CN111340332A - Student team model building method for subject competition based on campus big data platform - Google Patents

Student team model building method for subject competition based on campus big data platform Download PDF

Info

Publication number
CN111340332A
CN111340332A CN202010084261.1A CN202010084261A CN111340332A CN 111340332 A CN111340332 A CN 111340332A CN 202010084261 A CN202010084261 A CN 202010084261A CN 111340332 A CN111340332 A CN 111340332A
Authority
CN
China
Prior art keywords
student
competition
team
ability
utility
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.)
Pending
Application number
CN202010084261.1A
Other languages
Chinese (zh)
Inventor
曲大鹏
吴松林
吕国鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Liaoning University
CERNET Corp
Original Assignee
Liaoning University
CERNET Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Liaoning University, CERNET Corp filed Critical Liaoning University
Priority to CN202010084261.1A priority Critical patent/CN111340332A/en
Publication of CN111340332A publication Critical patent/CN111340332A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A student team model building method for subject competition based on a campus big data platform comprises the steps of collecting student information through three data platforms including a educational administration management module, a laboratory management module and a competition management module, building a competition model and a student model, and finally obtaining a final building team through a team matching algorithm. The method has excellent performance in the aspects of student effectiveness and team effectiveness, and meanwhile, by the method, the learning ability, practice ability, team cooperation ability and innovation ability of students are improved, and the ability gap among individual students is reduced.

Description

Student team model building method for subject competition based on campus big data platform
Technical Field
The invention relates to a student team model building method, in particular to a student team model building method facing subject competition based on a campus big data platform.
Background
With the rapid development of higher education and the continuous increase of social demands, the realization of innovative venture education has become the main trend of current higher education. The subject competition is an important component of teaching practice, is an important way for realizing innovative entrepreneurship education, and is a good way for culturing the practical, innovative and engineering abilities of students. Subject competitions are receiving more and more attention from schools, teachers and students. Students and teachers want to improve the practice ability of students after obtaining theoretical knowledge in class, and enterprises want to attract more students to participate in their practice projects, so more and more subject competitions are held by different organizations. Because there are many discipline contest projects, and most discipline contests require multiple students to form a team, how to build a team of students is an important issue. For example, how many team members a student team should have, and how to select the "appropriate" competitor. In addition, competition brings the highest profit, but "appropriateness" is a complex and fuzzy concept, and not only needs to consider various attributes of practical ability, leadership ability and the like of students, but also needs to consider the purpose of competition. Furthermore, when multiple competitions are conducted simultaneously, students should not choose to participate in all competitions due to limited efforts.
At present, the organization of student teams for subject competition has not been studied deeply. The current construction scheme generally has two aspects: on one hand, selecting students with better theoretical performance to build a student team; on the other hand, students who select the same class or the same bedroom constitute a team. In the prior art, only the theoretical performance of students is considered or only classmates familiar to the students are selected to form a team, so that the reasonability is lacked, certain blindness is realized, and the problems of team homogenization, low income of the team and team members and the like are easily caused.
Disclosure of Invention
In order to solve the technical problems, the invention provides a student team model building method facing subject competition based on a campus big data platform.
In order to achieve the purpose, the invention adopts the technical scheme that: a student team model building method for subject competition based on a campus big data platform comprises the following steps:
1) collecting information: collecting behavior data of students through a big data platform, wherein the behavior data comprises daily learning data, experimental data and competition data;
2) establishing a competition model: for each contest GjConstructing a competition model G, wherein the competition model is represented by using a six-tuple:
<Gpf，Gcl，Gta，Gpa，Gce，Gre>
gpf is the professional field of competition, Gcl is competition rating, Gta is theoretical capacity considered by competition, Gpa is application capacity considered by competition, Gce is competition experience obtained by competition participation, and Gre is fatigue value consumed by competition participation;
3) establishing a student model: for each student SiConstructing a student model, wherein the student model is represented by using a six-tuple:
<Sta，Spa，Sce，Sla，Sca，Sae>
sta is theoretical ability, Spa is application ability, Sce is competition experience, Sla is leader ability, Sca is cooperation ability, and Sae is essence value;
3.1) establishing a student ability model: calculating the theoretical ability Sta (S) of the studenti) And the operating capability Spa (S)i) Competition Gj1Race experience of (Sce) (S)i，Gj1) Leadership Sla (S)i) Collaboration capability Sca (S)i) Energy value Sre (S)i)；
3.2) establishing a student utility model:
the student utility model also uses a six-tuple representation for the six attributes of the student:
<Uta，Upa，Uce，Ula，Uca，Uae>
uta is the theoretical ability utility of the student, Upa is the application ability utility, Uce is the competition experience utility, Ula is the leader ability utility, Uca is the collaboration ability utility, Uae is the effort value utility;
calculating student S based on utility modeliContest G injUtility and team T ofkThe total utility of;
4) team matching:
optimization using particle swarmAlgorithm (PSO) for solving contest GjOf the optimal team TkMake team TkThe utility and team T of each studentkThe total utility of (a) is maximized.
In the step 2) described above, the step of,
gpf: equally dividing the competition into discipline fields according to the academic fields granted to the academic degree, and uniformly distributing the discipline fields in a (0, 1) interval, wherein two competitions with similar professional fields are adjacent in the attribute Gpf;
gcl: quantification using Analytic Hierarchy Process (AHP), which is a structured technique for organizing and analyzing complex decisions to quantify contest rating attributes; firstly, the attribute is divided into three layers by using an analytic hierarchy process, namely a target layer, a criterion layer and a scheme layer in sequence, and the specific values are shown in table 1:
table 1: competition rating hierarchical table
Figure BDA0002380374820000031
Secondly, establishing a comparison matrix for each layer, wherein the diagonal line of the comparison matrix is 1, and the lower triangular matrix element in the matrix satisfies
Figure BDA0002380374820000032
Wherein a isjiThe element of the jth row and ith column in the matrix; then, by the formula
Figure BDA0002380374820000033
Calculating a weight vector of each layer, and performing consistency verification; finally, calculating the total weight of the scheme layer according to the normalized product of the weight of each criterion layer and the weight vector of the corresponding scheme layer;
by modeling the contest G, by extracting contest attributes, creating contest attribute vectorsj1Competition Gj2Similarity Simi (G)j1，Gj2) The formula is as follows:
Figure BDA0002380374820000034
in the step 3.1), the specific method comprises the following steps:
3.1.1)Sta(Si): student SiTheoretical capability Sta (S) ofi) The formula is as follows:
Figure BDA0002380374820000035
wherein N isTC(Si) Representing student SiThe number of the participating theoretical courses; tg (S)i，Ch) Representing student SiIn theory course ChThe score of (2); te (S)i，Ch) Representing student SiIn theory course ChThe classroom performance evaluation is determined by the attendance condition of students and the completion condition of classroom work; tc (C)h) As a theoretical course ChCorresponding credit points; omegatgAnd ωteAs a weight parameter, ωtg+ωtg＝1；
3.1.2)Spa(Si): the application ability reflects the practical ability of the student, and the application ability is quantified by using the weighted performance of the experimental course, namely the student SiAbility to exercise Spa (S)i) The formula is as follows:
Figure BDA0002380374820000041
wherein N isPC(Si) Representing student SiThe number of experimental courses to be taken; pg (S)i，Ch) Representing student SiIn experiment course ChThe score of (2); pc (C)h) For experiment course ChCorresponding credit points;
3.1.3)Sce(Si，Gj1): when there are a plurality of races to choose from, the student S will have a high probability of participating in a race similar to that in which he participated and may achieve a good resultiParticipate in contest Gj1Race experience of (Sce) (S)i，Gj1) As follows:
Figure BDA0002380374820000042
wherein the content of the first and second substances,
Figure BDA0002380374820000048
indicating the number of students that have participated in the competition; gce (G)j2) Show contest Gj2The competition experience of (2); sami (G)j1，Gj2) Show contest Gj1Competition Gj2The similarity of (2);
3.1.4)Sla(Si) And Sca (S)i): after the competition is finished, the leadership ability and the collaboration ability of the competition participants are obtained by evaluating the members in the same group, so that the student SiLeader capacity Sla (S)i) And collaboration capability Sca (S)i) The formula is as follows:
Figure BDA0002380374820000043
Figure BDA0002380374820000044
wherein the content of the first and second substances,
Figure BDA0002380374820000049
representing student SiThe number of participating teams; ave (Ela (S)i，Tk) Represents a student SiIn team TkAverage leadership evaluation in (1); ave (Eca (S)i，Tk) Represents a student SiIn team TkThe average cooperative ability evaluation in (1);
3.1.5)Sre(Si): at the beginning of the school year, the student has the greatest energy and consumes the fatigue value corresponding to the competition when he participates in the competition, and therefore, the student SiEnergy value Sre (S)i) The formula is as follows:
Figure BDA0002380374820000046
wherein the content of the first and second substances,
Figure BDA0002380374820000047
representing student SiThe number of contests G participated in; gre (G)j) Show contest GjThe fatigue value of (a).
The specific calculation in the step 3.2) is as follows:
3.2.1)Uta：
Uta(Si，Gj)＝Sta(Si)×Gta(Gj)
3.2.2)Upa：
Upa(Si，Gj)＝Spa(Si)×Gpa(Gj)
3.2.3)Uce：
Uce(Si，Gj)＝Sce(Si，Gj)×Gce(Gj)
3.2.4)Ula：
Ula(Si，Tk)＝Sla(Si)×Ela(Tk)
wherein, Ela (T)k) Representation team TkThe leadership ability evaluation value of all students is expressed as follows:
Figure BDA0002380374820000051
wherein, ave (Sla (T)k) Represent team TkAverage of all student leadership, higher values mean better leadership of the team; var (Sla (T)k) Represent team TkThe variance of the leadership abilities of all students is higher, which means that the leadership abilities of the members of the team are different more, and a more harmonious team is easily formed;
3.2.5)Uca：
Uca(Si，Tk)＝Sca(Si)×Eca(Tk)
wherein Eca (T)k) Representation team TkThe cooperative ability evaluation value of all students is as follows:
Figure BDA0002380374820000052
wherein, ave (Sca (T)k) Represent team TkThe average value of all students' cooperative ability, and the higher the value is, the better the cooperative ability of the team is; var (Sca (T)k) Represent team TkThe variance of the cooperative ability of all students, the smaller the value is, the smaller the difference of the cooperative ability among the students is, the easier the cooperation is;
3.2.6) Uae: student SiContest G injThe effort value utility formula of (a) is as follows:
Figure BDA0002380374820000053
in the formula, the value range of the utility is limited by a sine function to be [ -1, 1], and when the energy value is too low or negative, the student participates in too much competition, so that the energy is insufficient, and negative effects are caused;
in the step 3.3), the concrete steps are as follows: based on the utility model, student SiContest G injThe utility formula of (a) is as follows:
Figure BDA0002380374820000061
wherein, ω ista、ωpa、ωce、ωaeAs a weight parameter, 0 ≦ ωta、ωpa、ωce、ωaeNot more than 1, and omegata+ωpa+ωce+ωaeα is a weight parameter, 0 is more than or equal to α and less than or equal to 1, and is used for balancing the proportion of the leadership ability utility and the cooperation ability;
team TkThe general utility formula of (a) is as follows:
Figure BDA0002380374820000062
the method has the advantages that: the invention collects and analyzes the behaviors of students by establishing a big data platform to describe the students, carries out modeling analysis on competitions and student teams, matches the proper teams for the students to participate in proper subject competitions so as to improve various abilities of the students, has the functions of reducing the difference of the whole ability level of the students, enhancing the innovation and creation ability of the students, avoiding the blindness of the students when participating in the subject competitions, and can effectively help teachers and students to establish proper subject competition teams.
Drawings
FIG. 1: big data platform schematic diagram.
FIG. 2 a: average team effectiveness graph.
FIG. 2 b: a team utility variance map.
FIG. 3 a: mean effectiveness chart.
FIG. 3 b: student utility variance map.
FIG. 4 a: average student performance.
FIG. 4 b: student ability variance plot.
Detailed Description
One, big data platform
For collecting and analyzing the behavior of students to describe their attributes. As shown in fig. 1, there are three layers, namely, a collection layer, an analysis layer, and a property layer. In addition, the acquisition layer also comprises three modules of teaching management, laboratory management and competition management. These modules are used to collect and analyze theoretical behavior data, experimental behavior data, and competition behavior data, respectively. And analyzing and converting the theoretical behavior data and the experimental behavior data into theoretical capacity and experimental capacity respectively. And (4) converting competition behavior data analysis into competition experience, leadership capability, cooperation capability and precision value.
Competition model
For each contest Gj，1≤j≤NG，NGIs the number of races. Competition model uses a six-tuple<Gpf，Gcl，Gta，Gpa，Gce，Gre>The representation represents the professional field of the competition, competition rating, theoretical ability and operational ability considered by the competition, competition experience obtained by the competition and fatigue value consumed.
The areas of expertise are equally divided into areas of discipline awarded to the degree and evenly distributed within the (0, 1) interval to ensure that two contests with similar areas of expertise will be a close distance apart in attribute Gpf.
Contest ratings are quantified using Analytic Hierarchy Process (AHP), which is a structured technique for organizing and analyzing complex decisions to quantify contest rating attributes. Firstly, the attribute is divided into three layers, namely a target layer, a criterion layer and a scheme layer in sequence by using an analytic hierarchy process, and the specific values are shown in table 1:
table 1: competition rating hierarchical table
Figure BDA0002380374820000071
Secondly, establishing a comparison matrix for each layer, wherein the diagonal line of the comparison matrix is 1, and the lower triangular matrix element in the matrix satisfies
Figure BDA0002380374820000072
Wherein a isjiIs the element in the jth row and ith column in the matrix. Then, by the formula
Figure BDA0002380374820000073
And calculating a weight vector of each layer, and performing consistency verification. And finally, calculating the total weight of the scheme layer according to the normalized product of the weight of each criterion layer and the weight vector of the corresponding scheme layer.
The theoretical ability and the operational ability considered by the competition, the competition experience obtained by the competition and the fatigue value consumed are determined by using an expert investigation method.
By modeling the contest G, by extracting contest attributes, creating contest attribute vectorsj1Competition Gj2Similarity Simi (G)j1，Gj2) As shown in equation (1):
Figure BDA0002380374820000081
similarity meterComputing Sce (S)i，Gj1) It is used at any time.
Model for student
For each student Si，1≤i≤NS，NSThe number of students. Using one hexahydric group for student models<Sta，Spa，Sce，Sla，Sca，Sae>And (3) expressions which respectively represent theoretical ability, application ability, competition experience, leadership ability, collaboration ability and power value of the students.
1. Capability model
The theoretical ability reflects the student' S ability to solve theoretical problems, quantified using the performance and classroom performance of a theoretical course, student SiTheoretical capability Sta (S) ofi) As shown in equation (2).
Figure BDA0002380374820000082
Wherein N isTC(Si) Representing student SiThe number of the participating theoretical courses; tg (S)i，Ch) Representing student SiIn theory course ChThe score of (2); te (S)i，Ch) Representing student SiIn theory course ChThe classroom performance evaluation is determined by the attendance condition of students, the completion condition of classroom work and the like; tc (C)h) As a theoretical course ChCorresponding credit points; omegatgAnd ωteAs a weight parameter, ωtg+ωtg＝1。
The application ability reflects the practical ability of the student, and the application ability is quantified by using the weighted performance of the experimental course, namely the student SiAbility to exercise Spa (S)i) As shown in equation (3).
Figure BDA0002380374820000083
Wherein N isPC(Si) Representing student SiThe number of experimental courses to be taken; pg (S)i，Ch) Representing student SiIn experiment course ChThe score of (2); pc (C)h) For experiments toCourse ChCorresponding credit points.
When there are multiple games available for selection, the student will have a high probability of participating in a race similar to that in which he has participated and may achieve a good performance. Thus, student SiParticipate in contest Gj1Race experience of (Sce) (S)i，Gj1) As shown in equation (4).
Figure BDA0002380374820000084
Wherein the content of the first and second substances,
Figure BDA0002380374820000094
indicating the number of students that have participated in the competition; gce (G)j2) Show contest Gj2The competition experience of (2); sami (G)j1，Gj2) Show contest Gj1Competition Gj2The similarity of (c).
When the competition is over, the leadership and collaboration ability of the competition participants will be evaluated by the other members of the same group, and therefore, student SiLeader capacity Sla (S)i) And collaboration capability Sca (S)i) As shown in equations (5) and (6).
Figure BDA0002380374820000091
Figure BDA0002380374820000092
Wherein the content of the first and second substances,
Figure BDA0002380374820000095
representing student SiThe number of participating teams; ave (Ela (S)i，Tk) Represents a student SiIn team TkAverage leadership evaluation in (1); ave (Eca (S)i，Tk) Represents a student SiIn team TkAverage cooperative ability evaluation in (1).
As the energy and time of the students are consumed in the competition, the aim is to ensureThe students are proved to have enough energy to participate in the competition, so the concept of the introduced energy value represents the energy state of the students in a learning year. At the beginning of the school year, the students have sufficient energy available to consume the fatigue value associated with the competition when they participate in the competition. Thus, student SiEnergy value Sre (S)i) As shown in equation (7).
Figure BDA0002380374820000093
Wherein the content of the first and second substances,
Figure BDA0002380374820000096
representing student SiThe number of contests G participated in; gre (G)j) Show contest GjThe fatigue value of (a).
2. Utility model
When a student participates in a competition, the student receives a corresponding profit, and the profit received by the student is defined as "utility". Corresponding to six attributes of the student, the student utility model is also represented by a six-tuple < Uta, Upa, Uce, Ula, Uca, Uae > which respectively represents the theoretical ability utility, the application ability utility, the competition experience utility, the leadership ability utility, the cooperation ability utility and the energy value utility of the student.
Student SiContest G injThe theoretical capacity utility of (c) is shown in equation (8).
Uta(Si，Gj)＝Sta(Si)×Gta(Gj) (8)
Student SiContest G injThe utility of the exercise capacity of (a) is shown in formula (9).
Upa(Si，Gj)＝Spa(Si)×Gpa(Gj) (9)
Student SiContest G injThe competition experience utility of (c) is shown in equation (10).
Uce(Si，Gj)＝Sce(Si，Gj)×Gce(Gj) (10)
Student SiContest G injThe leadership ability utility of (c) is shown in equation (11).
Ula(Si，Tk)＝Sla(Si)×Ela(Tk) (11)
Wherein, Ela (T)k) Representation team TkAnd (3) evaluating the leadership abilities of all students as shown in formula (12).
Figure BDA0002380374820000101
Wherein, ave (Sla (T)k) Represent team TkAverage of all student leadership, higher values mean better leadership of the team; var (Sla (T)k) Represent team TkThe variance of all student leadership, higher values mean that the difference of the leadership of the team members is larger, and a more harmonious team is easily formed.
Student SiContest G injThe collaboration capability utility of (c) is shown in equation (13).
Uca(Si，Tk)＝Sca(Si)×Eca(Tk) (13)
Wherein Eca (T)k) Representation team TkThe cooperation ability evaluation values of all students are shown by formula (14).
Figure BDA0002380374820000102
Wherein, ave (Sca (T)k) Represent team TkThe average value of all students' cooperative ability, and the higher the value is, the better the cooperative ability of the team is; var (Sca (T)k) Represent team TkThe variance of the cooperative ability of all students, the smaller the value means that the difference of the cooperative ability among the students is smaller, and the cooperation is easier.
Student SiContest G injThe effort value utility of (c) is shown in equation (15).
Figure BDA0002380374820000103
In the formula, the value range of the utility is limited by the sine function to be [ -1, 1], and when the energy value is too low or negative, the student participates in the competition too much, so that the energy is insufficient, and the participated competition is influenced negatively.
Based on the utility model, student SiContest G injThe utility of (c) is shown in equation (16).
Figure BDA0002380374820000104
Wherein, ω ista、ωpa、ωce、ωaeAs a weight parameter, 0 ≦ ωta、ωpa、ωce、ωaeNot more than 1, and omegata+ωpa+ωce+ωaeThe index is 1 and is used for balancing the proportion of theoretical ability utility, application ability utility, competition experience utility and energy value utility, α is a weight parameter, 0 is more than or equal to α is less than or equal to 1 and is used for balancing the proportion of leadership ability utility and cooperation ability.
Team TkThe total utility of (c) is shown in formula (17).
Figure BDA0002380374820000111
Four, team matching algorithm
Based on the above analysis, the utility of the best team and members is obtained by selecting the appropriate student building team to participate in the appropriate competition, with optimization objectives as shown in equation (18):
Figure BDA0002380374820000112
the problem is a multi-constraint multi-objective optimization problem, so that a feasible solution can be found by adopting an intelligent algorithm, and a Particle Swarm Optimization (PSO) algorithm is adopted as a solving method.
Using p for particles, PospAnd VelpIndicates the position and velocity of the particle p, and is a particleTwo attributes of child p, i.e. team to be solved TkMember set and member replacement speed. pbestfitRepresents the optimal fitness of single iteration of all particles in the iteration process, namely the team T to be solvedkLocal optimal utility of; gbestfitShows the optimal fitness of all the particles in the past iteration until the current iteration, namely the team T to be solvedkGlobal optimal utility. pbestposIndicating that in a single iteration, pbest is reachedfitThe position of the particle p, i.e. the team to be solved TkA local optimal team member set of (a); gbestposIndicating the optimal position of the particle in the previous iteration until the current iteration, i.e. the team to be solved TkA set of globally optimal team members.
The solving process is as follows: firstly, randomly initializing the position and speed of each particle, namely randomly initializing the member composition and member replacement speed of a team; second, the candidate solution is iteratively updated, the fitness of each particle is calculated according to equation (18), and pbest is updatedposAnd pbestfitIf pbestfitIs superior to gbestfitIf so, update the gbestposAnd gbestfit(ii) a Then, the Pos of each particle is updatedpAnd Velp(ii) a Finally, after multiple iterations, gbestposI.e. the optimal team member set, i.e. the solution team Tk。
Fifth, experimental setup and results
1. Experimental setup
Based on the model, simulation experiments are carried out on the scene that 88 computer science and technical professionals join in competition within one school year. The number of competitions is set to 5, each competition is set to 3 teams, and each team has 5 members.
We selected three currently mainstream student team construction methods FCTA, FCPA and FCTPA for comparative experiments with the CTB method herein. FCTA is a theoretical ability priority method, namely randomly selecting students with theoretical ability ranking ahead to form a team; FCPA is an application ability priority method, namely, students with top application ability ranking are randomly selected to form a team; FCTPA is a theoretical ability and application ability priority method, namely, a student building team with the theoretical ability and the application ability ranked at the top is randomly selected. We define a capability value above 0.75 as top ranked. In the sample, 57% of students ' theoretical ability ranks first, 47% of students ' exercise ability ranks first, and 38% of students ' theoretical ability and exercise ability rank first.
In equation (16), the weight ωta、ωpa、ωce、ωaeSet to 0.25, α to 0.5 the number of particles in the PSO algorithm is set to 10 and the number of iterations is set to 1000.
We used the following indices for performance evaluation and comparison: average Team Utility (ATU), i.e., the average utility of each team participating in the current contest; a team utility Variance (VTU), i.e., the variance of the utility of each team participating in the current competition; average Student Utility (ASU), i.e., the average utility of students participating in the current competition; student utility Variance (VSU), i.e., the variance of the utility of students participating in the current competition; average Student Ability (ASA), i.e. the average of all students' ability; student ability Variance (VSA), i.e. the variance of all student abilities, is shown in fig. 2 a-2 b.
2 a-2 b compare four team construction methods. The results show that the ATU of CTB is highest because six attributes of the student are considered in the team building process. While FCTA, FCPA and FCTPA have similar lower ATU because these methods only consider students with top ranking capabilities from a different perspective, which chooses to have a large number of overlapping students. Notably, in the five races, the restriction of selecting only the top ranked students reduces team utility for FCTA, FCPA and FCTPA, while CTB considers all students and improves team utility. Therefore, the VTU of CTB does not increase significantly, and other methods may increase rapidly.
As shown in fig. 3a, the ASU of the CTB remains stable, while the other three methods decrease as the number of races increases. Since CTB considers six attributes of students in the team building process, while FCTA, FCPA and FCTPA consider only top ranked students, and since FCTA, FCPA and FCTPA selectable students are limited, excellent students are repeatedly selected into teams to participate in different competitions, thereby causing the energy value of students to be consumed excessively and adversely affecting. As shown in fig. 3b, the VSU of the CTB remains stable, while the other three methods remain stable earlier, and rise sharply when the number of races is 5.
The abilities of the students are determined by the weighted sum of the six attributes of the students, and the abilities of all the students reflect the influence of the team building method on the overall ability level of the students. As shown in fig. 4a, ASA keeps increasing in all the methods, but CTB keeps steadily increasing at race number 5, and the other three methods are not effective due to insufficient numbers of students to choose from. As shown in fig. 4b, as the number of races increases, the VSA of the CTB remains down, while the other three methods remain up, because the CTB considers all students and tries to narrow the differences between students, while the other methods only consider top ranked students.
Sixth, conclusion
Compared with the traditional method of only selecting the top-ranked students to participate in the competition, the method is novel. The patent establishes a big data platform to collect and analyze student behavior data, and establishes a student and a competition model based on six attributes respectively. In addition, a utility function is defined for each attribute of the student to represent the benefit in that attribute when the student participates in the game. Finally, the utility of students and teams is maximized using the PSO algorithm. Simulation results show that the method has excellent performance in the aspects of student utility and team utility. In addition, the method reduces the capability gap among students on the basis of improving the overall capability level of the students, and is helpful for cultivating more students from more aspects in innovative entrepreneurship education.

Claims (5)

1. A student team model building method for subject competition based on a campus big data platform is characterized by comprising the following steps:
1) collecting information: collecting behavior data of students through a big data platform, wherein the behavior data comprises daily learning data, experimental data and competition data;
2) establishing a competition model: for each contest GjConstruction of a Competition modelG, the competition model uses a six-tuple representation:
<Gpf，Gcl，Gta，Gpa，Gce，Gre>
gpf is the professional field of competition, Gcl is competition rating, Gta is theoretical capacity considered by competition, Gpa is application capacity considered by competition, Gce is competition experience obtained by competition participation, and Gre is fatigue value consumed by competition participation;
3) establishing a student model: for each student SiConstructing a student model, wherein the student model is represented by using a six-tuple:
<Sta，Spa，Sce，Sla，Sca，Sae>
sta is theoretical ability, Spa is application ability, Sce is competition experience, Sla is leader ability, Sca is cooperation ability, and Sae is essence value;
3.1) establishing a student ability model: calculating the theoretical ability Sta (S) of the studenti) And the operating capability Spa (S)i) Competition Gj1Race experience of (Sce) (S)i，Gj1) Leadership Sla (S)i) Collaboration capability Sca (S)i) Energy value Sre (S)i)；
3.2) establishing a student utility model:
the student utility model also uses a six-tuple representation for the six attributes of the student:
<Uta，Upa，Uce，Ula，Uca，Uae>
uta is the theoretical ability utility of the student, Upa is the application ability utility, Uce is the competition experience utility, Ula is the leader ability utility, Uca is the collaboration ability utility, Uae is the effort value utility;
3.3) calculating student S based on utility modeliContest G injUtility and team T ofkThe total utility of;
4) team matching:
solving for contest G by particle swarm optimization algorithmjOf the optimal team TkMake team TkThe utility and team T of each studentkThe total utility of (a) is maximized.
2. The campus big data platform based subject competition oriented student team modeling method as claimed in claim 1, wherein: in the step 2) described above, the step of,
gpf: equally dividing the competition into discipline fields according to the academic fields granted to the academic degree, and uniformly distributing the discipline fields in a (0, 1) interval, wherein two competitions with similar professional fields are adjacent in the attribute Gpf;
gcl: quantification using Analytic Hierarchy Process (AHP), which is a structured technique for organizing and analyzing complex decisions to quantify contest rating attributes; firstly, the attribute is divided into three layers by using an analytic hierarchy process, namely a target layer, a criterion layer and a scheme layer in sequence, and the specific values are shown in table 1:
table 1: competition rating hierarchical table
Figure FDA0002380374810000021
Secondly, establishing a comparison matrix for each layer, wherein the diagonal line of the comparison matrix is 1, and the lower triangular matrix element in the matrix satisfies
Figure FDA0002380374810000022
Wherein a isjiThe element of the jth row and ith column in the matrix; then, by the formula
Figure FDA0002380374810000023
Calculating a weight vector of each layer, and performing consistency verification; finally, calculating the total weight of the scheme layer according to the normalized product of the weight of each criterion layer and the weight vector of the corresponding scheme layer;
by modeling the contest G, by extracting contest attributes, creating contest attribute vectorsj1Competition Gj2Similarity Simi (G)j1，Gj2) The formula is as follows:
Figure FDA0002380374810000024
3. the campus big data platform based subject competition oriented student team modeling method as claimed in claim 1, wherein: in the step 3.1), the specific method comprises the following steps:
3.1.1)Sta(Si): student SiTheoretical capability Sta (S) ofi) The formula is as follows:
Figure FDA0002380374810000031
wherein N isTC(Si) Representing student SiThe number of the participating theoretical courses; tg (S)i，Ch) Representing student SiIn theory course ChThe score of (2); te (S)i，Ch) Representing student SiIn theory course ChThe classroom performance evaluation is determined by the attendance condition of students and the completion condition of classroom work; tc (C)h) As a theoretical course ChCorresponding credit points; omegatgAnd ωteAs a weight parameter, ωtg+ωtg＝1；
3.1.2)Spa(Si): the application ability reflects the practical ability of the student, and the application ability is quantified by using the weighted performance of the experimental course, namely the student SiAbility to exercise Spa (S)i) The formula is as follows:
Figure FDA0002380374810000032
wherein N isPC(Si) Representing student SiThe number of experimental courses to be taken; pg (S)i，Ch) Representing student SiIn experiment course ChThe score of (2); pc (C)h) For experiment course ChCorresponding credit points;
3.1.3)Sce(Si，Gj1): when there are a plurality of races to choose from, the student S will have a high probability of participating in a race similar to that in which he participated and may achieve a good resultiParticipate in contest Gj1Race experience of (Sce) (S)i，Gj1) As follows:
Figure FDA0002380374810000033
wherein the content of the first and second substances,
Figure FDA0002380374810000034
indicating the number of students that have participated in the competition; gce (G)j2) Show contest Gj2The competition experience of (2); sami (G)j1，Gj2) Show contest Gj1Competition Gj2The similarity of (2);
3.1.4)Sla(Si) And Sca (S)i): after the competition is finished, the leadership ability and the collaboration ability of the competition participants are obtained by evaluating the members in the same group, so that the student SiLeader capacity Sla (S)i) And collaboration capability Sca (S)i) The formula is as follows:
Figure FDA0002380374810000035
Figure FDA0002380374810000036
wherein the content of the first and second substances,
Figure FDA0002380374810000037
representing student SiThe number of participating teams; ave (Ela (S)i，Tk) Represents a student SiIn team TkAverage leadership evaluation in (1); ave (Eca (S)i，Tk) Represents a student SiIn team TkThe average cooperative ability evaluation in (1);
3.1.5)Sre(Si): at the beginning of the school year, the student has the greatest energy and consumes the fatigue value corresponding to the competition when he participates in the competition, and therefore, the student SiEnergy value Sre (S)i) The formula is as follows:
Figure FDA0002380374810000041
wherein the content of the first and second substances,
Figure FDA0002380374810000042
representing student SiThe number of contests G participated in; gre (G)j) Show contest GjThe fatigue value of (a).
4. The campus big data platform based subject competition oriented student team modeling method as claimed in claim 1, wherein: the specific calculation in the step 3.2) is as follows:
3.2.1)Uta：
Uta(Si，Gj)＝Sta(Si)×Gta(Gj)
3.2.2)Upa：
Upa(Si，Gj)＝Spa(Si)×Gpa(Gj)
3.2.3)Uce：
Uce(Si，Gj)＝Sce(Si，Gj)×Gce(Gj)
3.2.4)Ula：
Ula(Si，Tk)＝Sla(Si)×Ela(Tk)
wherein, Ela (T)k) Representation team TkThe leadership ability evaluation value of all students is expressed as follows:
Figure FDA0002380374810000043
wherein, ave (Sla (T)k) Represent team TkAverage of all student leadership, higher values mean better leadership of the team; var (Sla (T)k) Represent team TkThe variance of the leadership abilities of all students is higher, which means that the leadership abilities of the members of the team are different more, and a more harmonious team is easily formed;
3.2.5)Uca：
Uca(Si，Tk)＝Sca(Si)×Eca(Tk)
wherein Eca (T)k) Representation team TkThe cooperative ability evaluation value of all students is as follows:
Figure FDA0002380374810000044
wherein, ave (Sca (T)k) Represent team TkThe average value of all students' cooperative ability, and the higher the value is, the better the cooperative ability of the team is; var (Sca (T)k) Represent team TkThe variance of the cooperative ability of all students, the smaller the value is, the smaller the difference of the cooperative ability among the students is, the easier the cooperation is;
3.2.6) Uae: student SiContest G injThe effort value utility formula of (a) is as follows:
Figure FDA0002380374810000051
in the formula, the value range of the utility is limited by the sine function to be [ -1, 1], and when the energy value is too low or negative, the result means that the students take too much competition to cause insufficient energy, and negative effects can be caused.
5. The campus big data platform based subject competition oriented student team modeling method as claimed in claim 1, wherein: in the step 3.3), the concrete steps are as follows: based on the utility model, student SiContest G injThe utility formula of (a) is as follows:
Figure FDA0002380374810000052
wherein, ω ista、ωpa、ωce、ωaeAs a weight parameter, 0 ≦ ωta、ωpa、ωce、ωaeNot more than 1, and omegata+ωpa+ωce+ωaeα is a weight parameter, 0 is more than or equal to α and less than or equal to 1, and is used for balancing the proportion of the leadership ability utility and the cooperation ability:
team TkIs given as the total utility formula
Figure FDA0002380374810000053
CN202010084261.1A 2020-02-06 2020-02-06 Student team model building method for subject competition based on campus big data platform Pending CN111340332A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010084261.1A CN111340332A (en) 2020-02-06 2020-02-06 Student team model building method for subject competition based on campus big data platform

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010084261.1A CN111340332A (en) 2020-02-06 2020-02-06 Student team model building method for subject competition based on campus big data platform

Publications (1)

Publication Number Publication Date
CN111340332A true CN111340332A (en) 2020-06-26

Family

ID=71183441

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010084261.1A Pending CN111340332A (en) 2020-02-06 2020-02-06 Student team model building method for subject competition based on campus big data platform

Country Status (1)

Country Link
CN (1) CN111340332A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116976544A (en) * 2022-04-14 2023-10-31 武汉东湖学院 Information management platform for college student discipline competition

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106503895A (en) * 2016-10-21 2017-03-15 西北工业大学 A kind of virtual contract network Team Member method for optimizing
CN106621327A (en) * 2016-12-22 2017-05-10 竞技世界（北京）网络技术有限公司 User grouping method and system based on extreme value
CN109559033A (en) * 2018-11-27 2019-04-02 河南科技大学 A kind of facing cloud manufactures and designs the social Team Member preferred method of mode
CN110288148A (en) * 2019-06-21 2019-09-27 福建师范大学 A kind of student's organizing method of open Practice Curriculum
CN110750733A (en) * 2019-10-15 2020-02-04 赛尔网络有限公司 Service recommendation method, device, equipment and medium based on school identity portrait

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106503895A (en) * 2016-10-21 2017-03-15 西北工业大学 A kind of virtual contract network Team Member method for optimizing
CN106621327A (en) * 2016-12-22 2017-05-10 竞技世界（北京）网络技术有限公司 User grouping method and system based on extreme value
CN109559033A (en) * 2018-11-27 2019-04-02 河南科技大学 A kind of facing cloud manufactures and designs the social Team Member preferred method of mode
CN110288148A (en) * 2019-06-21 2019-09-27 福建师范大学 A kind of student's organizing method of open Practice Curriculum
CN110750733A (en) * 2019-10-15 2020-02-04 赛尔网络有限公司 Service recommendation method, device, equipment and medium based on school identity portrait

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116976544A (en) * 2022-04-14 2023-10-31 武汉东湖学院 Information management platform for college student discipline competition

Similar Documents

Publication Publication Date Title
Lin et al. An automatic group composition system for composing collaborative learning groups using enhanced particle swarm optimization
CN107564391A (en) Public administration method and training platform based on simulation sand table
Chen et al. A group division method based on collaborative learning elements
Kong Innovative applications mode of network learning space in exercise physiology based on ubiquitous learning
Ma et al. Research on the construction and application of teacher-student interaction evaluation system for smart classroom in the post COVID-19
CN111340332A (en) Student team model building method for subject competition based on campus big data platform
Hussein Quasi-experimental method to identify the impact of ambiguity and urgency on project participants in the early project phase
Jiao et al. [Retracted] Football Teaching Quality Evaluation and Promotion Strategy Based on Intelligent Algorithms in Higher Vocational Colleges
Chen et al. Online cooperative teaching mode based on self-direction theory in method of sport science research
Zhang et al. RETRACTED: Novel evaluation approach for extracurricular sports lifestyle of college students
Mori et al. Optimising of support plans for new graduate employment market using reinforcement learning
Jamebozorg et al. Relationship between designing computer-based educational games, and learning motivation among elementary students
Zhang Research on the reform path of soccer special class for physical education majors in colleges and universities under the vision of intelligent sports
Huang et al. Construction of wisdom classroom for English majors based on back propagation
Sharmal et al. Fuzzy analytical hierarchy process to evaluate the curriculum of an undergraduate program with a vision to design an industry-ready undergraduate engineering program.
Zhao et al. Prediction model of sports results base on knowledge discovery in data-base
Learmonth et al. Participatory simulation as a tool of policy informatics: definitions, literature review, and research directions
Faraji Abdolmaleki et al. Bring Joy to Gamers: Adding Renewable Energy Alternative through Sustainable Development Indicators
Yang et al. Research on the Development Strategy of Ethnic Traditional Sports Based on Optimization Theory
Huang et al. Knowledge Map Automatic Update System Using Graph Convolutional Network
Zhang The Development Trend of Innovative Teaching of Dance in Colleges and Universities Based on Clustering Algorithm in the Context of Big Data
Xu Research on the Design of Online Teaching Platform of College Dance Course based on IGA Algorithm
Gao Research on Dynamic Early Warning Algorithm of Ideological Education Based on Genetic Algorithm
Chen Case Design and Analysis of Digital Media Courses Incorporating Aesthetic Education
Bokal Optimizing enjoyment of mathematics and OR education with introducing psychological concepts flow and grit using simulation-based model of emotional states of learning

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