulty factorTo update t The degree-of-difficulty factor of all exercise J of secondary iteration
(3.5), t+1 is assigned to t, and judges whether t=T is true, if so, then follow the steps (3.6)；Otherwise, it holds Row step (3.4)；
(3.6), enabling condition S is L3(Im,Jq∧ { the L of)=13(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；Item Part S+For L3(Im,Jq∧ { the L of)=13(Im,Jq) .flag=1 ∧ { L3(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；
(3.7), m-th of student's node I is calculated by formula (3)mTo p-th of knowledge point node KpGrasping level Kp.cognitionm:
Compared with prior art, the beneficial effects of the present invention are:
1, present invention employs the method for knowledge mapping, representing multiple graininess model and the student of knowledge point and exercise are constructed School work state table representation model solves knowledge point node in the prior art so as to carry out cognitive diagnosis analysis to student Without granularity hierarchy, so that it can be directed to cognitive process different at student when to student's cognitive diagnosis, at many levels, The school work state status of the student is analyzed in all directions；
2, for exercise, knowledge point and student data extensive in open education platform, lead to the pass in model between data System is far less than data number.The prior art indicates the school work of knowledge point and exercise model and student using the representation method of matrix State model causes matrix sparse.The present invention is based on the representing multiple graininess moulds of the method for knowledge mapping building knowledge point and exercise Type and the school work state table representation model of student, alleviate the Sparse Problems of data, effectively reduce the space complexity of data；
3, since the Different Cognitive stage in student, such as student are respectively at unit test, interim test, final test When, emphasis and have larger difference, more granularities proposed by the present invention to the granularity level of the knowledge point of grasp that student learns Indicate that model shows knowledge point with different granularity levels, so that the model can be for Different Cognitive process student's Emphasis difference carries out diagnostic analysis.
Detailed description of the invention
Fig. 1 is a kind of cognitive diagnosis method flow diagram of Student oriented cognitive process of the invention；
Fig. 2 is the representing multiple graininess model schematic of knowledge point of the invention, exercise；
Fig. 3 is student's school work state table representation model schematic diagram of the invention；
Fig. 4 is exercise, the knowledge dot matrix schematic diagram that traditional teacher uses；
Fig. 5 is the exercise of Student oriented cognitive process of the invention, knowledge dot matrix schematic diagram；
Fig. 6 is the relational graph of " sequence " knowledge cluster and exercise 1,2 in the specific embodiment of the invention.
Specific embodiment
In the present embodiment, a kind of cognitive diagnosis method of Student oriented cognitive process is: being primarily based on knowledge mapping method Construct the representing multiple graininess model of knowledge point and exercise；Secondly based on the representing multiple graininess model of knowledge point and exercise, building is learned Raw school work state table representation model；Again, for cognitive process at student, more granularity cognitive diagnosis point are carried out to student Analysis.Algorithm flow chart is as shown in Figure 1.Specifically, being to carry out as follows:
(1), the representing multiple graininess model of knowledge point and exercise is constructed:
The relationship between knowledge point and the relationship between exercise and knowledge point are indicated using the method for knowledge mapping.Such as Fig. 2 Shown, there are three types of the nodes of type, respectively knowledge point node, exercise node and student's node in knowledge mapping；Side then indicates Between node there are relationships.
(1.1), setting knowledge point number is P, exercise number is Q, student's number is M；
(1.2), creation of knowledge point node K={ K1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…,Jq,…, JQ, student node I={ I1,I2,…,Im,…,IM}；Wherein, KpIndicate p-th of knowledge point node, JqIndicate q-th of exercise section Point, ImIndicate m-th of student's node, p=1,2 ..., P, q=1,2 ..., Q, m=1,2 ..., M；
Specifically, as shown in Fig. 2, rectangle node table shows that knowledge point node, oval node indicate exercise node, pentagon Node table dendrography tight knot point.
(1.3), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, difficulty value Kp.difficuty；To define the attribute of P knowledge point node；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise is answered Case Jq.answer；To define the attribute of Q exercise node；
Define m-th of student's node ImAttribute be student name Im.name；To define the attribute of M student's node；
(1.4), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
(1.5) if, p-th of knowledge point node KpInclude v-th of knowledge point node Kv, then it represents that p-th of knowledge point node Kp With v-th of knowledge point node KvBetween there are sides, be denoted as L1(Kp,Kv), and L1(Kp,Kv)=1；
If p-th of knowledge point node KpNot comprising v-th of knowledge point node Kv, then L is enabled1(Kp,Kv)=0；V=1,2 ..., P, and v ≠ p；
The knowledge point node division that side is interconnected is a knowledge cluster, to be R by all knowledge point node divisions A knowledge cluster C={ C1,C2,…,Cr,…,CR}；CrIndicate r-th of knowledge cluster, r=1,2 ..., R；
Due to knowledge point node may comprising several sub- knowledge points, then by knowledge point and it includes sub- knowledge point formed The knowledge cluster of tree, that is, form the knowledge point structure of different grain size level.Minimum knowledge point in each knowledge cluster is known as Leaf node, i.e., the sub- knowledge point that can not be further segmented in knowledge cluster.
Specifically, as shown in Fig. 2, C1Comprising with knowledge point node K1For six knowledge points of father node, i.e. C1={ K1,K3, K4,K5,K8,K9}；C2Comprising with knowledge point node K2For three knowledge points of father node, i.e. C2={ K2,K6,K7}.Wherein, knowledge Point node K1Include knowledge point node K3、K4、K5, i.e. L1(K1,K3)=1, L1(K1,K4)=1, L1(K1,K5)=1；Knowledge point section Point K4Include knowledge point node K8、K9, i.e. L1(K4,K8)=1, L1(K4,K9)=1；Knowledge point node K2Include knowledge point node K6、 K7, i.e. L1(K2,K6)=1, L1(K2,K7)=1.But knowledge point node K1Not comprising knowledge point node K6And K8, then L1(K1, K6)=0, L1(K1,K8)=0, remaining knowledge point node do not include relationship similarly.
(1.6) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, and be r-th of knowledge cluster CrLeaf section Point, and q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then it represents that q-th of exercise node JqIt is saved with p-th of knowledge point Point KpBetween there are sides, be denoted as L2(Jq,Kp), and L2(Jq,Kp)=1；Otherwise, L is enabled2(Jq,Kp)=0；
Specifically, as shown in Fig. 2, knowledge point node K3It is knowledge cluster C1Leaf node, knowledge point node K6、K7It is to know Know cluster C2Leaf node；Exercise node J1It is related to knowledge point node K3、K6, then exercise node J1With knowledge point node K3、K6It Between there are sides, i.e. L2(J1,K3)=1, L2(J1,K6)=1.Knowledge point node K4It is not knowledge cluster C1Leaf node, then while Exercise node J1It is related to knowledge point node K4, but exercise node J1With knowledge point node K4Between be not present side, i.e. L2(J1,K4)= 0, remaining similar situation is similarly.Exercise node J1It is not related to knowledge point node K5, i.e. L2(J1,K5)=0, remaining similar situation are same Reason.
(1.7) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, but be not r-th of knowledge cluster CrLeaf section Point, and i-th of knowledge point node KiIt is r-th of knowledge cluster CrLeaf node, and p-th of knowledge point node KpKnow comprising i-th Know point node Ki, q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then q-th of exercise node JqWith i-th of leaf section Point KiBetween there are side L2(Jp,Ki), and L2(Jp,Ki)=1；Otherwise, L is enabled2(Jq,Kp)=0；
Since per pass exercise is related to one or more knowledge points, only student is in all knowledge points that grasp exercise is related to In the case of could correct answer, therefore, each exercise node is associated with the minimum knowledge point that it is related to.
Specifically, as shown in Fig. 2, knowledge point node K4It is not knowledge cluster C1Leaf node, knowledge point node K4Comprising knowing Know point node K8And K9, and knowledge point node K8And K9It is knowledge cluster C1Leaf node, if exercise node J1It is related to knowledge point section Point K4, then exercise node J1With knowledge point node K8And K9Between there are sides, i.e. L2(J1,K8)=1, L2(J1,K9)=1.Remaining feelings It is 0 under condition.
(1.8) if, m-th of student's node ImComplete q-th of exercise node JqAnswer, then it represents that m-th of student's node Im With q-th of exercise node JqBetween there are side L3(Im,Jq), and L3(Im,Jq)=1, otherwise L3(Im,Jq)=0；
Side L is set3(Im,Jq) attribute be L3(Im,Jq).ansm, indicate m-th of student's node ImIt answers q-th of exercise section Point JqAnswer；
Specifically, as shown in Fig. 2, student's node I1Complete exercise node J1Answer, then student's node I1With exercise section Point J1Between there are side L3(I1,J1), i.e. L3(I1,J1)=1；It can similarly obtain, L3(I2,J1)=1, L3(I2,J2)=1, L3(I3, J2)=1.Due to student's node I1Exercise node J is not completed2Answer, then student's node I1With exercise node J2Between be not present Side, i.e. L3(I1,J2)=0；It can similarly obtain, L3(I3,J1)=0.
Side L is respectively set3(I1,J1)、L3(I2,J1)、L3(I2,J2)、L3(I3,J2) attribute L3(I1,J1).ans1、L3 (I2,J1).ans2、L3(I2,J2).ans2、L3(I3,J2).ans3, and by student's node I1、I2、I3The answer of corresponding exercise of answering It is assigned to above four attribute values.
(1.9), side L is set3(Im,Jq) attribute be L3(Im,Jq) .flag, indicate m-th of student's node ImAnswer q A exercise node JqAnswer it is whether correct；If L3(Im,Jq).ansm=Jq.answer, then L is enabled3(Im,Jq) .flag=1, it is no Then enable L3(Im,Jq) .flag=0；
Specifically, as shown in Fig. 2, side L is respectively set3(I1,J1)、L3(I2,J1)、L3(I2,J2)、L3(I3,J2) attribute L3(I1,J1).flag、L3(I2,J1).flag、L3(I2,J2).flag、L3(I3,J2).flag.If student's node I1Correctly answer Exercise node J1, i.e. L3(I1,J1).ans1=J1.answer, then L3(I1,J1) .flag=1；If student's node I1Mistake is made Exercise node J is answered1, i.e. L3(I1,J1).ans1≠J1.answer, then L3(I1,J1) .flag=0；Remaining category similarly can be obtained Property value.
(1.10), it calculates and q-th of exercise node JqThere are student's node number n on sideq, to obtain and all exercise J There are the student node number n={ n on side1,n2,…,nq,…,nQ}；
It calculates and q-th of exercise node JqThere are student's node number n on sideqIn, m-th of student's node ImIt is practised with q-th Inscribe node JqBetween there are side L3(Im,Jq) attribute L3(Im,Jq) .flag=1 student's node numberTo obtain and institute With the presence of exercise J while attribute be " 1 " student's node number while number
Specifically, as shown in Fig. 2, with exercise node J1There are student's node number n on side1=2；N can similarly be obtained2=2.
If L3(I1,J1) .flag=1, L3(I2,J1) .flag=1, L3(I2,J2) .flag=0, L3(I3,J2) .flag= 1, then
(2), m-th of student's node I is constructedmSchool work state table representation model:
The relationship between student's exercise answered and knowledge point is indicated using the method for knowledge mapping.As shown in figure 3, knowledge There are two types of the nodes of type, respectively knowledge point node, exercise node in map；Side then indicate between node there are relationships. Assuming that building student's node I2School work state table representation model.
(2.1), knowledge point node K={ K is re-created1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…, Jq,…,JQ, the side L between the node of knowledge point1, side L between exercise node and knowledge point2；
Side between knowledge point node shown in Fig. 3, exercise node and node is multiplexed knowledge point node shown in Fig. 2, practises Inscribe the side between node and node.
(2.2), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, m-th of student ImTo p-th of knowledge point node KpGrasping level Kp.cognitionm；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise is answered Case Jq.answer, m-th of student ImAnswer Jq.ansm, m-th of student ImReaction time Jq.timem；
Compared with the representing multiple graininess model of knowledge point and exercise, in the school work state map of student, each knowledge point Node increases student's node I newly other than knowledge point title and detailed content2To the Grasping level attribute of knowledge point node；Each In exercise node other than storage topic information and answer, student's node I is increased newly2Answer and Reaction time attribute.
(2.3), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
By m-th of student's node ImTo p-th of knowledge point node KpGrasping level Kp.cognition it is set as " -1 "； Thus by m-th of student's node Im" -1 " is disposed as to the Grasping level of all knowledge point nodes；
(3), m-th of student's node ImCognitive diagnosis analysis:
(3.1), q-th of exercise node J is setqInitial difficulty coefficient beTo which the initial difficulty of all exercise J be arranged Spending coefficient is
Knowledge point, the exercise matrix that traditional cognitive diagnosis process uses are as shown in figure 4, different grain size level cannot be embodied Knowledge point grasp situation.Since each student cognitive process locating at present is different, in the new knowledge point study stage The raw granularity layers grasped to knowledge point time requirement is thinner, grasps in knowledge point combing and the student in review stage to knowledge point Granularity layers time require thicker.It therefore, can be according to different students to granularity level in the representing multiple graininess model that the present invention constructs It is different require to targetedly select required granularity level (as shown in Figure 5), to be pointedly more student setting The initial difficulty coefficient of exercise.
Contribution of the present invention on more granularity levels is illustrated as shown in fig. 6, exercise 1 and exercise 2 investigate student couple In the Grasping level of sort algorithm correlated knowledge point, however exercise 1 is the exercise in the sort algorithm study stage, knowledge point It should be divided into thinner granularity, i.e. { " Shell sorting basic conception ", " Shell sorting time complexity " }, to investigate student emphatically For the grasp situation of every sub- knowledge point.And exercise 2 is the examination paper in final examination, due to what is be related in end of term paper Knowledge point is more, if excessively high using the complexity that the knowledge point division methods of finer grain will lead to model solution, therefore, can make It obtains dividing compared with the knowledge point of coarseness with the father node of leaf node in knowledge cluster.For example, divide can be with for the coarseness of exercise 2 For { " Shell sorting ", " bubble sort ", " quicksort ", " heapsort " }；And if using most fine-grained division, every kind " basic conception ", " algorithm description ", " stability " of sort algorithm all should be used as knowledge point relevant to the exercise, lead to model Time complexity with higher when solution.It, can according to the proposed method, in Different Cognitive process based on this In student carry out personality analysis, formulate the initial difficulty coefficient of all exercises.
Assuming that exercise node J shown in Fig. 31、J2Initial difficulty coefficient be respectively
(3.2), setting iteration total degree is T, and current iteration number is t, and initializes t=1；
For simple declaration, it is assumed that iteration total degree is T=3.
(3.3), q-th of exercise node J is calculated by formula (1)qRegulation coefficient wq', to obtain the tune of all exercise J Integral coefficient w '={ w1′,w2′,…,wq′,…,wQ' }:
Exercise node J shown in Fig. 31Regulation coefficientExercise node J2Regulation coefficient
(3.4), q-th of exercise node J of the t times iteration is updated by formula (2)qDegree-of-difficulty factorTo update t The degree-of-difficulty factor of all exercise J of secondary iteration
(3.5), t+1 is assigned to t, and judges whether t=T is true, if so, then follow the steps (3.6)；Otherwise, it holds Row step (3.4)；
As shown in figure 3, exercise node J shown in Fig. 4 can be obtained by iteration T times1、J2Degree-of-difficulty factor As it can be seen that due to the exercise node J that answers1Student correctly answer the topic, exercise node J1Degree-of-difficulty factor substantially Degree reduces；Due to the exercise node J that answers2Student exist and correctly answer and mistake is answered situation, then exercise node J2Difficulty system Number decreases.
(3.6), enabling condition S is L3(Im,Jq∧ { the L of)=13(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；Item Part S+For L3(Im,Jq∧ { the L of)=13(Im,Jq) .flag=1 ∧ { L3(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；
Condition S is meant that: student's node ImAnswered knowledge point node KpThe exercise node J being related toq, or answer Knowledge point node KiThe exercise node J being related toqAnd knowledge point node KpInclude knowledge point node Ki。
Condition S+It is meant that: student's node ImCorrectly answered knowledge point node KpThe exercise node J being related toq, Huo Zhezheng Really answered knowledge point node KiThe exercise node J being related toqAnd knowledge point node KpInclude knowledge point node Ki。
As shown in figure 3, condition S are as follows: for student's node I2, answered and knowledge point node K4Included knowledge point node K8、K9The exercise node J being related to1；Condition S+Are as follows: for student's node I2, correctly answered and knowledge point node K4It include to know Know point node K8、K9The exercise node J being related to1；Student's node I similarly can be obtained2For the condition of other knowledge point nodes.
(3.7), m-th of student's node I is calculated by formula (3)mTo p-th of knowledge point node KpGrasping level Kp.cognitionm:
It can thus be concluded that student's node I2For knowledge point node K4Grasping level
Again as shown in fig. 6, exercise associated there has exercise by taking " basic conception " knowledge point of " Shell sorting " as an example 1, exercise 2, then according to formulaGrasp of this stage of the student to the knowledge point can be calculated Situation.If assuming, the student is in stage learning period Mo, need to have global grasp to " sequence " knowledge point, then should analyze this stage Grasp situation of the student to the knowledge point of " exchange sort " this kind of granularity.It is associated therewith by taking " exchange sort " knowledge point as an example The exercise of connection only has exercise 2, then according to formulaThe student stage in the end of term can be calculated to this The grasp situation of knowledge point.

Claims (1)

1. a kind of cognitive diagnosis method of Student oriented cognitive process, which comprises the following steps:
(1), the representing multiple graininess model of knowledge point and exercise is constructed:
(1.1), setting knowledge point number is P, exercise number is Q, student's number is M；
(1.2), creation of knowledge point node K={ K1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…,Jq,…,JQ, learn Tight knot point I={ I1,I2,…,Im,…,IM}；Wherein, KpIndicate p-th of knowledge point node, JqIndicate q-th of exercise node, Im Indicate m-th of student's node, p=1,2 ..., P, q=1,2 ..., Q, m=1,2 ..., M；
(1.3), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, difficulty value Kp.difficuty；To define the attribute of P knowledge point node；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise answer Jq.answer；To define the attribute of Q exercise node；
Define m-th of student's node ImAttribute be student name Im.name；To define the attribute of M student's node；
(1.4), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
(1.5) if, p-th of knowledge point node KpInclude v-th of knowledge point node Kv, then it represents that p-th of knowledge point node KpWith V knowledge point node KvBetween there are sides, be denoted as L1(Kp,Kv), and L1(Kp,Kv)=1；
If p-th of knowledge point node KpNot comprising v-th of knowledge point node Kv, then L is enabled1(Kp,Kv)=0；V=1,2 ..., P, and v ≠p；
The knowledge point node division that side is interconnected is a knowledge cluster, is known to be R for all knowledge point node divisions Know cluster C={ C1,C2,…,Cr,…,CR}；CrIndicate r-th of knowledge cluster, r=1,2 ..., R；
(1.6) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, and be r-th of knowledge cluster CrLeaf node, and Q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then it represents that q-th of exercise node JqWith p-th of knowledge point node Kp Between there are sides, be denoted as L2(Jq,Kp), and L2(Jq,Kp)=1；Otherwise, L is enabled2(Jq,Kp)=0；
(1.7) if, p-th of knowledge point node KpBelong to r-th of knowledge cluster Cr, but be not r-th of knowledge cluster CrLeaf node, And i-th of knowledge point node KiIt is r-th of knowledge cluster CrLeaf node, and p-th of knowledge point node KpInclude i-th of knowledge Point node Ki, q-th of exercise node JqIt is related to p-th of knowledge point node Kp, then q-th of exercise node JqWith i-th of leaf node KiBetween there are side L2(Jp,Ki), and L2(Jp,Ki)=1；Otherwise, L is enabled2(Jq,Kp)=0；
(1.8) if, m-th of student's node ImComplete q-th of exercise node JqAnswer, then it represents that m-th of student's node ImWith Q exercise node JqBetween there are side L3(Im,Jq), and L3(Im,Jq)=1, otherwise L3(Im,Jq)=0；
Side L is set3(Im,Jq) attribute be L3(Im,Jq).ansm, indicate m-th of student's node ImAnswer q-th of exercise node Jq Answer；
(1.9), side L is set3(Im,Jq) attribute be L3(Im,Jq) .flag, indicate m-th of student's node ImIt answers q-th of exercise Node JqAnswer it is whether correct；If L3(Im,Jq).ansm=Jq.answer, then L is enabled3(Im,Jq) .flag=1, otherwise enable L3 (Im,Jq) .flag=0；
(1.10), it calculates and q-th of exercise node JqThere are student's node number n on sideq, to obtain existing with all exercise J The student node number n={ n on side1,n2,…,nq,…,nQ}；
It calculates and q-th of exercise node JqThere are student's node number n on sideqIn, m-th of student's node ImWith q-th of exercise section Point JqBetween there are side L3(Im,Jq) attribute L3(Im,Jq) .flag=1 student's node numberTo obtain and all habits Inscribe J there are while attribute be " 1 " student's node number while number
(2), m-th of student's node I is constructedmSchool work state table representation model:
(2.1), knowledge point node K={ K is re-created1,K2,…,Kp,…,KP, exercise node J={ J1,J2,…,Jq,…, JQ, the side L between the node of knowledge point1, side L between exercise node and knowledge point2；
(2.2), p-th of knowledge point node K is definedpAttribute include: knowledge point title Kp.name, detailed content Kp.context, m-th of student ImTo p-th of knowledge point node KpGrasping level Kp.cognitionm；
Define q-th of exercise node JqAttribute include: exercise content Jq.name, exercise option Jq.option, exercise answer Jq.answer, m-th of student ImAnswer Jq.ansm, m-th of student ImReaction time Jq.timem；
(2.3), the attribute value of P knowledge point node, Q exercise node and M student's node is set；
By m-th of student's node ImTo p-th of knowledge point node KpGrasping level Kp.cognition it is set as " -1 "；To By m-th of student's node Im" -1 " is disposed as to the Grasping level of all knowledge point nodes；
(3), m-th of student's node ImCognitive diagnosis analysis:
(3.1), q-th of exercise node J is setqInitial difficulty coefficient beTo which the initial difficulty system of all exercise J be arranged Number is
(3.2), setting iteration total degree is T, and current iteration number is t, and initializes t=1；
(3.3), q-th of exercise node J is calculated by formula (1)qRegulation coefficient wq', to obtain the adjustment system of all exercise J Number w '={ w1′,w2′,…,wq′,…,wQ' }:
(3.4), q-th of exercise node J of the t times iteration is updated by formula (2)qDegree-of-difficulty factorTo update the t times repeatedly The degree-of-difficulty factor of all exercise J in generation
(3.5), t+1 is assigned to t, and judges whether t=T is true, if so, then follow the steps (3.6)；Otherwise, step is executed Suddenly (3.4)；
(3.6), enabling condition S is L3(Im,Jq∧ { the L of)=13(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；Condition S+ For L3(Im,Jq∧ { the L of)=13(Im,Jq) .flag=1 ∧ { L3(Jq,Kp∨ { the L of)=13(Jq,KiThe ∧ K of)=1pInclude Ki}}；
(3.7), m-th of student's node I is calculated by formula (3)mTo p-th of knowledge point node KpGrasping level Kp.cognitionm:
CN201811081743.0A 2018-09-17 2018-09-17 Cognitive diagnosis method for student cognitive process Active CN109191345B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811081743.0A CN109191345B (en) 2018-09-17 2018-09-17 Cognitive diagnosis method for student cognitive process

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811081743.0A CN109191345B (en) 2018-09-17 2018-09-17 Cognitive diagnosis method for student cognitive process

Publications (2)

Publication Number Publication Date
CN109191345A true CN109191345A (en) 2019-01-11
CN109191345B CN109191345B (en) 2021-06-29

Family

ID=64911594

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811081743.0A Active CN109191345B (en) 2018-09-17 2018-09-17 Cognitive diagnosis method for student cognitive process

Country Status (1)

Country Link
CN (1) CN109191345B (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110765278A (en) * 2019-10-24 2020-02-07 深圳小蛙出海科技有限公司 Method for searching similar exercises, computer equipment and storage medium
CN110969918A (en) * 2019-11-29 2020-04-07 广西师范大学 Method and system for reproducing wiring behavior process of student electrical experiment
CN111159419A (en) * 2019-12-09 2020-05-15 浙江师范大学 Knowledge tracking data processing method, system and storage medium based on graph convolution
CN111782815A (en) * 2019-04-04 2020-10-16 北京三好互动教育科技有限公司 Knowledge evaluation method and device, storage medium and electronic equipment
CN113221007A (en) * 2021-05-21 2021-08-06 合肥工业大学 Method for recommending answering behavior
CN113344204A (en) * 2021-06-10 2021-09-03 合肥工业大学 Cognitive data processing method and device for multiple logic problems
CN113361867A (en) * 2021-05-17 2021-09-07 山东师范大学 Concept importance judging method and system based on student answer records
CN117273130A (en) * 2023-11-13 2023-12-22 南京信息工程大学 Knowledge graph and individual capability-based cognitive diagnosis state machine implementation method

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665473A (en) * 2016-07-27 2018-02-06 科大讯飞股份有限公司 Learning path planning method and device

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107665473A (en) * 2016-07-27 2018-02-06 科大讯飞股份有限公司 Learning path planning method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
朱臻: ""基于知识图谱的初中英语选题系统及应用"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111782815A (en) * 2019-04-04 2020-10-16 北京三好互动教育科技有限公司 Knowledge evaluation method and device, storage medium and electronic equipment
CN110765278B (en) * 2019-10-24 2022-10-25 深圳小蛙出海科技有限公司 Method for searching similar exercises, computer equipment and storage medium
CN110765278A (en) * 2019-10-24 2020-02-07 深圳小蛙出海科技有限公司 Method for searching similar exercises, computer equipment and storage medium
CN110969918A (en) * 2019-11-29 2020-04-07 广西师范大学 Method and system for reproducing wiring behavior process of student electrical experiment
CN110969918B (en) * 2019-11-29 2021-07-13 广西师范大学 Method and system for reproducing wiring behavior process of student electrical experiment
CN111159419A (en) * 2019-12-09 2020-05-15 浙江师范大学 Knowledge tracking data processing method, system and storage medium based on graph convolution
CN113361867A (en) * 2021-05-17 2021-09-07 山东师范大学 Concept importance judging method and system based on student answer records
CN113221007A (en) * 2021-05-21 2021-08-06 合肥工业大学 Method for recommending answering behavior
CN113221007B (en) * 2021-05-21 2022-09-23 合肥工业大学 Method for recommending answering behavior
CN113344204A (en) * 2021-06-10 2021-09-03 合肥工业大学 Cognitive data processing method and device for multiple logic problems
CN113344204B (en) * 2021-06-10 2022-11-18 合肥工业大学 Cognitive data processing method and device for multiple logic problems
CN117273130A (en) * 2023-11-13 2023-12-22 南京信息工程大学 Knowledge graph and individual capability-based cognitive diagnosis state machine implementation method
CN117273130B (en) * 2023-11-13 2024-02-23 南京信息工程大学 Knowledge graph and individual capability-based cognitive diagnosis state machine implementation method

Also Published As

Publication number Publication date
CN109191345B (en) 2021-06-29

Similar Documents

Publication Publication Date Title
CN109191345A (en) A kind of cognitive diagnosis method of Student oriented cognitive process
CN113127731B (en) Personalized test question recommendation method based on knowledge graph
CN107862970B (en) Teaching quality evaluation model for turnover classroom
CN107122452A (en) Student's cognitive diagnosis method of sequential
CN111159419B (en) Knowledge tracking data processing method, system and storage medium based on graph convolution
CN112116092B (en) Interpretable knowledge level tracking method, system and storage medium
Bi et al. Quality meets diversity: A model-agnostic framework for computerized adaptive testing
Turabieh Hybrid machine learning classifiers to predict student performance
CN106203534A (en) A kind of cost-sensitive Software Defects Predict Methods based on Boosting
Minn et al. Improving knowledge tracing model by integrating problem difficulty
Huo et al. Towards personalized learning through class contextual factors-based exercise recommendation
Chen et al. Recommendation system based on rule-space model of two-phase blue-red tree and optimized learning path with multimedia learning and cognitive assessment evaluation
CN111898803B (en) Problem prediction method, system, equipment and storage medium
Dai et al. Knowledge tracing: A review of available technologies
CN111311997A (en) Interaction method based on network education resources
Budiman et al. Data mining implementation using naïve Bayes algorithm and decision tree J48 in determining concentration selection
Pattiasina et al. Comparison of data mining classification algorithm for predicting the performance of high school students
Kang et al. Personalized exercise recommendation via implicit skills
Xie et al. SQKT: A student attention-based and question-aware model for knowledge tracing
Wang et al. Gaskt: A graph-based attentive knowledge-search model for knowledge tracing
Chen Q-matrix optimization for cognitive diagnostic assessment
Dai et al. Knowledge Tracing: A Review of Available Techniques.
Chen et al. Design of Assessment Judging Model for Physical Education Professional Skills Course Based on Convolutional Neural Network and Few‐Shot Learning
Huang et al. An adaptive cellular genetic algorithm based on selection strategy for test sheet generation
CN114155124B (en) Test question resource recommendation method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant