CN109684436B - Knowledge correlation method and application - Google Patents
Knowledge correlation method and application Download PDFInfo
- Publication number
- CN109684436B CN109684436B CN201811459521.8A CN201811459521A CN109684436B CN 109684436 B CN109684436 B CN 109684436B CN 201811459521 A CN201811459521 A CN 201811459521A CN 109684436 B CN109684436 B CN 109684436B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- different
- learning
- relevance
- items
- 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.)
- Active
Links
Images
Abstract
The invention discloses a knowledge association method and application, wherein the knowledge association method comprises the following steps: acquiring two different knowledge items in the same knowledge system; acquiring test data of a plurality of testers under two different knowledge items; according to the test data, the learning performance of each tester on two different knowledge items is obtained; and determining the relevance between the two different knowledge items according to the learning performance of each tester on the two different knowledge items. The invention can acquire the relevance between two different knowledge items through the learning performance of a plurality of testers on the two different knowledge items in the same knowledge system, realizes the relevance mining of the knowledge items with uncertain relevance under the same knowledge system, and is convenient for the learning management of the two different knowledge items.
Description
Technical Field
The invention relates to the field of learning management, in particular to a knowledge association method and application.
Background
With the gradual improvement of living standard of people, parents and teachers pay more and more attention to the learning condition of students, and meanwhile, a learning system capable of interacting with learners and service systems such as personalized content recommendation or adaptive learning are also produced. Because the different knowledge has mutual dependency relationship and incidence relationship, the incidence analysis is carried out on the different knowledge, which is beneficial to the learner to improve the learning ability and enhance the learning efficiency, and other knowledge can be recommended to the learner by utilizing the incidence analysis.
At present, the traditional knowledge association method is mainly to manually associate various different knowledge contents by using education experts according to experience. Obviously, the manual association learning is performed according to experience, so that the consumption of human resources is increased to a great extent, and time and labor are wasted. Or, the knowledge association can be carried out according to different learning contents selected by the user. The autonomous selection is usually performed by combining an apolior algorithm and an FP-Tree algorithm to associate knowledge based on frequent pattern mining, so that the association of the knowledge corresponding to the learning content frequently selected by the user is high. Therefore, this method is only suitable for a case where the user can autonomously select the learning content. However, for knowledge learning that the user cannot select autonomously, such as learning of teaching materials, since each part of knowledge of the content of the teaching materials needs to be learned, each part of knowledge appears frequently, and under the premise, the association learning method depending on autonomous selection cannot be used to learn the association between different knowledge.
Disclosure of Invention
Therefore, the technical problem to be solved by the embodiments of the present invention is that the association method of knowledge in the prior art depends on human association, which results in large human resource consumption and time and labor waste, or the traditional method of mining knowledge association based on frequent patterns cannot know the association between different knowledge with the same frequency.
Therefore, the embodiment of the invention provides the following technical scheme:
the embodiment of the invention provides a knowledge correlation method, which comprises the following steps:
acquiring two different knowledge items in the same knowledge system;
acquiring test data of a plurality of testers in the two different knowledge items;
according to the test data, acquiring the learning performance of each tester on the two different knowledge items;
and determining the relevance between the two different knowledge items according to the learning performance of each tester on the two different knowledge items.
Optionally, after the step of determining the association between the two different pieces of knowledge, the method further comprises:
and correcting the relevance between the two different knowledge items.
Optionally, the step of correcting the correlation between the two different pieces of knowledge further includes:
extracting text features and semantic features of each knowledge;
calculating text correlation between the two different knowledge items according to the text characteristics of each knowledge item; and calculating semantic correlation between the two different knowledge items according to the semantic features of each knowledge item;
calculating the content correlation between the two different knowledge items according to the text correlation and the semantic correlation;
and correcting the relevance between the two different knowledge items according to the content relevance.
Optionally, the step of obtaining, according to the test data, the learning performance of each tester on the two different knowledge items includes:
acquiring the testing capability value of each tester for each knowledge test question;
and determining the learning performance of each tester on the two different knowledge items according to the testing capability value.
Optionally, the test ability value is calculated by a project reaction theory algorithm.
Optionally, the method for associating knowledge further includes: the number of answers of each tester forms a data set, and the data set comprises a first preset number threshold and a second preset number threshold.
Optionally, in the knowledge association method, the multiple testers with the number of answers meeting the preset condition are screened out between the first preset number threshold and the second preset number threshold.
Optionally, the first preset quantity threshold and the second preset quantity threshold are obtained by an abnormal value truncation point of a box chart formed by data distribution of the answer quantity of the data group.
Optionally, the determining of the association between the two different knowledge items according to the learning performance of each tester on the two knowledge items is performed by an association algorithm.
Optionally, the two different pieces of knowledge are selected from textbooks or curriculum contents or teaching schemas in the same knowledge system.
Optionally, the test data is selected from a class assignment or post-class assignment or class test problem for the plurality of testers.
Optionally, the method for associating knowledge further includes:
acquiring the sequence of learning of the two different knowledge items;
and according to the sequential learning sequence, taking the prior knowledge as the prepositive knowledge of the later learning knowledge, or taking the later learning knowledge as the postpositive knowledge of the prior learning knowledge.
The embodiment of the invention provides a recommendation method for knowledge learning, which comprises the following steps:
acquiring current learning knowledge of a target learner in a knowledge system;
acquiring the relevance between a plurality of groups of two different knowledge in the knowledge system by using the knowledge relevance method;
searching post knowledge having the relevance with the current learning knowledge according to the relevance between the multiple groups of two different knowledge;
recommending the post-knowledge to the target learner.
Optionally, in the knowledge learning recommendation method, the association between the post-knowledge and the current learning knowledge is greater than a first preset threshold.
The embodiment of the invention provides an intelligent volume combination method, which comprises the following steps:
acquiring the knowledge to be checked of a target learner in a knowledge system;
acquiring the relevance between a plurality of groups of two different knowledge in the knowledge system by using the knowledge relevance method;
according to the relevance between the multiple groups of two different pieces of knowledge, searching first knowledge which has the relevance with the knowledge to be examined, wherein the first knowledge is the advanced knowledge of the knowledge to be examined;
respectively setting a first preset distribution proportion of the knowledge to be assessed and the first knowledge;
and intelligently grouping the volumes according to the first preset distribution proportion.
Optionally, the intelligent volume group method further includes:
according to the relevance between the multiple groups of two different pieces of knowledge, searching for second knowledge which has the relevance with the first knowledge, wherein the second knowledge is the advanced knowledge of the first knowledge;
respectively setting a second preset distribution proportion of the knowledge to be assessed, the first knowledge and the second knowledge;
and intelligently assembling the volumes according to the second preset distribution proportion.
An embodiment of the present invention provides a storage medium having stored thereon computer instructions that, when executed by a processor, perform the steps of the associated learning method for knowledge; or, implementing the recommendation method for knowledge learning; or, the steps of the intelligent volume combination method are realized.
The embodiment of the invention provides knowledge association equipment, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the steps of the knowledge association method when executing the program.
The embodiment of the invention provides recommendation equipment, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor executes the program to realize the steps of the recommendation method for knowledge learning.
The embodiment of the invention provides a volume assembling device, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the steps of the intelligent volume assembling method when executing the program.
The technical scheme of the embodiment of the invention has the following advantages:
the invention discloses a knowledge association method and application, wherein the knowledge association method comprises the following steps: acquiring two different knowledge items in the same knowledge system; acquiring test data of a plurality of testers under two different knowledge items; according to the test data, the learning performance of each tester on two different knowledge items is obtained; and determining the relevance between the two different knowledge items according to the learning performance of each tester on the two different knowledge items. The invention can acquire the relevance between two different knowledge items through the learning performance of a plurality of testers on the two different knowledge items in the same knowledge system, realizes the relevance mining of uncertain relevance of the two association relations in the same knowledge system, and is convenient for the learning management of the two different knowledge items.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a first flow chart of a method of associating knowledge in an embodiment of the present invention;
FIG. 2 is a second flow chart of a method of associating knowledge in an embodiment of the present invention;
FIG. 3 is a third flowchart of a method of associating knowledge in an embodiment of the invention;
FIG. 4 is a fourth flow chart of a method of associating knowledge in an embodiment of the invention;
FIG. 5 is a box diagram of a method of associating knowledge in an embodiment of the invention;
FIG. 6 is a first flowchart of a recommendation method for knowledge learning according to an embodiment of the present invention;
FIG. 7 is a first flowchart of an intelligent volume group method according to an embodiment of the present invention;
FIG. 8 is a second flowchart of the intelligent volume group method in an embodiment of the present invention;
FIG. 9 is a hardware diagram of a knowledge correlation device in an embodiment of the invention;
FIG. 10 is a hardware diagram of a recommendation device in an embodiment of the invention;
fig. 11 is a hardware schematic diagram of an intelligent volume group device in an embodiment of the present invention.
Detailed Description
The technical solutions of the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings, and it is to be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the embodiments of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience in describing the embodiments of the present invention and simplifying the description, but do not indicate or imply that the referred devices or elements must have specific orientations, be configured in specific orientations, and operate, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the embodiments of the present invention, it should be noted that, unless explicitly stated or limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; the two elements may be directly connected or indirectly connected through an intermediate medium, or may be communicated with each other inside the two elements, or may be wirelessly connected or wired connected. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Example 1
The embodiment of the invention provides a knowledge association method, as shown in fig. 1, comprising the following steps:
and S11, acquiring two different knowledge items in the same knowledge system. The same knowledge system here is of the same knowledge type, for example: the math textbook or the Chinese textbook or the geography textbook or the English textbook represent knowledge contained in the same knowledge type. The two different knowledge items are selected from textbooks or course contents or teaching outlines in the same knowledge system, but the association between the two is not determined. For example: a multiplication knowledge and a division knowledge are obtained in the math teaching material. Therefore, two different pieces of knowledge are objects of knowledge association.
And S12, acquiring test data of a plurality of testers at two different knowledge items. The test data herein is selected from the group consisting of classroom or post-classroom tasks or classroom test problems for a plurality of testers. The multiple testers can be a batch of test people who have learned the two different knowledge items, and the test data represents the learning performance of the multiple testers in classroom operation or post-classroom operation or classroom test exercises.
The test data also includes the ID of each tester, the ID of the test questions to be answered, the answering situation, the ID number of each knowledge test question and the attribute information thereof, and the attribute information includes the discrimination parameter, the project difficulty parameter and the guess parameter in the project reaction theory IRT.
The test data for the two different knowledge items can be represented by the following tables 1 and 2:
table 1 tester answer sheet
| Tester ID | Topic ID | Whether it is correct or not | |
| Answer record 1 | 1 | 10000001 | 0 |
| Answer recording 2 | 2 | 10000002 | 1 |
TABLE 2 topic information Table
Both tables 1 and 2 above present test data for each knowledge, and the test data for each knowledge is obtained to confirm whether there is an association between two uncertain knowledge.
And S13, acquiring the learning performance of each tester to two different knowledge items according to the test data. Learning here is expressed as how well each tester has mastered both of these knowledge items.
Specifically, as shown in fig. 2, the step S13 of obtaining the learning performance of two different knowledge items for each tester according to the test data includes:
s131, obtaining the testing capability value of each tester for each knowledge test question. The test ability value here is the mastery degree or the comprehension degree of each tester to each knowledge, the learning condition of each tester to each knowledge can be reflected through the test ability value, the test ability value can also be reflected through the test score, generally, the higher the score of each tester in the test question of each knowledge is, the higher the test ability value of the tester is, and generally, the lower the score of each tester in the test question of each knowledge is, the lower the test ability value of the tester is. The test data of the test capability value is mainly used for obtaining whether a certain incidence relation exists between two different knowledge items.
The test ability value is calculated by a project reaction theory algorithm. The specific formula is as follows:
wherein, P (theta) is the question answering probability, a is the discrimination parameter, b is the project difficulty parameter, c is the guess parameter, theta is the testing ability value, and D is the constant 1.7.
When the actual answer condition (whether correct) of each tester and the discrimination parameter a, the project difficulty parameter b and the guess parameter c of the corresponding questions are known numbers, the testing capability value of each tester on each knowledge can be calculated according to the maximum likelihood estimation algorithm.
Specifically, the guess parameter in the above formula (1) can be directly determined by the answer type, for example: the guess parameter of the question is 1/2, and the guess parameter of the radio question is 1/number of options. For a multiple selection of four options, all its answers may be:
the total number of the 14 types (A/B/C/D/AB/AC/AD/BC/BD/CD/ABC/ABD/BCD/ABCD) is 1/14, and the guessing degree parameter of the question-answer question is 0.
The item difficulty parameter, the discrimination parameter, and the test capability value in the above formula (1) are set to be a default value of 0 if one or more of the three items are vacant according to the item reaction theory IRT algorithm.
The formula (1) can also estimate the discrimination parameter and the project difficulty parameter through an iterative algorithm, and repeat iteration until the result is converged to obtain a final result. For example: and (3) 5000 testers answering the to-be-tested questions perform independent tests on the 5000 testers, and if the absolute values of the test capability values/project difficulty parameters/discrimination parameters of more than 95% of the testers, which change between the nth iteration and the (n + 1) th iteration, are less than a certain threshold, such as 0.01, the n +1 iteration results are considered to be converged. For example: if the testing capability value of the tester is unknown, acquiring response data of all the test questions answered by the tester, and estimating the testing capability value of the tester; if the project difficulty parameter is unknown, the answer data of each tester answering the test question and the discrimination parameter are obtained, and the project difficulty parameter is estimated according to the answer data and the discrimination parameter.
And S132, determining the learning performance of each tester on two different knowledge items according to the testing capability value. The higher the testing capability value is, the better the learning performance of the tester is, and the lower the testing capability value is, the worse the learning performance of the tester is, so that the learning performance of each tester to two different knowledge items can be represented through the testing capability value. For example: and (3) calculating the testing capability value of the first knowledge of each tester to be A and the testing capability value of the second knowledge to be B through the formula (1), and obtaining the learning expression of the plurality of testers on the two knowledge.
And S14, calculating the relevance between the two different knowledge items according to the learning performance of each tester on the two different knowledge items.
Specifically, the step S14 of determining the correlation between two different knowledge items is calculated by the correlation algorithm according to the learning performance of each tester on the two different knowledge items. The correlation algorithm comprises a Pearson correlation coefficient algorithm and a Spearman algorithm. Wherein, the formula of the Pearson correlation coefficient is as follows:
wherein r is a Pearson correlation coefficient between two different knowledge items, XiFor the ith tester's learning performance on the first knowledge, YiFor the learning performance of the ith tester on the second knowledge,
the learning performance of a plurality of testers on the first knowledge is mean,
the learning performance mean value of a plurality of testers on the second knowledge, and n is the number of the testers.
Wherein the Spearman algorithm is calculated by the following formula,
where ρ is the Spearman's correlation coefficient between two different pieces of knowledge, n is the number of multiple testers, diThe difference in learning performance ranking was presented for each tester at two different knowledge items. For example: when 100 testers are tested, n is 100, and the ith tester is at the ithThe learning expression ranking on one knowledge is 3, the learning expression ranking on the second knowledge is 5, di＝3-5＝-2。
The method for associating knowledge in the embodiment of the present invention, as shown in fig. 3, further includes, after the step S14 of determining the association between two different pieces of knowledge:
and S15, correcting the relevance between the two different knowledge items. The correction is to adjust the correlation between the two different knowledge items calculated in step S14 to make the estimation of the correlation more accurate.
Specifically, as shown in fig. 4, in the method for associating knowledge in the embodiment of the present invention, the step S15 of correcting the association between two different pieces of knowledge further includes:
and S151, extracting text features and semantic features of each knowledge. The text features are subjected to feature extraction through the Word description content of each knowledge, the semantic features are subjected to feature extraction through the semantic content of each knowledge, the semantic features can be obtained by extracting the Word description of each knowledge through the Word2vec algorithm, and the result is expressed in the form of Word vectors.
S152, calculating the text correlation between two different knowledge items according to the text characteristics of each knowledge item; and calculating semantic correlation between two different knowledge items according to the semantic features of each knowledge item.
Specifically, the calculation of the text relevance between two different knowledge items is performed by the following formula,
T(x,y)＝1-d(x,y)/maxLen(x,y) (4)
wherein T (x, y) is text relevance, x is a text feature of the first knowledge, y is a text feature of the second knowledge, d (x, y) is an edit distance between the first knowledge and the second knowledge, and maxLen (x, y) is a length value corresponding to a longer character in the first knowledge x and the second knowledge y. For example: the text feature of the first knowledge x is "calculation of area", the text feature of the second knowledge y is "calculation of volume", the edit distance d (x, y) between x and y is 1, and maxLen (x, y) is 5, so that T (x, y) is 4/5 calculated by the above formula (4).
Calculating the semantic correlation between two different pieces of knowledge is calculated by the following formula,
wherein cos (X, Y) is the semantic correlation of two different knowledge items, XjIs the value of the j dimension in the word vector corresponding to the first knowledge, XjThe j-th dimension value in the word vector corresponding to the second knowledge is shown, and m is the dimension of the word vector.
And S153, calculating the content correlation between the two different knowledge items according to the text correlation and the semantic correlation.
Specifically, the calculation formula of the content correlation is as follows:
content correlation ═ W1 × T (X, Y) + (1-W1) × cos (X, Y); (6)
wherein, for text relevance, cos (X, Y) is semantic relevance of two different knowledge items, W1 is weight corresponding to text relevance, (1-W1) is weight corresponding to semantic relevance, and W1 has weight range of 0-1.
And S154, correcting the relevance between the two different knowledge items according to the content relevance. The content relevance calculated by the text relevance and the semantic relevance adjusts the relevance between two different knowledge items.
Specifically, the final correction correlation formula is as follows:
the final correlation is max (g, (W2 Xg + (1-W2). times.h); (7)
Wherein g is the relevance between two different knowledge items, h is the content relevance, and W2 is the weight corresponding to the relevance between two different knowledge items, and the weight range is 0-1.
The association method of knowledge in the embodiment of the invention further comprises the following steps: the number of answers of each tester constitutes a data set, and the data set comprises a first preset number threshold and a second preset number threshold. For example: the number of answers for each tester is formed into a data set [55,70,75,76,77,78,150], and the data set includes an upper threshold and a lower threshold for the data set, where the lower threshold may be a first preset number threshold and the upper threshold may be a second preset number threshold.
Specifically, a plurality of testers with the number of answers meeting the preset conditions are screened between a first preset number threshold and a second preset number threshold. For example: the first threshold value of the number of the data sets is 58, and the second threshold value of the number of the data sets is 90, so that a plurality of testers with answer numbers meeting the preset conditions are screened out between 58 and 90, and therefore, abnormal values 55 and 150 in the data sets can be screened out. The purpose of obtaining the first preset number threshold and the second preset number threshold in this embodiment is to eliminate the answer data that do not meet the preset condition, so as to retain the required answer data.
Specifically, the first preset quantity threshold and the second preset quantity threshold are obtained by the abnormal value truncation point of the box chart formed by the data distribution of the data group. The box chart is also called box whisker chart, box chart or box chart, and is a statistical chart for displaying a group of data dispersion condition data. The boxplot typically includes an upper quartile point, a lower quartile point, an upper outlier truncation point, and a lower outlier truncation point. As shown in fig. 5, Q1 is the lower quartile, Q3 is the upper quartile, P1 is the lower outlier cut-off point, P2 is the upper outlier cut-off point, Q3 represents the upper quartile of the whole data, and Q1 represents the lower quartile of the whole data. Wherein, the IQR is Q3-Q1, the IQR is the quartering distance of a box diagram, the P1 is Q1-1.5IQR,
p2 ═ Q3+1.5 IQR. For example: the number of questions in the data set is [55,70,75,76,77,78,150], the first quartile point Q1 is the 2 nd number of questions in the set of numbers 70, and the second quartile point Q3 is the 6 th number of questions in the set of numbers 78. Therefore, IQR is Q3-Q1 is 78-70 is 8. The first outlier cutoff point P1-Q1-1.5 IQR-70-1.5 i 8-58; second outlier truncation point
P2-Q3 +1.5 IQR-78 +1.5 × 8-90. The box-shaped graph is generated according to the answer quantity of a plurality of testers, and the upper abnormal value truncation point and the lower abnormal value truncation point of the box-shaped graph are used as answer quantity thresholds to screen the plurality of testers, so that the problems of low estimation precision of the test capability value caused by too few answers of the plurality of testers or similar brushing, machine answering and the like reflected by too many answers of the plurality of testers are avoided. A first preset quantity threshold value is respectively set as an upper limit threshold value of the number of answers and a second preset quantity threshold value is set as a lower limit threshold value to screen out a plurality of testers, so that the calculation accuracy errors of the testing capability values of the plurality of testers are controlled.
The association method of knowledge in the embodiment of the invention further comprises the following steps:
firstly, the sequential learning sequence of two different knowledge items is obtained. The sequential learning order herein refers to the sequential order in which the target learner learns knowledge at each learning stage, for example: the students in the first grade need to learn the teaching material contents in the first grade before learning the teaching material contents in the second grade, for example: the contents of the multiplication are learned prior to the contents of the division. The learning sequence can determine which knowledge content of the two knowledge contents is the front knowledge and which knowledge content is the rear knowledge. Typically, the postknowledge is pre-knowledge dependent, such as: division is multiplication with postknowledge, i.e. division depends on multiplication knowledge.
Then, according to the sequential learning sequence, the first-learning knowledge is used as the front-end knowledge of the later-learning knowledge, or the later-learning knowledge is used as the rear-end knowledge of the first-learning knowledge. Once the sequential learning order is determined, the pre-knowledge and the post-knowledge can be determined between two different knowledge items.
The knowledge correlation method in the embodiment of the invention can acquire the correlation between two different pieces of knowledge aiming at the learning performance of a plurality of testers on the two different pieces of knowledge in the same knowledge system, realizes the correlation mining on the knowledge with uncertain correlation under the same knowledge system, and is convenient for the learning management on the two different pieces of knowledge.
Example 2
An embodiment of the present invention provides a recommendation method for knowledge learning, as shown in fig. 6, including:
and S61, acquiring the current learning knowledge of the target learner in a knowledge system. A knowledge system herein is a type of knowledge, such as: a math textbook or a chinese textbook or a geographic textbook or an english textbook, i.e. represents knowledge contained in one type of knowledge. The current learning knowledge is the knowledge that the target learner is learning. For example: the knowledge that the target learner is learning is the calculation of the cylinder surface area in the knowledge of the six-grade mathematical textbook.
S62, obtaining the relevance between a plurality of groups of two different knowledge in a knowledge system by using the relevance method of knowledge. The knowledge association method in embodiment 1 is used to obtain the association between different groups of knowledge in the same knowledge system. For example: A. b, C, D, E, F, G, H are provided. There is an association between knowledge a and knowledge B, an association between knowledge a and knowledge C, an association between knowledge a and knowledge D, an association between knowledge C and knowledge E, an association between knowledge E and knowledge F, and an association between knowledge G and knowledge H. Therefore, the relevance among a plurality of groups of different knowledge in the same knowledge system can be acquired.
And S63, searching the post knowledge which has relevance with the current learning knowledge according to the relevance between the multiple groups of two different knowledge. For example: the current learning knowledge is C, and the post knowledge which is related to C is searched in multiple groups of related data AB, AC, AD, CD, CE, EF and GH with relevance. For example: if the knowledge correlation method in embodiment 1 is used to know that the post-knowledge of the current learning knowledge of the knowledge C is E, the post-knowledge of the current learning knowledge can be obtained as E by searching the plurality of sets of data.
And S64, recommending post knowledge to the target learner. The target learner is a learner who learns under a certain knowledge system, for example: the individual who studies under a certain knowledge system or a certain class with consistent study progress. After the post-knowledge is acquired, in order to help the target learner to pre-learn the new knowledge, quickly understand new contents in the learning process, enhance the learning efficiency and learning enthusiasm of the target learner, and recommend the post-knowledge which is associated with the current learning knowledge to the target learner.
In the knowledge learning recommendation method in the embodiment of the invention, the relevance between the post-knowledge and the current learning knowledge is greater than a first preset threshold value. The first preset threshold value here represents one correlation value at which the degree of correlation between the post-knowledge and the current learning knowledge is the greatest.
By using the knowledge association method in the embodiment of the invention, when the target learner learns the knowledge of a certain knowledge system, the interdependency between two different knowledge items can be obtained according to the association between the two different knowledge items, so that the target learner can recommend the post-knowledge of the current learning knowledge. To a great extent, the learning efficiency of the target learner in a certain knowledge system can be improved, the learning interest of the target learner is enhanced, and the good learning habit of the target learner is developed.
Example 4
An embodiment of the present invention provides an intelligent volume group method, as shown in fig. 7, including:
and S71, acquiring the knowledge to be checked of the target learner in a knowledge system. A knowledge system herein is a type of knowledge, such as: a math textbook or a chinese textbook or a geographic textbook or an english textbook, i.e. represents knowledge contained in one type of knowledge. The knowledge to be examined is the knowledge that the target learner has learned and needs to be examined, for example: the core-to-be-examined knowledge that the target learner has learned is the calculation of the cylinder surface area in the knowledge of the six-grade mathematical textbooks.
And S72, acquiring the relevance between a plurality of groups of two different knowledge in the same knowledge system by using a knowledge relevance method. The knowledge association method in embodiment 1 is used to obtain the association between different groups of knowledge in the same knowledge system. For example: A. b, C, D, E, F, G, H are provided. There is an association between knowledge a and knowledge B, an association between knowledge a and knowledge C, an association between knowledge a and knowledge D, an association between knowledge C and knowledge E, an association between knowledge E and knowledge F, and an association between knowledge G and knowledge H. Therefore, the relevance among a plurality of groups of different knowledge in the same knowledge system can be acquired.
And S73, searching first knowledge which has relevance with the knowledge of the core to be examined according to the relevance between the two different knowledge groups, wherein the first knowledge is the advanced knowledge of the core to be examined. The foreknowledge here uses the association method of knowledge in example 1 to obtain the foreknowledge of the knowledge to be examined, i.e. the first knowledge.
And S74, setting a first preset distribution proportion of the to-be-checked knowledge and the first knowledge respectively. The first preset distribution proportion is the proportion of the knowledge to be examined and the proportion of the first knowledge. For example: the method comprises the following steps that the to-be-checked knowledge is C, the first knowledge is A, the first preset distribution proportion of A is preset to be 30%, and the first preset distribution proportion of C is preset to be 70%.
And S75, intelligently assembling the rolls according to the first preset distribution proportion. And intelligently assembling the examination paper according to the first preset distribution proportion of the to-be-examined knowledge of 70 percent and the first preset distribution proportion of the first knowledge of 30 percent, namely, the proportion of the to-be-examined knowledge in the examination paper is 70 percent and the proportion of the first knowledge in the examination paper is 30 percent. Intelligently grouping the volumes according to the proportion of 30 percent and 70 percent.
In the embodiment of the invention, the relevance between the knowledge to be examined and the first knowledge is greater than a second preset threshold value, and the second preset threshold value represents the maximum relevance value between the knowledge to be examined and the preposed knowledge thereof.
The intelligent test paper combining method in the embodiment of the invention can combine a knowledge association method to intelligently combine the knowledge to be checked of the target learner and the first knowledge according to different proportions, is beneficial to the knowledge test of the target learner, and is further convenient to analyze the mastery degree of the target learner on the current learning knowledge and locate weak links in the knowledge learning of the target learner.
Example 5
As shown in fig. 8, the intelligent volume assembling method in the embodiment of the present invention further includes:
and S81, acquiring the knowledge to be checked of the target learner in a knowledge system. The knowledge system here is a type of knowledge, such as: a math textbook or a chinese textbook or a geographic textbook or an english textbook, i.e. represents knowledge contained in one type of knowledge. The knowledge to be examined is the knowledge that the target learner has learned and needs to be examined. For example: the core-to-be-examined knowledge that the target learner has learned is the calculation of the cylinder surface area in the knowledge of the six-grade mathematical textbooks.
And S82, acquiring the relevance between a plurality of groups of two different knowledge in the same knowledge system by using a knowledge relevance method. The knowledge association method in embodiment 1 is used to obtain the association between different groups of knowledge in the same knowledge system. For example: A. b, C, D, E, F, G, H are provided. There is an association between knowledge a and knowledge B, an association between knowledge a and knowledge C, an association between knowledge a and knowledge D, an association between knowledge C and knowledge E, an association between knowledge E and knowledge F, and an association between knowledge G and knowledge H. Therefore, the relevance among a plurality of groups of different knowledge in the same knowledge system can be acquired.
And S83, searching second knowledge which has relevance with the first knowledge according to the relevance between the plurality of groups of two different knowledge, wherein the second knowledge is the advanced knowledge of the first knowledge. The knowledge correlation method in embodiment 1 can be used to obtain the pre-knowledge of the core knowledge to be examined, for example, the core knowledge to be examined is C, the first knowledge is B, B is the pre-knowledge of the core knowledge to be examined, the second knowledge is a, and a is the pre-knowledge of the first knowledge B.
And S84, setting a second preset distribution proportion of the to-be-checked knowledge, the first knowledge and the second knowledge respectively. The second preset distribution proportion is the proportion of the knowledge to be examined, the proportion of the first knowledge and the proportion of the second knowledge. For example: the core knowledge to be examined is C, the first knowledge is B, the second knowledge is A, the second preset distribution proportion of A is 20%, the second preset distribution proportion of B is 30%, and the second preset distribution proportion of the core knowledge to be examined is 50%.
And S85, intelligently assembling the rolls according to a second preset distribution proportion. And intelligently composing the examination paper according to a second preset distribution proportion of the to-be-examined knowledge of 50 percent, a second preset distribution proportion of the first knowledge of 30 percent and a second preset distribution proportion of the second knowledge of 20 percent, wherein the examination paper comprises the to-be-examined knowledge of 50 percent, the first knowledge of 30 percent and the second knowledge of 20 percent. Intelligently grouping the volumes according to the proportion of 50 percent, 30 percent and 20 percent.
In the intelligent volume combination method in the embodiment of the invention, the relevance of the knowledge to be checked and the first knowledge and the relevance of the first knowledge and the second knowledge are greater than a second preset threshold value. The second preset threshold value represents the maximum correlation value between the knowledge to be assessed and the preposed knowledge thereof.
The intelligent test paper combining method in the embodiment of the invention can combine a knowledge association method to intelligently combine the pre-knowledge of the core knowledge to be examined and the pre-knowledge of the core knowledge to be examined according to different proportions, thereby being beneficial to the knowledge test of the target learner and further positioning the problem shown in the knowledge learning of the target learner.
Through the methods in the embodiment 4 and the embodiment 5, the pre-knowledge of the knowledge to be examined can be obtained circularly, and then the pre-knowledge of the knowledge to be examined is intelligently organized according to different proportions, so that the knowledge to be examined of the target learner is formed, the knowledge test of the target learner is facilitated, and the mastery degree of the target learner on the current learning knowledge is conveniently analyzed, and weak links in knowledge learning of the target learner are positioned.
Example 6
Embodiments of the present invention provide a storage medium having stored thereon computer instructions that, when executed by a processor, implement the steps of the methods of embodiments 1, 2, 3, 4, and 5. The storage medium is also stored with test data of each tester for two different knowledge items, learning expression of each tester for two different knowledge items, text characteristics, semantic characteristics, text relevance, semantic relevance and the like of each knowledge item.
The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a flash Memory (FlashMemory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a Random Access Memory (RAM), or the like.
Example 7
An embodiment of the present invention provides a knowledge association apparatus, as shown in fig. 9, including a memory 920, a processor 910, and a computer program stored on the memory 920 and executable on the processor 910, where the processor 910 implements the steps of the method in embodiment 1 when executing the program.
Fig. 9 is a schematic diagram of a hardware structure of an association device for performing a processing method for list item operations according to an embodiment of the present invention, as shown in fig. 9, the knowledge association device includes one or more processors 910 and a memory 920, where one processor 910 is taken as an example in fig. 9.
The apparatus for performing the processing method of the list item operation may further include: an input device 930 and an output device 940.
The processor 910, the memory 920, the input device 930, and the output device 940 may be connected by a bus or other means, and fig. 9 illustrates an example of a connection by a bus.
Example 8
An embodiment of the present invention provides a recommendation device, as shown in fig. 10, including a memory 1020, a processor 1010 and a computer program stored in the memory 1020 and executable on the processor 1010, where the processor 1010 executes the computer program to implement the steps of the method in embodiment 1.
Fig. 10 is a hardware schematic diagram of a recommendation device for executing a processing method of list item operations according to an embodiment of the present invention, as shown in fig. 10, the recommendation device includes one or more processors 1010 and a memory 1020, where one processor 1010 is taken as an example in fig. 10.
The apparatus for performing the processing method of the list item operation may further include: an input device 1030 and an output device 1040.
The processor 1010, the memory 1020, the input device 1030, and the output device 1040 may be connected by a bus or other means, and fig. 10 illustrates an example of connection by a bus.
Example 9
An embodiment of the present invention provides an intelligent volume group device, as shown in fig. 11, which includes a memory 120, a processor 110, and a computer program stored on the memory 120 and operable on the processor 110, where the processor 110 implements the steps of the methods in embodiments 4 and 5 when executing the program.
Fig. 1 is a hardware schematic diagram of an intelligent volume device for performing a processing method of list item operations according to an embodiment of the present invention, as shown in fig. 1, the intelligent volume device includes one or more processors 110 and a memory 120, where one processor 110 is taken as an example in fig. 1.
The apparatus for performing the processing method of the list item operation may further include: an input device 130 and an output device 140.
The processor 110, the memory 120, the input device 130, and the output device 140 may be connected by a bus or other means, and fig. 1 illustrates the connection by a bus as an example.
The processor 110 may be a Central Processing Unit (CPU). The Processor 110 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or any combination thereof. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.
Claims (19)
1. A method for associating knowledge, comprising the steps of:
acquiring two different knowledge items in the same knowledge system, wherein the same knowledge system represents the knowledge contained in the same knowledge type;
acquiring test data of a plurality of testers in the two different knowledge items;
according to the test data, acquiring the learning performance of each tester on the two different knowledge items;
determining the relevance between the two different knowledge items according to the learning performance of each tester on the two different knowledge items;
correcting the correlation between the two different pieces of knowledge;
the step of correcting the correlation between the two different pieces of knowledge further comprises:
extracting text features and semantic features of each knowledge;
calculating text correlation between the two different knowledge items according to the text characteristics of each knowledge item; and calculating semantic correlation between the two different knowledge items according to the semantic features of each knowledge item;
calculating the content correlation between the two different knowledge items according to the text correlation and the semantic correlation;
correcting the correlation between the two different pieces of knowledge according to the content correlation;
and finally determining the relevance between the two different knowledge items by adopting the following formula:
the final correlation is max (g, (W2 × g + (1-W2) × h);
wherein g is the relevance between two different knowledge items, h is the content relevance, and W2 is the weight corresponding to the relevance between two different knowledge items, and the weight ranges from 0 to 1.
2. The method for associating knowledge according to claim 1, wherein the step of obtaining learning performance of each tester on the two different knowledge items according to the test data comprises:
acquiring the testing capability value of each tester for each knowledge test question;
and determining the learning performance of each tester on the two different knowledge items according to the testing capability value.
3. The method of claim 2, wherein the test ability value is calculated by a project reaction theory algorithm.
4. The method of associating knowledge of claim 2, further comprising: the number of answers of each tester forms a data set, and the data set comprises a first preset number threshold and a second preset number threshold.
5. The method according to claim 4, wherein the plurality of testers with the number of answers meeting the preset condition are screened between the first preset number threshold and the second preset number threshold.
6. The method of claim 5, wherein the first predetermined number threshold and the second predetermined number threshold are obtained by an outlier cutoff point of a box chart formed by data distribution of answer numbers of the data sets.
7. The method for associating knowledge according to claim 1, wherein the determining of the association between the two different knowledge items according to the learning performance of each tester on the two knowledge items is calculated by an association algorithm.
8. The method of claim 1, wherein the two different pieces of knowledge are selected from textbooks or curriculum contents or teaching schemas in the same knowledge system.
9. The method of claim 1, wherein the test data is selected from the group consisting of a classroom or post-classroom task or a classroom test problem for the plurality of testers.
10. The method of associating knowledge of claim 1, further comprising:
acquiring the sequence of learning of the two different knowledge items;
and according to the sequential learning sequence, taking the prior knowledge as the prepositive knowledge of the later learning knowledge, or taking the later learning knowledge as the postpositive knowledge of the prior learning knowledge.
11. A recommendation method for knowledge learning, comprising:
acquiring current learning knowledge of a target learner in a knowledge system;
acquiring the relevance between a plurality of groups of two different knowledge in the knowledge system by using the knowledge relevance method of any one of claims 1-10;
searching post knowledge having the relevance with the current learning knowledge according to the relevance between the multiple groups of two different knowledge;
recommending the post-knowledge to the target learner.
12. The knowledge learning recommendation method according to claim 11, wherein the association between the post-knowledge and the current learning knowledge is greater than a first preset threshold.
13. An intelligent volume assembling method is characterized by comprising the following steps:
acquiring the knowledge to be checked of a target learner in a knowledge system;
acquiring the relevance between a plurality of groups of two different knowledge in the knowledge system by using the knowledge relevance method of any one of claims 1-10;
according to the relevance between the multiple groups of two different pieces of knowledge, searching first knowledge which has the relevance with the knowledge to be examined, wherein the first knowledge is the advanced knowledge of the knowledge to be examined;
respectively setting a first preset distribution proportion of the knowledge to be assessed and the first knowledge;
and intelligently grouping the volumes according to the first preset distribution proportion.
14. The intelligent volume group method of claim 13, further comprising:
according to the relevance between the multiple groups of two different pieces of knowledge, searching for second knowledge which has the relevance with the first knowledge, wherein the second knowledge is the advanced knowledge of the first knowledge;
respectively setting a second preset distribution proportion of the knowledge to be assessed, the first knowledge and the second knowledge;
and intelligently assembling the volumes according to the second preset distribution proportion.
15. The intelligent volume organizing method of claim 14, wherein the association between the knowledge to be qualified and the first knowledge and the association between the first knowledge and the second knowledge are greater than a second preset threshold.
16. A storage medium having stored thereon computer instructions, characterized in that the instructions, when executed by a processor, carry out the steps of the associated learning method of knowledge according to any one of claims 1-10; or, the step of implementing the recommendation method for knowledge learning of any one of claims 11-12; or, the steps of implementing the intelligent volume group method of any of claims 13-15.
17. Knowledge correlation apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of correlation of knowledge according to any of claims 1-10 when executing the program.
18. Recommendation device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the recommendation method of knowledge learning according to any of claims 11-12 when executing the program.
19. A volume device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the intelligent volume method according to any one of claims 13-15 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459521.8A CN109684436B (en) | 2018-11-30 | 2018-11-30 | Knowledge correlation method and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811459521.8A CN109684436B (en) | 2018-11-30 | 2018-11-30 | Knowledge correlation method and application |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109684436A CN109684436A (en) | 2019-04-26 |
| CN109684436B true CN109684436B (en) | 2021-08-03 |
Family
ID=66185616
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811459521.8A Active CN109684436B (en) | 2018-11-30 | 2018-11-30 | Knowledge correlation method and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109684436B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110021213A (en) * | 2019-05-14 | 2019-07-16 | 上海乂学教育科技有限公司 | Mathematics preamble learning method in artificial intelligence study |
| CN110618390B (en) * | 2019-09-25 | 2021-08-31 | 深圳力维智联技术有限公司 | Method, device and equipment for identifying abnormal single battery of storage battery pack and storage medium thereof |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512214A (en) * | 2015-11-28 | 2016-04-20 | 华中师范大学 | Knowledge database, construction method and learning situation diagnosis system |
| CN106599999A (en) * | 2016-12-09 | 2017-04-26 | 北京爱论答科技有限公司 | Evaluation method and system for using small amount of questions to accurately detect segmented weak knowledge points of student |
| CN107943940A (en) * | 2017-11-23 | 2018-04-20 | 网易（杭州）网络有限公司 | Data processing method, medium, system and electronic equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180061999A (en) * | 2016-11-30 | 2018-06-08 | 한국전자통신연구원 | Apparatus and method for adaptive e-learning |
-
2018
- 2018-11-30 CN CN201811459521.8A patent/CN109684436B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512214A (en) * | 2015-11-28 | 2016-04-20 | 华中师范大学 | Knowledge database, construction method and learning situation diagnosis system |
| CN106599999A (en) * | 2016-12-09 | 2017-04-26 | 北京爱论答科技有限公司 | Evaluation method and system for using small amount of questions to accurately detect segmented weak knowledge points of student |
| CN107943940A (en) * | 2017-11-23 | 2018-04-20 | 网易（杭州）网络有限公司 | Data processing method, medium, system and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109684436A (en) | 2019-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110399541B (en) | Topic recommendation method and device based on deep learning and storage medium | |
| Seaton et al. | The reciprocal relations between self-concept, motivation and achievement: Juxtaposing academic self-concept and achievement goal orientations for mathematics success | |
| Vie et al. | A review of recent advances in adaptive assessment | |
| CN110990691A (en) | Online course recommendation method and device and computer storage medium | |
| CN107481170A (en) | A kind of course recommends method, apparatus, curricula-variable server and storage medium | |
| CN105355111A (en) | After-class reinforced learning system based on learning situation analysis | |
| CN111651677B (en) | Course content recommendation method, apparatus, computer device and storage medium | |
| CN109816265B (en) | Knowledge characteristic mastery degree evaluation method, question recommendation method and electronic equipment | |
| Hu et al. | Evaluation of model fit in cognitive diagnosis models | |
| Herman et al. | A psychometric evaluation of the digital logic concept inventory | |
| CN110399558B (en) | Test question recommendation method and system | |
| CN111858906B (en) | Problem recommendation method and device, electronic equipment and computer readable storage medium | |
| CN111090809A (en) | Topic recommendation method and device, computer equipment and storage medium | |
| KR20190006409A (en) | Learning and Scheduling Apparatus and Method of Word Recognition State Quantification and Smart Devices using Memorizing Learning Data | |
| CN109684436B (en) | Knowledge correlation method and application | |
| CN108665394A (en) | Adaptive learning method and system | |
| Lynch et al. | Real world usage of an adaptive testing algorithm to uncover latent knowledge | |
| CN116796802A (en) | Learning recommendation method, device, equipment and storage medium based on error question analysis | |
| Pérez et al. | Automatic classification of question difficulty level: Teachers' estimation vs. students' perception | |
| CN115081912A (en) | Hierarchical reading method and system for helping children enhance reading ability and increase reading interest | |
| CN113239669B (en) | Test Question Difficulty Prediction Method | |
| CN109840867B (en) | Intelligent teaching method, equipment and device | |
| Smith et al. | Adjusting for guessing and applying a statistical test to the disaggregation of value-added learning scores | |
| CN111008916A (en) | Knowledge point grasp inference method, knowledge point grasp inference system, computer device, and storage medium | |
| Bolsinova et al. | Using expert knowledge for test linking. |
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 |