=Paper=
{{Paper
|id=Vol-2105/10000365
|storemode=property
|title=Curriculum Optimization by the Criteria of Maximizing Professional
Value and the Connection Coefficient of Educational Elements, Using
Software Tools
|pdfUrl=https://ceur-ws.org/Vol-2105/10000365.pdf
|volume=Vol-2105
|authors=Nataliia Kravchenko,Hanna Alekseeva,Larysa Gorbatyuk
|dblpUrl=https://dblp.org/rec/conf/icteri/KravchenkoAG18
}}
==Curriculum Optimization by the Criteria of Maximizing Professional
Value and the Connection Coefficient of Educational Elements, Using
Software Tools==
https://ceur-ws.org/Vol-2105/10000365.pdf
Curriculum Optimization by the Criteria of Maximizing
Professional Value and the Connection Coefficient of
Educational Elements, Using Software Tools
N. Kravchenko [0000-0002-9642-5403], H. Alekseeva[0000-0003-3204-3139], L. Gorbatyuk
Berdyansk State Pedagogical University, Schmidt Str. 4, 71100 Berdyansk, Ukraine
natalyvkravchenko@gmail.com, alekseeva@ukr.net, loravas@ukr.net
Abstract. In the article, the process of designing a curriculum is formalized. The
structure of the curriculum for the training of specialists is proposed as a set of
structural-logical schemes and a set of educational elements. The filling of the
developed structural-logical scheme by educational elements allows: to clearly
define the content of the educational material and the purpose of training; to pro-
vide a clear logical sequence of educational disciplines; to use the automated
presentation of the structure of the curriculum. The method for improving the
structure of the training curriculum by the criterion of the general significance of
educational disciplines content modules has been developed. The informational
system for improving the structure of the curriculum of training according to the
criteria of general significance has been developed.
Keywords: Curriculum optimization, Curriculum development, Vocational
training.
1 Introduction
The main purpose of universities is to produce well-qualified specialists. At the same
time, the quality of the training of specialists depends on how well the graduate can
meet the modern workplace requirements. The quality of specialist’s training is largely
determined by his course. The training of professionals that meet modern demand en-
tails the continuous improvement of educational programs. Therefore, the curriculum
must be flexible enough to quickly adapt to changing requirements [1, 2].
Currently, the process of designing a curriculum is based on the experience and in-
tuition of university staff, therefore it needs improvement and scientific evidence/sup-
port for the decisions made.
The development of modern technical and software, methods of constructing deci-
sion support systems makes it possible to improve the efficiency of work in educational
institutions through the introduction of modern methods for processing information [3],
construction and implementation of mathematical methods and models, tools for the
search, analysis, formation and optimization of managerial decisions [4, 5].
The problem of automating curricula designing was considered in many scientific
papers. How to optimize the logical structure of curricula based on the apparatus of
graph theory was considered in the following works [6, 7, 8]. Improvement of making
curricula of higher educational institutions using hierarchical trees of specialist training
purposes was considered in the work [9]. In the work [10] algorithms of curricula im-
provement based on relationships between modules were proposed.
Scientific publications devoted to the optimization of curricula, considering the nec-
essary condition that the volume of modules should be equal to the volume of the cur-
riculum [11]. The problem of optimization the logical structure of curricula on the basis
of the theory of graph apparatus, was considered in works [6, 7]. Improving the com-
pilation of curricula of universities using hierarchical trees for the purposes of specialist
training was considered in the works [9]. Algorithms of optimization of curricula on
the basis of relationships between modules were offered in work [10]. It should be noted
that most works fix only the presence or absence of relationships between modules.
Analysis of works on optimization of curricula [6, 9, 11, 12] showed that the main
methods of optimizing curriculum include: matrix method, graph method, method of
constructing tree of targets.
According to I. M. Morgunov [6], J. Johnson [7], the development of the structure
of the curriculum should be based on the theory of graphs and matrix methods. The
authors propose, first of all, to establish the logic of the learning, that is, the order of
studying the training modules, based on the graph of module’s connectivity. For each
content module, the time period for a possible study is determined. Then the modules
are distributed on weeks according to the chosen optimization criterion.
A. Moskvychenko [9] proposes to improve the structure of curricula with the help
of hierarchical trees for the purpose of training a specialist. Target Tree has several
hierarchical levels. Main goals of the training are graduate’s knowledge and abilities.
Each goal is put in accordance to one or more disciplines of the curriculum. Each dis-
cipline can be divided into modules. The educational process tree of goals has three
levels: the goals of the educational process; blocks of curriculum disciplines; thematic
modules of discipline. Thematic modules have equal amount of time.
The inputs are the coefficients of the relative importance of the goals for the educa-
tional process, as well as the importance of the second level objectives for the objectives
of the first level. Based on these data, the coefficients of the relative importance of the
goals of the second level, the importance of the goals of the third level for the purposes
of the second level and the ratios of the relative importance of the objectives of the third
level are calculated. The modules are placed by the reduction of group weights. It is
necessary to select in the curriculum V of the first modules, where V - the volume of
the curriculum in the modules. Then an expert survey is conducted to establish relation-
ships between the selected modules in the curriculum.
Such an algorithm does not take into account the relationships between the modules.
The relationships between the modules included in the curriculum are evaluated after
the content is selected. Therefore, in the process of designing a curriculum, there may
be a lack of information for students to master the teaching material of some modules,
since the modules required for them as an information base may have insufficiently
high group weight.
In the works of E. Herman [12], A. Ovchinnikova [11] algorithms of optimization
of curricula on the basis of the graph method are proposed, taking into account the
relationships between the content modules. In these works, optimization of the plan is
possible provided that the total volume of the modules of educational disciplines is
equal to the scope of the curriculum. And only the presence or absence of relationships
is noted.
In work O. Trofimova [10] highlighted the possibility of improving the curriculum
by several criteria, taking into account the density of the relationship between the mod-
ules and the significance of each module, but the algorithm built by the author does not
allow to take into account the dual vocational training of future engineers-teachers.
The peculiarity of curriculum designing in engineering and pedagogical higher edu-
cational institutions is associated with the direction of the professional training at two
sites of the activity: professional training and engineering activities. In the work [13] it
is stated that the training of specialists for engineering and educational activities should
take into account engineering and teaching components, moreover these components
must be interlinked and integrated, that is to make a dual system.
Thus there is a need to develop a qualitatively new approach to design a curriculum
for future engineers-teachers of computer profile built on the model of dual content of
engineers-teachers’ training.
2 Formalization Contents of Training
We will formalize the process of designing a curriculum for the training of engineers-
teachers of the computer profile. The curriculum is a set of disciplines that are interre-
lated and must be studied over a certain period of time. The structure of the educational
material of the discipline consists of smaller elements - modules.
ZD ZM 1, ZM 2 ,..., ZM j ,..., ZM L (1)
where ZD - content of the discipline;
ZMj - j- content module of discipline;
L - number of modules in the discipline.
The number of modules for each discipline varies and depends on the volume of
discipline.
Each module consists of single elements - educational elements. Educational ele-
ment is the minimum dose of educational information that preserves the properties of
the training object. Educational elements can be represented in the form: concept, phe-
nomenon, relation, and algorithm.
ZМ N1, N2 ,..., N j ,..., Nt (2)
where Ni – i-educational element of the module.
Since one educational element can belong to different content modules of different
disciplines, it is advisable to design the training of a specialist as a set of objects of the
lowest level - educational elements.
N N1, N2 ,..., N j ,..., N g (3)
where N - content of specialist training, Ni - educational element.
To structure the dual content of the professional training of engineer-teachers, we
will use the graph method [14]. The structural-logical scheme of professional training
of future engineers-educators should be represented in the form of a graph whose ver-
tices are educational disciplines, fig. 1
Fig. 1. Fragment of the content graph of the training
We will analyze the discipline for the presence or absence of mutual relationships be-
tween the teaching elements. Disciplines between which there is no connection we shall
isolate into layers [14]. A group of disciplines joined by a layer will compile a list of
disciplines for the academic semester.
Fig. 2. Fragment of the structural-logical scheme of training engineers-teachers of the computer
profile is divided by layers
In fig. 2 shows a fragment of a structural-logical scheme structured in layers. Thus, part
of the curriculum is depicted in Fig. 1. can be represented by five layers. Therefore,
these disciplines will be studied within five semesters.
The combination of the graph and set of educational elements form components for
content model of specialist’s training. This model allows to:
─ clearly define the content of educational material and learning goals;
─ to present the content in a clear and foreseeable manner;
─ involve experts in discussion about the completeness and target indicators at the in-
itial design stage;
─ ensure the logic of teaching;
─ go to computerization of presentation of the content model;
─ form a systematic notion of the content of educational material.
3 Mathematical Formulation of the Problem
The process of forming a training plan for future engineer educators should take into
account the high dynamics of the development of information and communication tech-
nology. This is possible due to the principles of modular selectivity and modular redun-
dancy. The principle of modular selectivity lies in the fact that in each discipline there
is a certain part of important modules, which must necessarily constitute the content of
vocational training, the so-called invariants of disciplines. The principle of modular
redundancy is the ability to form discipline with a large number of related modules.
Compliance with these principles allows you to adapt the curriculum to the require-
ments of the present and provide the most effective professional training.
Thus, when forming a curriculum we will consider two types of modules. Norma-
tive modules - modules that are required to study and fall into the curriculum. The
number of these modules should not exceed the maximum number of modules L for
this discipline. Variant modules - modules that may not fit into the curriculum.
The task of improving the curriculum is to select the most important for the pro-
fessional activity of the material and place it in semesters in an optimal way.
The modules of the curriculum are interconnected that is in the following ones the
material from the previously studied modules is used. If the module Mi uses information
from the module Mj, Mj is called an ancestor relative to Mi, Mi is called a descendant
relative to Mj.
Density of connection P (i, j) between the modules Mi and Mj may be character-
ized, considering which part of the module Mi is used in the material of module Mj.
P(i, j ) k1 *k2
(4)
where k1 – the number of study elements used in the module of Mj descendant of the
module Mi;
k2 – the number of these connections.
After evaluating all connections of all modules the density of communications is
normalized from 0 to 10. In determining the coefficient of module significance for study
of further material we will take into account not only the module contribution to the
study of its direct descendants, but also the later studied modules. For this purpose let’s
enumerate all the modules, giving one index to each. Let the number of modules be M.
Then the graph of connectivity can be provided by a two-dimensional matrix A of MxM
dimension, each element of which is a (i, j), which is equal to the density coefficient of
communication between modules i and j P (i, j).
Introduce the concept of iterative power of k order of m – pk(m) module. Iterative
power of m module of the first order characterizes the value of the contribution of a
module-ancestor to explore its descendants and is equal to the sum of weights of con-
nections that go from it.
M
p1(m) a(m,i)
i 1 (5)
where a(m,i) – coefficient of density connection of module-ancestor with the module-
descendant.
Iterative power of module of the second order describes its contribution to the study
of its descendants and the descendants of the second generation
M
p 2 (m) p1(m) p1 (i) P(m,i)
i 1 (6)
Usually just a few iterations to rank all the elements of the matrix are needed. After
the ranking of the elements ceases to change, one can finish the calculation. Note
through kc the significance coefficient of module for the further material study.
p max (m)
kc
M max
p (i)
i 1 (7)
The significance coefficient for the further study of the material – descendants
comes to the scale from 0 to 10.
Construction of the curriculum is based on the data obtained from experts. In the
training process of engineers-teachers of computer profile it is important to consider
the educational component of each engineering module and the engineering component
in each pedagogical module. Therefore, the task of experts is to determine for each
module the coefficient of significance for the future activity of computer systems engi-
neer and the coefficient of significance for the future activity of a teacher.
In this study, special attention was given to criteria weighting, considering the in-
fluence of weight assignment on the final result [15].
The coefficient of professional significance is found with the formula:
1 4
kпр kп kі
5 5 (8)
where kn– the average assessment of the module by experts-teachers;
ki – the average assessment of the module by experts-engineers.
For each module the coefficient of the total value is calculated.
By solving the task of optimizing the training curriculum, we will understand the
multicriterial task for maximizing the professional value and maximizing the connec-
tion [16, 17]. Let us reduce the multicriteria problem to a single criterion [18, 19].
k з 0.5 * k с 0.4 * k п 0.1* kі
(9)
The formation of the content in the form of a directed graph and a set of educational
elements that form the maximum value of the generalized significance will be consid-
ered to be the solution for the task of optimizing specialist’s training content.
4 Stages of Automation of the Design of the Curriculum of
the Disciplines of Professional Training
Method of optimization of the curriculum for future engineers-teachers after for the
criterion of maximizing the generalized significance considering the connectivity of
modules consists of the following ten steps:
1. Getting expert data. Linking educational elements (data that reflects the basic edu-
cational elements for each educational element that is necessary to study). Assess-
ment of pedagogical and engineering significance of modules.
2. Building the modules dependency graph.
3. Calculation and ranging the coefficient of density dependence between modules.
4. Removing contours. Finding the cycle in the graph and removing the edge, which
belongs to the cycle and has the lowest coefficient of density dependence between
modules.
5. Calculating the overall significance coefficient of the module.
6. Removing excessive modules from the graph. For each discipline compare the total
number of modules and the permissible number of modules, and if there are too
many modules, we remove the excessive ones, choosing the module with the lowest
coefficient of general significance.
7. Building the graph of dependence between disciplines.
8. Calculating and ranking the values of density dependence coefficients between dis-
ciplines.
9. Removing the contours from the graph.
10. Dividing the contours into layers.
The algorithm for optimization of the curriculum for future engineers-teachers af-
ter the criterion of maximizing the generalized significance is presented in Fig. 3
Fig. 3. An algorithm for optimization of the curriculum
Based on the described mathematical apparatus the information system of management
of curriculum designing in terms of dual training was designed and implemented.
Information system is a software tool based on the use of .NET Framework and
ASP.NET MVC. Microsoft SQL Server is used to store data. The MVC model is pre-
sented in fig. 4.
Fig. 4. MVC model
The Web service provides the following functionalities:
─ authorization of users and differentiation of the rights of their access to the possibil-
ities of the site;
─ management of discipline directory: the ability to create, edit and remove disci-
plines;
─ management of content modules directory: the ability to create, modify and delete
modules;
─ management of the Directory of Educational Elements: the ability to create, edit and
delete educational elements;
─ linking to the module of the list of learning elements studied within the framework
of this module;
─ linking to the learning elements of other learning elements, which are determined by
the basic elements for studying this learning element;
─ to make expert evaluation by engineering and pedagogical specialists of variational
content modules, which are part of disciplines with an excessive number of modules;
─ management of user lists: create, edit, delete users, assign roles to differentiate
rights;
─ processing expert data and forming a structural and logical scheme for the training;
─ the isolation of the layers of the structural-logical scheme of training;
─ constructing an image of a structurally-logical scheme for the training, broken down
into layers.
To work with the database, we chose .NET Entity Framework, namely the Code First
approach. This approach involves describing database models with the help of a pro-
gram code, and from it the database is generated. As a result, we received the database
depicted in Figure 5.
Fig. 5. Structure of the database
Now let's look at the basic classes of database models. Class "Discipline" contains
information about the discipline in fig. 6.
Disciplin
int Id
string Name
double NumberCredits
int CycleId
public virtual Cycle Cycle
public virtual ICollection Moduls
public virtual ICollection ApplicationUsers
public float CalcStandartModul()
public bool CanAddStandart()
public bool IsOverfull()
public static ValidationResult ValidateNameUnique(Disciplin disciplin)
public static ValidationResult ValidateCredits(Disciplin disciplin)
public static ValidationResult ValidateAllCredits(Disciplin disciplin)
Fig. 6. Diagram of the Discipline class
Field Id - stores a unique object index.
Field Name - keeps the name of the discipline.
Field NumberCredits corresponds to the number of credits assigned to the disci-
pline.
CycleId field - saves the unique key referencing to the table of cycles (Cycle).
Cycle property - the virtual field returns an object of the type Cycle (a training cycle)
whose key corresponds to CycleId.
Property Moduls - returns a collection of Modul type objects (modules) that refer to
this discipline.
CalcStandartModul () function returns the number of credits reserved by standard
modules.
CanAddStandart () function returns true values if it is possible to attach another nor-
mative module in the discipline, the value false otherwise.
IsOverfull () function returns the true value if the number of modules is greater than
the maximum value of the number of modules for this discipline, the value false other-
wise.
Functions ValidateNameUnique, ValidateCredits, ValidateAllCredits - check the
boundary values for the saved object to the database.
ValidateNameUnique - checks for the uniqueness of the name of the discipline.
ValidateCredits - checks that the number of credits is multiple of 0.5.
ValidateAllCredits - Does not allow you to keep discipline if the number of credits
for the course is over.
Depending on the user's role ("Developer", "Expert" ("Educator" or "Engineer"),
"Administrator"), "Web-site" has different functionality.
The functional capabilities of the system for the user in the role of "Curriculum de-
veloper" allow the creation and editing of disciplines, modules of teaching disciplines,
educational elements, as well as establishing links between educational elements. Based
on these connections, the system calculates the coefficient of connectivity between the
content modules.
The functional capabilities of the system for the user in the role of "Expert" ("The
teacher" or "Engineer") allow to view and edit any discipline, content module, training
elements and determine the professional importance of the content modules of the dis-
ciplines. This assessment is taken into account by the system to determine the duality
factor of the content module and to decide on the feasibility of including this content
module in the curriculum. Depending on the role of "Teacher" or "Engineer", the pro-
fessional pedagogical or professional engineering significance of the content module is
determined.
The system function for the user as Administrator allows you to add, edit and delete
users of the system. The administrator also has every opportunity to develop a curricu-
lum. However, he cannot make an expert assessment of content modules.
The "Structure of the curriculum" requires the presence of Microsoft Windows
Server operating system and Microsoft SQL Server database management systems.
Workstations must have Web browsers and Microsoft Excel installed.
Web interface for the main page of the "Structure of the curriculum" is depicted in
Fig. 7.
Fig. 7. Web-interface of the "Structure of the curriculum"
Inputs for the information system "Structure of the curriculum" are: a directory of dis-
ciplines, a directory of modules, a directory of educational elements, links between ed-
ucational elements, expert assessment of the professional significance of each module.
Fig. 8. The structural-logical scheme of training
Output - there are two documents: the first document - a graphical representation of
the structural-logical scheme of training (Fig. 8); The second document is the Excel
spreadsheet file, which contains a list of disciplines with semesters in which it is rec-
ommended to study these disciplines.
5 Conclusion
The structure of the curriculum for the training of specialists is proposed as a set of
structural-logical schemes and a set of educational elements. The filling of the devel-
oped structural-logical scheme by educational elements allows: to clearly define the
content of the educational material and the purpose of training; to present the contents
of the educational material in a visual and accessible form; to engage experts to discuss
the completeness of content and targets at the initial design stage; to provide a clear
logical sequence of educational disciplines; to use the automated presentation of the
structure of the curriculum; to form a system representation of the content of the edu-
cational material both from the developers and the teachers.
The method for improving the structure of the training curriculum by the criterion
of the general significance of educational disciplines content modules has been devel-
oped.
The informational system for improving the structure of the curriculum of training
according to the criterion of general significance has been developed. This system pro-
vides the possibility of distance work on the development of the structure of the curric-
ulum.
With the help of the information system, the structure of the curriculum of training
of the bachelor's degree in major "Professional Education. Computer Technology" is
improved.
References
1. Sosnitskiy, O., Kravchenko, N., et al.: The concept of reform of the system of national edu-
cation based on thinking. Information Technologies & Knowledge 6(3), 283–299 (2012).
2. Gholami M. H., Seyyed-Esfahani M.: An integrated framework for competitive market strat-
egy selection by using fuzzy AHP. Tehnicki vjesnik/Technical Gazette 1(4), 769-780
(2012).
3. Stoev L., et al.: Preparation of video lectures for distance education. Tehnički vjesnik:
znanstveno-stručni časopis tehničkih fakulteta Sveučilišta u Osijeku 23(3), 925-928 (2016).
4. Alekseeva, G. M.: Practical aspects of using of computer technology in the process of pro-
fessional preparation of students at educational institutions. Zbirnik naukovih prats (Ak-
tualni pitannya fIziko-matematichnoyi osviti) l.3, 139-145 (2014).
5. Hrynovets, M. V.: Information model of educational process management in conditions of
weak structuring. Informatsiyni systemy ta merezhi, 129-136 (2010).
6. Morhunov, Y. B.: Fundamentals of discrete optimization of some ordering tasks (on the ex-
ample of the educational process), Мoscow: Issledovatelskiy tsentr problem kachestva pod-
gotovki spetsialistov, 1994.
7. Parker, R. G., Rardin, R. L.: Discrete optimization, Elsevier, 2014.
8. Johnstone, J.: Mathematical Models Developed for Use in Educational Planning: A Review.
Review of Educational Research 44(2), 177–201 (1974).
9. Moskovychenko, A. L.: Tree of purpose engineering activities. Cybernetics and high school,
13, 123-129 (1987).
10. Trofimova, O.K.: Avtomatizatsija processa sostavlenija uchebnykh planov vuzov: Diss.
Kand. Tekhn. Nauk, Moscow (1999). (in Russian)
11. Ovchinnikov, A.A.; Puchinskiy, V.S.; Petrov, G.F.: Networking methods of planning and
organizing the educational process. Мoscow, Vyisshaya shkola, (1972)
12. German, E.I.: Development of models and algorithms for multi-objective optimization of
the educational process plans: thesis ... cand. tech. sciences, Tomsk, 1975.
13. Khomenko, V.: General scientific principles of developing a system of dual content of train-
ing of future engineers-teachers, problems of engineering and pedagogical education. Col-
lected Works. Ukrainian Engineering and Pedagogical Academy (UEPA), 42-43, (2014).
229 p.
14. Deo, N.: Graph theory with applications to engineering and computer science, Courier Do-
ver Publications, 2017.
15. Simunovic, K., Draganjac, T., Simunovic, G.: Application of different quantitative tech-
niques to inventory classification. Tehnicki Vjesnik - Technical Gazette 15(4), 41-47 (2008).
16. Eschenauer, H., Koski, J., Osyczka, A.: Multicriteria design optimization. Berlin, Spring-
errVerlag, (1990).
17. Mladineo, M.; Veža, I.: Optimization of the Selection of Competence Cells in Regional Pro-
duction Network. Tehnicki Vjesnik Technical Gazette 18(4), 581-488 (2011).
18. Ferreira, José C., et al.: Methodology to select solutions for multiobjective optimization
problems: Weighted stress function method. Journal of Multi‐Criteria Decision Analysis 24,
103-120 (2017).
19. Noghin, V. D.: Linear scalarization in multi-criterion optimization. Scientific and Technical
Information Processing 42(6), 463-469 (2015).