CN111930960A - Knowledge graph technology-based optical transport network knowledge testing method - Google Patents

Abstract

The invention relates to a method for testing knowledge of an optical transport network based on a knowledge graph technology, which comprises the steps of constructing a knowledge point graph in the field of optical transport network OTN, and designing a method for representing a knowledge space by combining knowledge points and incidence relations thereof; further designing rules for extracting knowledge state sets from the knowledge space, wherein each knowledge state element in the knowledge state sets represents a possible knowledge state of the learner; and finally, designing a calculation method of the current knowledge state of the learner, deducing the knowledge point mastering level of the learner according to the knowledge point test result on the basis of automatically designing the knowledge point test sequence, and determining the current knowledge state of the learner from the knowledge state set.

Description

Knowledge graph technology-based optical transport network knowledge testing method

Technical Field

The invention relates to an intelligent testing method for the mastery level of Optical Transport Network (OTN) related knowledge points organized in the form of a knowledge graph, belonging to the fields of network learning and knowledge evaluation. The method is mainly applied to the intelligent testing of each knowledge point and the rapid determination of the learner's competence level by means of the structural systematization characteristics of the knowledge map in the process of learning the knowledge related to the optical transmission network.

Background

The Optical Transport Network (OTN) technology flexibly provides various high-speed interfaces, supports transparent transmission of various service data, has strong network management capability and network protection capability, and has replaced the conventional Synchronous Digital Hierarchy (SDH) technology to form a high-capacity high-speed power communication backbone transmission network. Compared with SDH, the OTN technology principle, the equipment structure and the operation method are more complex, and the related knowledge points are fine and complicated. In order to better understand, learn and master the OTN technology, knowledge points related to the OTN technology need to be structured and organized, and on the basis of the structural and structural knowledge points, the OTN knowledge level of a learner is quickly and accurately evaluated, so as to develop learning training activities in a targeted manner, and improve learning efficiency and learning effect.

The concept of Knowledge Graph (Knowledge Graph) was formally proposed by Google (Google) 5, 17 months in 2012, and the construction of the next-generation intelligent search engine based on the concept was announced. The key technology comprises the steps of extracting entities and attribute information thereof from information resources such as web pages of the Internet and the like, and creating a brand-new information retrieval mode by the relationship between the entities. The domain knowledge map is constructed based on professional domain data, has strict and rich data modes, and can be regarded as a professional knowledge base based on semantic technology. The OTN knowledge point diagram spectrum established based on the domain knowledge diagram concept and technology can be used for accurately expressing a complex knowledge system in the OTN domain. Because the knowledge point atlas gives the incidence relation among the knowledge points while expressing knowledge, the relation reflects the cognitive logic of the learner on each knowledge point to a certain extent, and a high-efficiency test algorithm and a test scheme can be designed based on the relation, so that the whole knowledge mastering level of the learner can be quickly obtained under the condition of testing as few knowledge points as possible, and a basis is provided for targeted knowledge learning and skill drilling. At present, a mature rapid test method suitable for the OTN knowledge graph has not been developed yet.

Disclosure of Invention

The invention aims to provide an intelligent testing method of Optical Transport Network (OTN) knowledge based on knowledge-graph technology, which realizes the rapid and accurate testing of the knowledge level of the OTN of learners by constructing and updating a knowledge state set in a domain knowledge space.

The technical scheme of the invention is as follows:

a testing method of knowledge of an optical transport network based on knowledge graph technology comprises the steps of constructing a knowledge point graph in the field of optical transport network OTN, and designing a representation method of a knowledge space by combining knowledge points and incidence relations thereof; further designing rules for extracting knowledge state sets from the knowledge space, wherein each knowledge state element in the knowledge state sets represents a possible knowledge state of the learner; and finally, designing a calculation method of the current knowledge state of the learner, deducing the knowledge point mastering level of the learner according to the knowledge point test result on the basis of automatically designing the knowledge point test sequence, and determining the current knowledge state of the learner from the knowledge state set.

Preferably, the method for testing knowledge of the optical transport network based on the knowledge graph technology specifically comprises the following steps:

constructing an OTN knowledge point map and a knowledge space of an optical transport network;

secondly, constructing a knowledge state set;

thirdly, designing a testing sequence of knowledge points;

and (IV) determining the current knowledge state of the learner.

Preferably, the step (one) of constructing the OTN knowledge point map and knowledge of the optical transport network specifically includes: designing an overall architecture of an OTN knowledge point map based on a knowledge map technology; forming nodes in the knowledge point map by the knowledge points of the optical transport network, and forming edges in the knowledge point map by the incidence relation among the knowledge points of the optical transport network; the types of the association relation before the knowledge point in the knowledge point map are 2 types, namely, the inclusion relation and the leading relation.

Preferably, the incidence relation before the knowledge point in the knowledge point map is the edge between the node and the node in the map;

the inclusion relationship is: the knowledge points containing other knowledge points in the containing relation are called parent knowledge points, and the contained knowledge points are called child knowledge points; the pilot relationship is: the pilot relation has directionality, starts from the pilot knowledge point and ends at the subsequent knowledge point, and the precondition for mastering the subsequent knowledge point is to master the pilot knowledge point.

Preferably, the step (ii) of constructing the knowledge state set specifically includes: the learner's mastery level at a certain knowledge point is set to be in two states of mastered and not mastered; and (3) setting a question for the learner aiming at the knowledge point, and checking the answer accuracy to determine which state the learner is in on the knowledge point, if the answer accuracy is higher than a set accuracy threshold, marking the state of the knowledge point as 'mastered', otherwise, marking the state of the knowledge point as 'not mastered'.

Preferably, the knowledge state is a set of all knowledge points that characterize the learner's level of mastery in the knowledge space to "mastered", and the knowledge state set is a set of all possible knowledge states, and is constructed as follows:

(1) the knowledge points contained in the knowledge state have the knowledge points of the children if the knowledge points of the parents exist, namely the knowledge points of the parents have the knowledge points of the children if the knowledge points of the parents are mastered;

(2) if there is a successor knowledge point, there must be its predecessor knowledge point, that is, the successor knowledge point must have its predecessor knowledge point.

Preferably, in the step (iii), the association relationship between knowledge points in the OTN knowledge point atlas of the optical transport network is designed in the knowledge point test sequence, and the cognitive logic of the learner on each knowledge point is also implied, so as to draw the following inference:

(1) if the precursor knowledge point is not mastered, the subsequent knowledge point is not mastered with a certain or large probability;

(2) if the child knowledge point is not mastered, the parent knowledge point is not mastered with a certain or large probability;

(3) all child knowledge points contained in the parent knowledge points are mastered, and the parent knowledge points are mastered with a certain probability or a large probability.

Preferably, the inference rule for setting the knowledge point grasp level is as follows:

(1) if the test result of the subsequent knowledge point is mastered, the pilot knowledge point is not required to be tested again and is directly marked as mastered;

(2) if the test result of the child generation knowledge point is not mastered, the parent generation knowledge point does not need to be tested again and is directly marked as not mastered;

(3) if the test results of all the child knowledge points contained in the parent knowledge points are mastered, the parent knowledge points do not need to be tested again and are directly marked as mastered;

in order to reduce the number of the tested knowledge points, leaf knowledge points are tested firstly, pilot knowledge points corresponding to the leaf knowledge points are tested secondly, ancestor knowledge points of the leaf knowledge points are tested secondly, and then the knowledge point testing sequence of the pilot knowledge points corresponding to the ancestor knowledge points is tested secondly.

Preferably, the step (four) of determining the current knowledge state of the learner includes: further designing a method for updating the knowledge state set based on the test result to finally determine the current knowledge state of the learner; if the knowledge level of a certain knowledge point is marked as mastered, the knowledge state containing the knowledge point is reserved in the knowledge state set; if the knowledge level of a certain knowledge point is marked as not mastered, removing the knowledge state containing the knowledge point from the knowledge state set; and after all knowledge points are marked with the mastery level, taking the union of all elements in the current knowledge state set as the current knowledge state of the learner, and ending the test process.

Preferably, the complete test procedure is as follows:

step 1, determining a knowledge space according to a knowledge point diagram;

determining a knowledge state set according to knowledge points in the knowledge space and the incidence relation thereof;

determining a knowledge point testing sequence according to the child knowledge point, the child precursor knowledge point, the parent knowledge point and the parent precursor knowledge point;

starting the test, and updating the knowledge state set according to the test result until the test is terminated;

and 5, outputting the final knowledge state of the learner.

Compared with the prior art, the invention has the advantages that:

the method established in the invention can effectively infer the possible knowledge state of a learner in the knowledge space of the Optical Transport Network (OTN), greatly reduce the number of knowledge points to be tested by combining the inclusion relation and the leading relation among the knowledge points in the knowledge map, simplify the testing process and realize the rapid and efficient testing of the knowledge mastering level. The method provides a feasible solution for intelligent testing of an Optical Transport Network (OTN) knowledge system established in a knowledge map form, and accords with the future development direction of network evaluation and adaptive learning. The method can be applied to other professional fields of knowledge system construction based on the knowledge map in an expanded way, and a test scheme of learner competence level is designed, so that a foundation is laid for personalized learning and intelligent evaluation.

Drawings

FIG. 1 is a block diagram of a representative knowledge point map segment of an Optical Transport Network (OTN);

FIG. 2 is a flow diagram of an intelligent test based on an Optical Transport Network (OTN) knowledge point graph;

FIG. 3 is a diagram of knowledge mastery levels after the knowledge point 3 test;

FIG. 4 is a diagram of knowledge mastery levels after testing of knowledge points 7;

FIG. 5 is a diagram of knowledge mastery levels after the knowledge point 4 test;

fig. 6 is a knowledge grasp level test termination chart after the knowledge point 5 test.

Detailed Description

Referring to fig. 1-6, the present invention first constructs a knowledge point map in the field of Optical Transport Networks (OTNs), and designs a representation method of knowledge space by combining knowledge points and their association relations. Rules are further devised for extracting a set of knowledge states from the knowledge space, each knowledge state element in the set of knowledge states representing a possible knowledge state for the learner. And finally, designing a calculation method of the current knowledge state of the learner, deducing the knowledge point mastering level of the learner according to the knowledge point test result on the basis of automatically designing the knowledge point test sequence, and determining the current knowledge state of the learner from the knowledge state set. The specific implementation method is as follows:

firstly, constructing an Optical Transport Network (OTN) knowledge point map and knowledge space

And designing the overall architecture of an Optical Transport Network (OTN) knowledge point map based on knowledge map technology. The optical transmission network knowledge points form nodes in the knowledge point map, and the incidence relations among the optical transmission network knowledge points form edges in the knowledge point map. The types of association relations before a knowledge point in the knowledge point graph (i.e., "edges" between "nodes" and "nodes" in the graph) are 2:

the middle and the side contain the relationship: the knowledge points containing other knowledge points in the containing relation are called parent knowledge points, and the contained knowledge points are called child knowledge points;

in the leading relationship: the pilot relation has directionality, starts from the pilot knowledge point and ends at the subsequent knowledge point, the precondition for mastering the subsequent knowledge point is to master the pilot knowledge point,

according to the scheme, the optical transport network related technical specification, the national power grid company communication professional skill level evaluation standard and the post capability training specification are referred, the main knowledge points corresponding to the optical transport network related knowledge are summarized, the main knowledge points are subdivided under the guidance of experts until the main knowledge points are divided into atomic knowledge points (knowledge points which do not need to be continuously split from the knowledge learning or evaluation angle), a knowledge map construction method of 'from top to bottom' is adopted, the knowledge map structure of the optical transport network is designed, duplication removal and disambiguation of the knowledge points are carried out, and an Optical Transport Network (OTN) knowledge point map is constructed.

In order to facilitate the explanation of the intelligent testing method designed by the present invention, a representative map segment in an Optical Transport Network (OTN) knowledge point map is selected for the description of the testing method, as shown in fig. 1. The selected map segments have 7 knowledge points, and the knowledge points 1-7 respectively represent the OTN frame structure and overhead, the OTN frame structure, the OTN overhead, the OTUk frame structure, the ODUk frame structure, the OTN layered structure and the protection and maintenance management functions of the OTN. The selected map covers the condition that the child knowledge point (knowledge point 3) has a pilot knowledge point (knowledge point 7) and the parent knowledge point (knowledge point 2) has a pilot knowledge point (knowledge point 6), covers the atomic knowledge points 3, 4 and 5 (hereinafter referred to as leaf knowledge points) which cannot be split, has all basic characteristics of an Optical Transport Network (OTN) knowledge point map, and has typicality and representativeness when being used for explaining a test scheme.

Searching all descendant knowledge points contained in the knowledge point 1 and all leading knowledge points of the knowledge point 1 and the descendant knowledge points thereof, and constructing a knowledge space on the basis of all the knowledge points, wherein the knowledge points contained in the knowledge space are as follows: {1, 2, 3, 4, 5, 6, 7}.

And sequentially counting the parent knowledge point and the precursor knowledge point of each knowledge point in the knowledge space, and if the knowledge point has no parent knowledge point or precursor knowledge point, marking the knowledge point as 0. Only the case that only the only one leading knowledge point exists in the following knowledge points is considered here, and the statistical results are shown in table 1. The knowledge space is characterized by using { knowledge points, parent knowledge points and precursor knowledge points } as basic elements, the representation mode covers the knowledge points contained in the knowledge space and information of incidence relations among the knowledge points, and the knowledge space is represented by { {1, 0, 0}, {2, 1, 0}, {3, 1, 7}, {4, 2, 0}, {5, 2, 0}, {6, 0, 0}, and {7, 0, 0} }.

TABLE 1 knowledge points and their associated relation table contained in knowledge space

Knowledge points	Ancestor knowledge points (0 means nothing)	Pilot knowledge point (0 means nothing)
			1	0	0
2	1	6
			3	1	7
4	2	0
			5	2	0
6	0	0
			7	0	0

Secondly, establishing a knowledge state set

In the scheme, the mastery level of the learner at a certain knowledge point is set to be in two states of mastered state and mastered state. The method can determine which state a learner is in on a knowledge point by setting the knowledge point for the learner and checking the answer accuracy, and if the answer accuracy is higher than a set accuracy threshold, the state of the knowledge point is marked as mastered, otherwise, the knowledge point is marked as not mastered.

The knowledge state is a set of all knowledge points representing that the learner reaches mastery level in the knowledge space, and the knowledge state set is a set of all possible knowledge states, and the construction rule is as follows:

(1) the knowledge points contained in the knowledge state have the knowledge points of the children if the knowledge points of the parents exist, namely the knowledge points of the parents have the knowledge points of the children if the knowledge points of the parents are mastered;

(2) if there is a successor knowledge point, there must be its predecessor knowledge point, that is, the successor knowledge point must have its predecessor knowledge point.

For the knowledge points and their associations contained in the knowledge space described in table 1, there are:

if the knowledge point 1 is contained in the middle-edge knowledge state, the knowledge point 2 and the knowledge point 3 are necessarily contained;

if the knowledge point 2 is contained in the forward and backward knowledge state, the knowledge point 4 and the knowledge point 5 are necessarily contained;

if the knowledge point 2 is contained in the forward and backward knowledge state, the knowledge point 6 is necessarily contained;

if the knowledge point 3 is included in the forward/backward knowledge state, the knowledge point 7 is necessarily included.

Starting from the leaf knowledge points of the representative knowledge point map, searching all possible knowledge states in the knowledge space to form a knowledge state set which is: {{4}, {5}, {6}, {7}, {4, 5}, {4, 6}, {4, 7}, {5, 6}, {5, 7}, {6, 7}, {3, 7}, {4, 5, 6}, {4, 5, 7}, {5, 6, 7}, {3, 7, 4}, {3, 7, 5}, {3, 7, 6}, {4, 5, 6, 7}, {3, 7, 4, 5}, {3, 7, 4, 6}, {2, 4, 5, 6}, {2, 4, 5, 6, 7}, {3, 7, 4, 5, 6}, {2, 3, 4, 5, 6, 7}, {1, 2, 3, 4, 5, 6, 7}}. The learner's current knowledge state (the set of learned knowledge points) is an element of the set of knowledge states.

Thirdly, designing a testing sequence of knowledge points

The association relationship between knowledge points in the knowledge point map of the Optical Transport Network (OTN) also implies the cognitive logic of learners on each knowledge point, and can summarize the following inference:

(1) if the precursor knowledge point is not mastered, the subsequent knowledge point is not mastered to a certain extent (or a large probability);

(2) if the child knowledge point is not mastered, the parent knowledge point is not mastered to a certain extent (or a high probability);

(3) all child knowledge points contained in the parent knowledge points are mastered, so that the parent knowledge points are mastered to a certain extent (or with a high probability).

According to the logical inference described above, the inference rule that sets the knowledge point grasp level is as follows:

(1) if the test result of the subsequent knowledge point is 'mastered', the precursor knowledge point is directly marked as 'mastered' without retest;

(2) if the test result of the child generation knowledge point is 'not mastered', the parent generation knowledge point is directly marked as 'not mastered' without retest;

(3) if the test result of all the child knowledge points contained in the parent knowledge point is 'mastered', the parent knowledge point does not need to be tested again and is directly marked as 'mastered';

in order to reduce the number of the tested knowledge points, leaf knowledge points are tested firstly, pilot knowledge points corresponding to the leaf knowledge points are tested secondly, ancestor knowledge points of the leaf knowledge points are tested secondly, and then the knowledge point testing sequence of the pilot knowledge points corresponding to the ancestor knowledge points is tested secondly. With reference to the Optical Transport Network (OTN) knowledge point map in fig. 1, the knowledge point test sequence determined by the above method is:

1) knowledge points 3, 4, 5: the knowledge points 3, 4 and 5 are leaf knowledge points which cannot be continuously divided, and the leaf knowledge points are sequentially (or randomly) tested;

2) knowledge point 7: knowledge point 7 is a leading knowledge point of knowledge point 3 (leaf knowledge point);

3) knowledge point 2: since knowledge point 2 is a child knowledge point of knowledge point 1;

4) knowledge point 6: knowledge point 6 is a leading knowledge point of knowledge point 2;

5) knowledge point 1.

Fourthly, determining the current knowledge state of the learner

And further designing a method for updating the knowledge state set based on the test result so as to finally determine the current knowledge state of the learner. If the knowledge level of a certain knowledge point is marked as 'mastered', the knowledge state containing the knowledge point is reserved in the knowledge state set; if the knowledge level of a certain knowledge point is marked as 'not mastered', the knowledge state containing the knowledge point is removed from the knowledge state set. And after all knowledge points are marked with the mastery level, taking the union of all elements in the current knowledge state set as the current knowledge state of the learner, and ending the test process.

In the actual process, if the knowledge point to be tested has no test question, the method can be used for solving the problem of the knowledge point to be tested. In order not to affect the test, the test of the knowledge point is skipped, and the knowledge point is marked as "not mastered". With reference to the Optical Transport Network (OTN) knowledge point map in fig. 1, a complete testing process is shown in fig. 2, which is as follows:

step 1, determining a knowledge space according to a knowledge point diagram:

{{1, 0, 0}, {2, 1, 0}, {3, 1, 7}, {4, 2, 0}, {5, 2, 0}, {6, 0, 0}, {7, 0, 0}}

and 2, determining a knowledge state set according to knowledge points in the knowledge space and the incidence relation thereof:

{{4}, {5}, {6}, {7}, {4, 5}, {4, 6}, {4, 7}, {5, 6}, {5, 7}, {6, 7}, {3, 7}, {4, 5, 6}, {4, 5, 7}, {5, 6, 7}, {3, 7, 4}, {3, 7, 5}, {3, 7, 6}, {4, 5, 6, 7}, {3, 7, 4, 5}, {3, 7, 4, 6}, {2, 4, 5, 6}, {2, 4, 5, 6, 7}, {3, 7, 4, 5, 6}, {2, 3, 4, 5, 6, 7}, {1, 2, 3, 4, 5, 6, 7}}

and 3, determining the testing sequence of the knowledge points according to the 'child knowledge point-child precursor knowledge point-parent precursor knowledge point':

3-7-4-5-2-6-1

and 4, starting the test, updating the knowledge state set according to the test result until the test is terminated:

a) the knowledge point 3 is tested, and if the answer accuracy is lower than the threshold, the knowledge point is marked as "not mastered", and meanwhile, the parent knowledge point 1 of the knowledge point 3 is marked as "not mastered", as shown in fig. 3. And removing the knowledge state containing the knowledge point 3 and the knowledge point 1, and updating the knowledge state set into: {4}, {5}, {6}, {7}, {4, 5}, {4, 6}, {4, 7}, {5, 6}, {5, 7}, {6, 7}, {4, 5, 6}, {4, 5, 7}, {5, 6, 7}, {4, 5, 6, 7}, {2, 4, 5, 6}, {2, 4, 5, 6, 7} };

b) because the knowledge point 3 is not mastered, the leading knowledge point 7 needs to be tested, and if the answer accuracy of the knowledge point 7 is lower than the threshold value, the answer is marked as "not mastered", as shown in fig. 4, the knowledge state containing the knowledge point 7 is removed, and the knowledge state set is updated as follows: { {4}, {5}, {6}, {4, 5}, {4, 6}, {5, 6}, {4, 5, 6}, {2, 4, 5, 6} };

c) testing a knowledge point 4, marking as mastered if the answer accuracy is higher than a threshold value, and keeping a knowledge state set unchanged as shown in fig. 5;

d) testing a knowledge point 5, marking as mastered if the answer accuracy is higher than a threshold value, and keeping the knowledge state set unchanged;

e) since the knowledge points 4 and 5 are mastered, the test of the knowledge point 2 is skipped, the test is marked as mastered, and the knowledge state is unchanged;

f) the knowledge point 2 is mastered, the test of the pilot knowledge point 6 is skipped, the test is marked as mastered, and the knowledge state is unchanged;

g) the average of the mastery levels of all knowledge points is marked, as shown in fig. 6, and the test is ended.

And 5, outputting the final knowledge state of the learner.

The union of all elements in the final knowledge state set is {2, 4, 5, 6}, the complement of knowledge points contained in the knowledge space is {1, 3, 7}, which indicates that the learner has mastered the knowledge points 2, 4, 5, 6, does not have the mastered knowledge points 1, 3, 7, and the learner's current knowledge state is {2, 4, 5, 6 }.

The above steps realize the whole process of the construction of the knowledge point atlas and knowledge space, the construction of the knowledge state set and the intelligent testing of the knowledge point of the Optical Transport Network (OTN).

Claims (10)

1. A method for testing knowledge of an optical transport network based on knowledge graph technology is characterized by comprising the steps of constructing a knowledge point graph in the field of Optical Transport Networks (OTNs), and designing a representation method of a knowledge space by combining knowledge points and incidence relations thereof; further designing rules for extracting knowledge state sets from the knowledge space, wherein each knowledge state element in the knowledge state sets represents a possible knowledge state of the learner; and finally, designing a calculation method of the current knowledge state of the learner, deducing the knowledge point mastering level of the learner according to the knowledge point test result on the basis of automatically designing the knowledge point test sequence, and determining the current knowledge state of the learner from the knowledge state set.

2. The method for testing knowledge of an optical transport network based on a knowledge-graph technology according to claim 1, comprising the following steps:

constructing an OTN knowledge point map and a knowledge space of an optical transport network;

secondly, constructing a knowledge state set;

thirdly, designing a testing sequence of knowledge points;

and (IV) determining the current knowledge state of the learner.

3. The method for testing knowledge of an optical transport network based on the knowledge graph technology as claimed in claim 2, wherein the step (one) of constructing the OTN knowledge point graph and knowledge of the optical transport network specifically comprises: designing an overall architecture of an OTN knowledge point map based on a knowledge map technology; forming nodes in the knowledge point map by the knowledge points of the optical transport network, and forming edges in the knowledge point map by the incidence relation among the knowledge points of the optical transport network; the types of the association relation before the knowledge point in the knowledge point map are 2 types, namely, the inclusion relation and the leading relation.

4. The method according to claim 3, wherein the prior association relationship of the knowledge points in the knowledge point graph is an edge between a node and a node in the graph;

the inclusion relationship is: the knowledge points containing other knowledge points in the containing relation are called parent knowledge points, and the contained knowledge points are called child knowledge points; the pilot relationship is: the pilot relation has directionality, starts from the pilot knowledge point and ends at the subsequent knowledge point, and the precondition for mastering the subsequent knowledge point is to master the pilot knowledge point.

5. The method for testing knowledge of an optical transport network based on a knowledge graph technology according to claim 2, wherein the step (ii) of constructing the knowledge state set specifically comprises: the learner's mastery level at a certain knowledge point is set to be in two states of mastered and not mastered; and (3) setting a question for the learner aiming at the knowledge point, and checking the answer accuracy to determine which state the learner is in on the knowledge point, if the answer accuracy is higher than a set accuracy threshold, marking the state of the knowledge point as 'mastered', otherwise, marking the state of the knowledge point as 'not mastered'.

6. The method of claim 5, wherein the knowledge state is a set of all knowledge points that characterize the learner's level of mastery in knowledge space, and the knowledge state set is a set of all possible knowledge states, and is constructed as follows:

(1) the knowledge points contained in the knowledge state have the knowledge points of the children if the knowledge points of the parents exist, namely the knowledge points of the parents have the knowledge points of the children if the knowledge points of the parents are mastered;

(2) if there is a successor knowledge point, there must be its predecessor knowledge point, that is, the successor knowledge point must have its predecessor knowledge point.

7. The method as claimed in claim 2, wherein the step (iii) of designing the relationship between the knowledge points in the OTN knowledge point map in the knowledge point testing sequence also implies the learner's cognitive logic for each knowledge point, and makes the following inference:

(1) if the precursor knowledge point is not mastered, the subsequent knowledge point is not mastered;

(2) if the child knowledge point is not mastered, the parent knowledge point is not mastered;

(3) all child knowledge points contained in the parent knowledge points are mastered, and the parent knowledge points are mastered.

8. The method for testing knowledge of an optical transport network based on the knowledge-graph technology as claimed in claim 7, wherein the inference rule for setting the mastery level of the knowledge points is as follows:

(1) if the test result of the subsequent knowledge point is mastered, the pilot knowledge point is not required to be tested again and is directly marked as mastered;

(2) if the test result of the child generation knowledge point is not mastered, the parent generation knowledge point does not need to be tested again and is directly marked as not mastered;

(3) if the test results of all the child knowledge points contained in the parent knowledge points are mastered, the parent knowledge points do not need to be tested again and are directly marked as mastered;

in order to reduce the number of the tested knowledge points, leaf knowledge points are tested firstly, pilot knowledge points corresponding to the leaf knowledge points are tested secondly, ancestor knowledge points of the leaf knowledge points are tested secondly, and then the knowledge point testing sequence of the pilot knowledge points corresponding to the ancestor knowledge points is tested secondly.

9. The method for testing knowledge of optical transport networks based on knowledge-graph technology as claimed in claim 2, wherein the step (four) of determining the current knowledge state of the learner comprises the specific steps of: further designing a method for updating the knowledge state set based on the test result to finally determine the current knowledge state of the learner; if the knowledge level of a certain knowledge point is marked as mastered, the knowledge state containing the knowledge point is reserved in the knowledge state set; if the knowledge level of a certain knowledge point is marked as not mastered, removing the knowledge state containing the knowledge point from the knowledge state set; and after all knowledge points are marked with the mastery level, taking the union of all elements in the current knowledge state set as the current knowledge state of the learner, and ending the test process.

10. The method for testing knowledge of an optical transport network based on a knowledge-graph technology according to claim 9, wherein the complete testing process is as follows:

step 1, determining a knowledge space according to a knowledge point diagram;

determining a knowledge state set according to knowledge points in the knowledge space and the incidence relation thereof;

determining a knowledge point testing sequence according to the child knowledge point, the child precursor knowledge point, the parent knowledge point and the parent precursor knowledge point;

starting the test, and updating the knowledge state set according to the test result until the test is terminated;

and 5, outputting the final knowledge state of the learner.

Images