CN115186014B - Data processing method for educational training - Google Patents
Data processing method for educational training Download PDFInfo
- Publication number
- CN115186014B CN115186014B CN202211107383.3A CN202211107383A CN115186014B CN 115186014 B CN115186014 B CN 115186014B CN 202211107383 A CN202211107383 A CN 202211107383A CN 115186014 B CN115186014 B CN 115186014B
- Authority
- CN
- China
- Prior art keywords
- child
- children
- degree
- classroom
- time
- 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
- 238000012549 training Methods 0.000 title claims abstract description 19
- 238000003672 processing method Methods 0.000 title claims abstract description 13
- 230000003993 interaction Effects 0.000 claims abstract description 26
- 238000004088 simulation Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 16
- 239000000126 substance Substances 0.000 claims description 11
- 230000003542 behavioural effect Effects 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 4
- 230000007547 defect Effects 0.000 abstract description 3
- 238000012552 review Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 26
- 230000000694 effects Effects 0.000 description 22
- 238000004364 calculation method Methods 0.000 description 8
- 230000002452 interceptive effect Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 206010046542 Urinary hesitation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/011—Arrangements for interaction with the human body, e.g. for user immersion in virtual reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2203/00—Indexing scheme relating to G06F3/00 - G06F3/048
- G06F2203/01—Indexing scheme relating to G06F3/01
- G06F2203/012—Walk-in-place systems for allowing a user to walk in a virtual environment while constraining him to a given position in the physical environment
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Human Computer Interaction (AREA)
- Remote Sensing (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a data processing method for educational training, wherein related electronic equipment is adopted to identify each child in a target classroom, the behavior membership degree corresponding to each child is determined based on the prestige frequency of each child in the classroom, the typical hesitation time of interaction and the hesitation stability degree, the child with the largest behavior membership degree is taken as a representative child in a target time period and is taken as an object for recommending simulation when a young teacher reviews the target time, the quality of a copy can be improved compared with a mode of randomly selecting a simulation object, the young teacher is helped to find the defects when interacting with the child, the training of professional ability is further deepened, the workload of the copy of the young teacher can be reduced compared with a mode of simulating all children one by one, and the copy efficiency is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a data processing method for education training.
Background
Preschool education is foundation education of life development, people attach more and more importance to preschool education, and practical infant teacher training market prospect is wide.
At present, professional training of preschool education is difficult to realize by scene simulation. As most preschool education forms are indoor interactive activities, for a large number of education contents depending on children sitting on the ground, children are easy to be distracted, and the interactive ability of teachers is tested. Present VR (virtual reality) technique can provide an immersive experience, can guarantee that the teacher looks back, understand the administrative effect of education in-process, discover the problem that its classroom in-process exists, and then improve follow-up classroom quality, but there is effectual data drive mode temporarily, data processing mode promptly, lead to the teacher not to know when looking back that to put into which children's role and experience, be unfavorable for the teacher to discover the problem that its classroom in-process exists, thereby the effect that the teacher utilized the VR technique to look back in class has been reduced.
Disclosure of Invention
In order to solve the problem that the existing data processing mode can not enable a teacher to not know which child role to be placed for experience when the teacher looks back in a classroom by utilizing a VR technology, and therefore the effect of looking back in the classroom is reduced, the invention aims to provide a data processing method for educational training.
The invention discloses a data processing method for educational training, which comprises the following steps:
acquiring the frequent degree of the prestige of each child in a classroom, the typical hesitation time for interaction and the stability degree of the presthesion in the target classroom;
calculating the attraction degree of each child in the classroom to each child according to the typical delay time and the delay stability degree of each child to interaction in the target classroom;
calculating the behavior difference distance between any two children according to the attraction degree of the classroom in the target classroom to each child, the frequent degree of prestige and the lagged time sequence; classifying the children in the target classroom according to the behavior difference distance to obtain a plurality of types; the delay time sequence is a sequence formed by the interaction delay times of the corresponding children in the classroom;
for either type: calculating the spatial distance between each child in the type and other children of the type according to the prestige frequency of each child in the type and the prestige state sequence corresponding to the target time period, calculating the behavior membership degree corresponding to each child in the type according to the spatial distance, taking the child with the maximum behavior membership degree as a representative child of the type corresponding to the target time period, and taking the representative child of the type corresponding to the target time period as a recommended simulation object when a young teacher looks back at the target time; the periscopic state sequence corresponding to the target time period is a sequence formed by the periscopic states corresponding to the collecting moments of the corresponding children in the target time period.
Further, the method for obtaining the stability degree of the late doubt comprises the following steps:
the degree of hesitation stability is calculated using the following formula:
wherein the content of the first and second substances,the degree of hesitation stability for a child,the typical late time to doubt for that child,in order to find the maximum value,is the hesitation time series for that child.
Further, the late suspected time is typically the average of the first 30% of the maximum late suspected times in the late suspected time sequence.
Further, the attraction degree of the classroom to each child is calculated by using the following formula:
wherein, the first and the second end of the pipe are connected with each other,in order to attract a child in a classroom,represents a typical suspicion time of the child, K represents a suspicion stability degree of the child, tanh () is a hyperbolic tangent function, alpha is a neighborhood correction parameter,,is the maximum of the typical hesitation times for children in the eight neighbourhood of the child.
Further, the behavior difference distance between any two children is calculated by using the following formula:
wherein the content of the first and second substances,being the behavioural difference distance between child a and child B,to the extent of attraction of the class to the child a,to the extent of attraction of the class to the child B,the frequent degree of the look for the child a,the frequent degree of the look of the child B,is the late-suspected time series of child a,is the late-suspected time series of child B,the cosine similarity is solved.
Further, the calculating a spatial distance between each child in the category and another child in the category according to the prestige frequency of each child in the category and the prestige status sequence corresponding to the target time period includes:
wherein the content of the first and second substances,is the spatial distance between the child p and the child q,for the sequence of states of the child p in the target time period,for the sequence of the children's q's state of view within the target time period,the frequent degree of the look of the child p,the frequent degree of prestige of the children q.
Further, calculating behavior membership corresponding to each child in the type according to the spatial distance, including:
the behavioral membership is calculated using the following formula:
wherein the content of the first and second substances,the behavior membership of the child sample p,is the spatial distance between the child sample p and the child sample q,kth reachable distance for child sample pThe set of child samples covered in, K being the set value.
Furthermore, the DBSCAN algorithm is used for classifying the children in the target classroom.
Has the beneficial effects that: the method obtains parameters capable of reflecting behaviors, attention and performance of children in an activity classroom, wherein the parameters comprise frequent degree of prestige of each child in the classroom, typical delay time for interaction and delay stability degree; classifying children in the classroom according to the parameters to obtain different types of children; next, analyzing the children included in each type, and selecting representative children in each type as recommended simulation objects in the later period of driving VR, compared with a mode of randomly selecting simulation objects, the method can improve the quality of the reply, help the young students find the defects in the interaction with the children, further deepen the training of professional ability, and improve the VR review effect; compared with a mode of simulating all children one by one, the tray-duplicating work load of the young teachers can be reduced, and the tray-duplicating efficiency is improved.
Drawings
Fig. 1 is a flow chart of a data processing method for educational training of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.
As shown in fig. 1, the data processing method for educational training of the present embodiment includes:
step 1: and acquiring the frequent degree of the hope of each child in the classroom, the typical delay time for interaction and the delay stability degree in the target classroom.
This embodiment represents a study classroom as a target classroom with multiple children in the target area. In the embodiment, two cameras are arranged in a target classroom, wherein the first camera is a common camera for shooting a child and is used for acquiring and calculating the head direction of the child; the second is a VR camera, which is set up behind the sitting area of the children seat for the children to simulate the situation of the children watching the attention reduction.
In order to determine the listening and speaking habits of each child in the target classroom from the active classroom, the implementationFor example, the listening and speaking modes of all children are counted, and the frequent degree of looking and looking of all children is calculatedTypical hesitation time of a child to an interactionAnd degree of tardy stability. The specific acquisition process of the parameters is as follows:
the head posture of each child can be estimated by collecting the facial key points and the three-dimensional Mesh through the first camera, and a more common way is to determine the ear key points by using OpenPose and roughly calculate the head posture by combining the recognition result of the facial key points. The deflection angle A of the head relative to the position of the young teacher can be obtained through conversion of the pose of the camera, and the specific conversion process is not repeated because the existing estimation technology of the head deflection angle is widely applied. For the head deflection angle of the child, the tolerance angle interval can be set based on the polar coordinate space of the head center of the childTo distinguish the state of the child watching the teachers at the time t from the state of looking aroundRecording the state of each child's view of the whole classSampling is performed every 5 seconds until time T is reached and the activity classroom is over. When the child is in the state of watching the young teacher,=1; when the child does not belong to the state of watching the young teacher, i.e., the state of showing a look-up,=0, the child belongs to the state of watching the young teacher, namely the child has the head deflection angle in the tolerance angle intervalThe child is not in the condition of watching the young teacher, i.e. the head deflection angle of the child is not in the tolerance angle interval. Tolerance angle intervalCan be set by the user when in application.
Counting the frequent degree of observation: when the state of looking at the children does not belong to the state for 2 times continuously from a certain moment, the state of looking at the children is considered to be finished once, and the children continue to pay attention to the content of the activity classroom of the teachers. Calculating the time length of the child in the state of looking at, and then looking at the frequency degreeThe time that the deflection angle of the head posture exceeds the tolerance angle (namely the time in the state of looking at) accounts for the whole class time, and the greater the frequent degree of looking at, the less interested the children in the activity class information of the current teacher.
Calculating the typical hesitation time of a child to an interaction: the interaction time interval is specified by the teachers in each class, so that the late-doubt time of the children in the activity class is calculated. Every class all generally comprises a plurality of times of interdynamic, and when the child was in the state of looking after before the interaction began, the beginning of each time of interdynamic can attract children's attention to turn to the young teacher, when there is a plurality of interactive hesitancies in this in-processIn between, these late times may constitute a sequence of late timesWherein, in the process,the hesitation time for a child at the first interaction,the hesitation time of a certain child at the nth interaction, N is the total number of interactions involved in the class.
Typical hesitation times can be calculated for multiple attention shifts of children in an activity classroomAnd late stability degree K. For children, multiple times of attention inattention are common phenomena, how to attract children is a part of training of preschool education specialties, and the method is also an important link for guaranteeing education quality. For the phenomenon of inattention, existence of multiple interactive hesitation, and obvious hesitation for multiple times enough to influence the quality of the current activity class of the child, the typical hesitation time is calculated by the embodiment,Time series for question delayThe average of the first 30% of the maximum late times included in the late times. In this embodiment, the typical interaction time is calculated according to the last 30% of the maximum late doubt time, and as another embodiment, the typical interaction time may be calculated by selecting the last 25% or 40% of the maximum late doubt time.
The lagged-suspected stability degree K can reflect children in the current activity classroomThe degree of stability of the late suspicion time of the child, in this exampleIf the ratio of the typical latency to the maximum interactive latency is smaller, then this is saidAndthe difference is large; if the ratio of the typical hesitation time to the maximum interactive hesitation time is large, it is statedAndthe difference is small, and the difference is small,and can represent the attraction degree of the classroom to the children. Implementer settingsThe threshold value of (a) is set,below the threshold, the hearing and speaking quality of the child is considered to be not greatly affected, and K =1; for children without hesitation time, K =1 was set.
Therefore, the frequent degree of the hope of each child in the target classroom, the typical delay time for interaction and the delay stability degree can be obtained.
Step 2: and calculating the attraction degree of the classroom to each child according to the typical delay time and the delay stability degree of each child to the interaction in the target classroom.
The shorter the time to doubt, i.e. the less distracted the child is, the more attracted the interaction and listening to the lessee in the activity class, the greater the child's attraction to the quality of the activity class. In this embodiment, the following formula is specifically used to calculate the attraction degree of the classroom to each child:
wherein the content of the first and second substances,to the attraction degree of a classroom to a certain child,the larger the attraction, the larger the attraction degree;indicating the typical suspicion time of the child and K indicating the suspicion stability of the child.The smaller, theCompared withLarger, if K is not considered, but onlySince the calculated attraction degree is large as the size of the attraction degree of the child in the class, the attraction degree of the child listening and speaking is corrected by using K as the correction index in this embodiment. tanh () is a hyperbolic tangent function, used here for normalization of parameters.
The method comprises the following steps of calculating the maximum typical hesitation length of eight neighborhoods, wherein alpha is a neighborhood correction parameter, the eight neighborhoods of the positions where children are located need to be referenced in a setting mode, and the eight neighborhoods can be mutually influenced, so that the maximum typical hesitation length of the eight neighborhoods is counted and used as a reference value. When a child is in a certain position, the typical hesitation time of the child around the child is greater than that of the child, and the child is considered to be relative to the weekIn the case where the surrounding child is attracted by other things, the activity classroom contents of the young teacher can be more concerned. In this exampleWhereinIndicating a typical late-to-doubt time for a child,is the maximum of the typical hesitation times for children in the eight neighbourhood of the child. When the child is at an edge location, no non-existent neighborhoods are included in the calculation. The degree of attraction of the children to the contents of the teacher's activity classroom in the activity classroom is obtained through analysis so far。
And step 3: calculating the behavior difference distance between any two children according to the attraction degree, the prestige frequency degree and the delay time sequence of each child in the classroom of the target classroom; classifying the children in the target classroom according to the behavior difference distance to obtain a plurality of types; the delay time sequence is a sequence formed by the interaction delay times of the corresponding children in the classroom.
The embodiment analyzes the in-class data of the children in the target classroom and determines the difference of the interaction behaviors of the children of different types in the activity classroom. In the embodiment, based on a dbscan algorithm, r and minpts are set for clustering to obtain more clustering clusters, wherein each clustering cluster is a type; wherein the child behavior difference distance function is constructed as follows:
wherein, the first and the second end of the pipe are connected with each other,between child A and child BThe distance of the difference in the behavior is,in order to the attraction degree of the class to the child a,to the extent of attraction of the class to the child B,the frequent degree of the look for the child a,the frequent degree of the look of the child B,is the late-suspected time series of child a,is the late-suspected time series of child B,the cosine similarity is solved. In the embodiment, the cosine similarity distance is used for secondary comparison, and the attraction degree of the children during listening and speaking is determined according to the similarityFrequent degree of observationAnd constructing a behavior vector. The present embodiment utilizesThe calculation is carried out for the purpose of avoiding coincidence of similar attraction degrees caused by different behavior modes when the attraction degrees are determined, so that the final calculation results of the values of different parameters are the same.
Thus, the difference in the degree of attraction to the activity class between any two children can be obtained. According to the mode, the distance comparison is carried out on any two sampled children at present, and the difference distance between any two children is obtained. Thus, a child group assumed space based on psychological behaviors is obtained, and based on the assumed space, children in different listening and speaking modes are divided into different child groups based on the improved DBSCAN, and each child group corresponds to one type. For example, some children can well draw their attention when the interaction between the teachers starts, so the late time is short and the attraction degree value is large, but the attention is often drawn when the children are not in the interaction ring, and the children in the listening and speaking mode are classified into one type. In addition, there is also a large attraction degree value, and children who continuously pay attention to the young teacher are classified into one type. Of course, there may be other listening and speaking modes of children classified as other types. The r and the minpts can be set by themselves during application, and the difference of the r and the minpts during application can cause the difference of the type quantity and the classification result of the later-stage division, but the division of the child types can be realized.
And 4, step 4: for either type: calculating the spatial distance between each child in the type and other children of the type according to the prestige frequency of each child in the type and the prestige state sequence corresponding to the target time period, calculating the behavior membership degree corresponding to each child in the type according to the spatial distance, taking the child with the maximum behavior membership degree as a representative child of the type corresponding to the target time period, and taking the representative child of the type corresponding to the target time period as a recommended simulation object when a young teacher looks back at the target time; the periscopic state sequence corresponding to the target time period is a sequence formed by enough periscopic states corresponding to the acquisition moments of the corresponding children in the target time period.
For each mode of listening and speaking, the teacher can simulate the head direction of the child in a classroom in a VR mode to simulate and experience the behavior of the child for teaching. This is inefficient, however, and does not allow for a good analysis of the key reference views and a determination of when the view may represent a location where the teacher needs to have increased expertise, resulting in the teacher having to traverse through many times to find the cause. Considering that for a plurality of children belonging to a listening and speaking mode, the young teacher only needs to select one of the most representative children for analysis, so that the present embodiment next selects the child representative in each listening and speaking mode, so that the young teacher only needs to simulate the child representative in each listening and speaking mode to quickly find the reason that the attention of the corresponding type of children is affected when reviewing at the later stage.
The present embodiment takes 8 minutes as a time period, which is a sliding sampling window, and calculates the following contents at each data update time as the time of the activity class changes. The present embodiment performs sampling once every 5 seconds, so the present embodiment updates the following calculation process every 5 seconds.
For any period of time in an activity classroom, a state space of attention of a child in a certain listening and speaking mode in the period of time is constructed. The present embodiment calculates the spatial distance between two children belonging to the same type at a certain time using the following formula:
wherein the content of the first and second substances,is the spatial distance between the child p and the child q,the child p is in the periscopic state sequence in the period of time, the periscopic state sequence is the set of the periscopic states corresponding to the sampling moments in the period of time,for the sequence of the states of attentiveness of the child q during this time, since the included attentiveness states of a group of children are not completely similar but may be approximately of the same type, when the attentiveness state of a child belongs to the situation represented by most children,the state features that can represent attention are similar, so distances are scaled to close distances in the hypothetical space, i.e., the term goes to 1, whereas it is kept farther away to longer distances.Can reflect the condition that the attention of the children is lost in a tidy classroom,the frequent degree of the look of the child p,the frequent degree of the children q, the times of the children p and q may be different, so that the state space of attention can further reflect the concentration and the ignorant behavior difference of one type of children. In this embodiment, the period of time is recorded as a target time, and when a time period corresponding to the period of time changes, the target time period also changes correspondingly.
For any child sample in the state space, calculating the Kth reachable distance of each childKth reachable distanceI.e. the distance at which a child sample radiates outward in the associated state space until the K-th adjacent sample is covered. In this embodiment, K is 5, and for K which cannot reach 5, calculation is not required, and K is regarded as an isolated sample; for K up to 5, the analysis was continued. When the system is applied specifically, an implementer can set the K value according to the number of children in a class.
At Kth reachable distance of child sample pCan cover K childrenSample q, constructing a set of all child samples q covered in the covered space. Therefore, the behavior membership degree of the child sample p can be obtained:
wherein, the first and the second end of the pipe are connected with each other,the behavior membership degree of the child sample p,is the spatial distance between the child sample p and the child sample q. When the behavior of the child is unified with other children, the higher the concentration degree of the characteristics of the concentration condition of the child p and the surrounding of the state space of the child p is, the smaller the reachable distance is, and the higher the membership degree is, which means that the behavior of the child is representative in the time period and has a higher reference value. Conversely, low concentration means that the degree of membership is lower, meaning that the behavior of the child is unique in the time period and has a smaller reference value.
The embodiment takes the child most representative in the period of time as the child representative to be observed in the period of time in the mode. And taking the set of most representative children corresponding to the time period in each mode as the child roles recommended to be simulated when the teacher looks back at the time period in the later period. The child representatives corresponding to different time periods are calculated, the roles of the recommended and simulated children are different when the young teacher looks back at the different time periods, the membership degree calculation is performed every 5 seconds, and the recommendation data are more real-time. As other embodiments, the membership level calculation may be performed at longer intervals to avoid large calculations or large fluctuations in the recommended data.
The embodiment obtains parameters capable of reflecting behaviors, attention and performance of children in an activity classroom, wherein the parameters comprise frequent degree of prestige of each child in the classroom, typical lagged time for interaction and lagged stability degree; classifying children in the classroom according to the parameters to obtain different types of children; next, analyzing the children included in each type, and selecting representative children in each type as recommended simulation objects for driving VR at a later stage, wherein compared with a mode of randomly selecting simulation objects, the method can improve the quality of a reply, help a young teacher find the defects when interacting with the children, and further deepen the training of professional ability; compared with a mode of simulating all children one by one, the tray-duplicating work load of the young teachers can be reduced, and the tray-duplicating efficiency is improved.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.
Claims (5)
1. A data processing method for educational training, comprising:
acquiring the frequent degree of prestige, typical delay time for interaction and delay stability degree of each child in a classroom of a target classroom;
the method for acquiring the frequent degree of the prospect specifically comprises the following steps: the proportion of the time that the deflection angle based on the head posture exceeds the tolerance angle to the whole class period; the typical late key time is the mean value of the first 30% of the maximum late key times in the late key times included in the late key time sequence; the late-suspected stability degree is calculated using the following formula:
wherein the content of the first and second substances,for a certain child's degree of hesitation to stabilize,the typical late time to doubt for that child,in order to find the maximum value,a late-suspected time series for the child;
calculating the attraction degree of each child in the classroom according to the typical delay time and the delay stability degree of each child to the interaction in the target classroom, comprising the following steps:
wherein the content of the first and second substances,in order to attract a child in a classroom,represents the typical hesitation time of the child, K represents the hesitation stability degree of the child, tanh () is a hyperbolic tangent function, alpha is a neighborhood correction parameter,,maximum of typical hesitation times for children within eight neighbourhoods of the child;
calculating the behavior difference distance between any two children according to the attraction degree, the prestige frequency degree and the delay time sequence of each child in the classroom of the target classroom; classifying the children in the target classroom according to the behavior difference distance to obtain a plurality of types; the delay time sequence is a sequence formed by the interaction delay times of the corresponding children in the class;
for either type: calculating the spatial distance between each child in the type and other children of the type according to the prestige frequency of each child in the type and the prestige state sequence corresponding to the target time period, calculating the behavior membership degree corresponding to each child in the type according to the spatial distance, taking the child with the maximum behavior membership degree as a representative child of the type corresponding to the target time period, and taking the representative child of the type corresponding to the target time period as a recommended simulation object when a young teacher looks back at the target time; the periscopic state sequence corresponding to the target time period is a sequence formed by the periscopic states corresponding to the acquisition moments of the corresponding children in the target time period.
2. The data processing method for educational training according to claim 1, wherein the behavior difference distance between any two children is calculated using the following formula:
wherein the content of the first and second substances,being the behavioural difference distance between child a and child B,to the extent of attraction of the class to the child a,to the extent of attraction of the class to the child B,to the extent that child a is looking at frequently,the frequent degree of the prestige of the child B,is the late-suspected time series for child a,is the late-suspected time series of child B,the cosine similarity is solved.
3. The data processing method for educational training according to claim 1, wherein the calculating the spatial distance between each child in the category and other children in the category according to the frequent degree of views of each child in the category and the sequence of views corresponding to the target time period comprises:
wherein the content of the first and second substances,is the spatial distance between the child p and the child q,for the sequence of states of the child p in the target time period,for children q in a target time periodThe sequence of states of the internal views,the frequent degree of the look of the child p,is the frequent degree of the child q.
4. The data processing method for educational practical training according to claim 3, wherein calculating the behavioral membership corresponding to each child in the category according to the spatial distance comprises:
the behavioral membership is calculated using the following formula:
5. The data processing method for educational training according to claim 1, wherein each child in the target classroom is classified using DBSCAN algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211107383.3A CN115186014B (en) | 2022-09-13 | 2022-09-13 | Data processing method for educational training |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211107383.3A CN115186014B (en) | 2022-09-13 | 2022-09-13 | Data processing method for educational training |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115186014A CN115186014A (en) | 2022-10-14 |
CN115186014B true CN115186014B (en) | 2022-12-02 |
Family
ID=83524489
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211107383.3A Active CN115186014B (en) | 2022-09-13 | 2022-09-13 | Data processing method for educational training |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115186014B (en) |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8930685B2 (en) * | 2011-12-13 | 2015-01-06 | International Business Machines Corporation | Deployment of a software image on multiple targets with streaming technique |
CN108389000A (en) * | 2018-03-09 | 2018-08-10 | 新华网股份有限公司 | Processing method, device and the computing device of child attention concentration degree information |
CN110070295B (en) * | 2019-04-25 | 2024-03-05 | 平安科技(深圳)有限公司 | Classroom teaching quality evaluation method and device and computer equipment |
CN110916631B (en) * | 2019-12-13 | 2022-04-22 | 东南大学 | Student classroom learning state evaluation system based on wearable physiological signal monitoring |
-
2022
- 2022-09-13 CN CN202211107383.3A patent/CN115186014B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN115186014A (en) | 2022-10-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104023230B (en) | A kind of non-reference picture quality appraisement method based on gradient relevance | |
US9219998B2 (en) | Participant grouping for enhanced interactive experience | |
CN109034036A (en) | A kind of video analysis method, Method of Teaching Quality Evaluation and system, computer readable storage medium | |
CN111915148A (en) | Classroom teaching evaluation method and system based on information technology | |
CN106204292A (en) | The methods of exhibiting of a kind of topological relation and device | |
CN110493803B (en) | Cell scene division method based on machine learning | |
Lance et al. | Sampling and evaluation | |
CN111507467A (en) | Neural network model training method and device, computer equipment and storage medium | |
Yang et al. | A multi-feature weighting based K-means algorithm for MOOC learner classification | |
CN113190747B (en) | Personalized resource recommendation method based on learning style and cognitive level | |
CN113591678B (en) | Classroom attention determination method, device, apparatus, storage medium, and program product | |
CN115186014B (en) | Data processing method for educational training | |
CN111259844A (en) | Real-time monitoring method for examinees in standardized examination room | |
CN113076885A (en) | Concentration degree grading method and system based on human eye action characteristics | |
CN109934188B (en) | Slide switching detection method, system, terminal and storage medium | |
Chen et al. | Intelligent teaching evaluation system integrating facial expression and behavior recognition in teaching video | |
Carr-Cornish et al. | Exploring the orientations which characterise the likely public acceptance of low emission energy technologies | |
Viana | Statistical methods for summarizing independent correlational results | |
CN114265954B (en) | Graph representation learning method based on position and structure information | |
Cheng et al. | The role of environmental education in waste segregation-at-source behaviour among households in Putrajaya | |
Huang et al. | Research on learning state based on students’ attitude and emotion in class learning | |
Yu et al. | Measuring acoustics with collaborative multiple agents | |
CN111738285A (en) | Evaluation method for evaluating learning efficiency and improving and optimizing learning efficiency of students | |
CN111369124A (en) | Image aesthetic prediction method based on self-generation global features and attention | |
Ahmad et al. | Requirements framework for engineering human-centered artificial intelligence-based software systems |
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 |