=Paper=
{{Paper
|id=Vol-2683/paper2
|storemode=property
|title=Toward the Methodology for Considering Mentality Properties in eGgovernment Problems
|pdfUrl=https://ceur-ws.org/Vol-2683/paper2.pdf
|volume=Vol-2683
|authors=Alexander Makarenko
}}
==Toward the Methodology for Considering Mentality Properties in eGgovernment Problems ==
https://ceur-ws.org/Vol-2683/paper2.pdf
Toward the Methodology for Considering Mentality
Properties in eGovernment Problems
Alexander Makarenko
Institute for Applied System Analysis
National Technical University of Ukraine Technologies
Kiev, Ukraine
makalex51@gmail.com
Abstract—A general framework for eGovernment is The structure of the paper is next. At section 1 we
considered. The results of system analysis of different propose the general scheme of eGovernment droving from
components of eGovernment are proposed. Also the the point of view proposed by author concepts. Some
background for considering and modeling of human properties detalization of such concepts is proposed at section 2.
of individuals is described. It is proposed also the models for Section 3 devotes for considering transformations in society
considering spreading and development of eGovernment in the and of eGovernment subsystem.
society. The approach allows forecasting the dynamics of
opinion formation, and leading to modeling of the behavior of
eGovernment participants. Our approach is based on the II. CENERAL FRAMEWORK
attempt to utilize the principles of associative memory from eGovernment is the society part. So it should be
neural networks. Also the models with internal mental considered in the general frames accepted for considering
structures structure of individuals are considered and results of society and social systems. Usually in general problems of
computer experiments are discussed. Different kinds of opinion large social systems three ‘pillars’ had been considered
evolution are discussed including punctuated equilibrium. (Figure 1)
Indexes for power distribution in eGovernment are proposed.
Further research problems just as recommendations for All such components (and restrictions on corresponding
practical implementations are proposed. recourses) also should be considered in eGovernment
problems. Remark that scientific community agrees that
Keywords—eGovernment, opinion formation, associative ‘ecology’ and ‘economy’ ‘pillars’ have more or less
memory, reputation, mental patterns, participants, evolutionary developed models. But ‘social’ ‘pillar’ has less adequate
approaches, cybersecurity models. So in discussion of general framework for
eGovernment we will concentrates on the methodologies for
‘social’ aspects. At first stage we will accept that the models
I. INTRODUCTION for ‘ecological’ and ’economical’ components will supply
Recently eGovernment became more and more common the forecasts for ‘social’ components environment. (This is
technologies for society tasks and for society only the approximation because ‘social’ pillar has impact on
transformations. But practical experience in eGovernment other). Following approach from [5, 6] we suppose at the
using is far ahead of theoretical foundations of eGovernment. first approximation that he social part of eGovernment
Before in the series of papers [1-4] we had proposed outline consists from N individuals with bonds between them. The
of the problems of eGovernment. For example we had individual posses own dynamics of some parameters of
considered the eGovernment from the point of view of social type.
system analysis [1]; some presumable methodologies for
eGovernment considering [2,3]; sustainability of society and
of eGovernment [4] ; general models of large social systems
[5,6]. But for deep understanding of eGovernment and
moreover for practical implementation of eGovernment
systems more elaborated concepts, models and
methodologies should be developed.
Thus in given paper we propose some approach for
accounting mental properties of eGovernment participants,
the ways of transformations and the number of related
properties, including investigation of system elasticity,
calculating power indexes, supply the security of the system
etc.
Fig.1. Three ‘pillars’ of social system
Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0)
We suppose that the ‘Social’ part of government also has example the Scales of projects may expand from local to the
the ‘technical’ part. ‘Technical’ part includes interfaces country or international level.
between participants of eGovernment and administrative
(electronic and classical) part. For example ‘technical’ part It had been stressed by many researchers including author
may include communication lines, computers, analytical and [1-4] that the eGovernment development require the
security centres personal interfaces etc. Administration may searching of optimal ways for design and financing of
include top-level leaders, decision-making departments, data eGovernment. Recently it is impossible with applications of
collection and processing departments, press centres and mathematical models and approaches. The models are
many others. Thus at first approximation the eGovernment necessary as for global problems (for example for sustainable
system may be represented by schemes on the Figures 2, 3. development) as for searching more local regional
Figures 2 corresponds to traditional arrangement of commercial projects and solutions. Of course a lot of
government. But the Figure 3 display the origin some new mathematical models exist for different components of
aspects of government which include the ‘electronic’ remarked above pillars of system (it may be the goals of
government. The essentially new elements are individuals separate papers). So here we will concentrate on the aspects
with access to servers (S) through communications lines and most closely related to eGovernment especially to the less
separate departments for decision- making. formalized (just theoretically).
Of course such pictures are oversimplified. So it is
possible to pose more detailed scheme which can help to
understand the structure and role of eGovernment in social
system. Remark that evidently hierarchical nature of
considered social systems. Such pictures may also help to
pose the tasks of investigation and design of eGovernment
systems of different level and scales.
Fig. 3. Scheme with ‘classical’ and ‘electronic‘ government
Namely below we will consider the components related
Fig.2. Simple scheme of ‘classical’ government
with ‘population’ and ‘government’ blocks from Figures 2,3.
Of course such presumable schemes also are some Remark that usually any of components of eGovernment
approximations for real system. For example because a lack include as ‘classical’ as ‘new’ component (‘new’ means
of place we doesn’t show explicitly infrastructures, related to ‘electronic’ part of eGovernment). The share of
organizations, forms and industry, cities and villages, social ‘new’ components may be evaluated by some formal
networks and many others. But just such schemes allows for procedures and indexes. The fracture F (%) of population
stress some components and aspects of eGovernment. Such which use the interfaces (external and through PC) of
pictures illustrate the different presumable scales of eGovernment may serves as one of the simple examples. The
eGovernment systems; non-homogeneous character of fracture FG (%) of government departments involved in
systems especially of population; hierarchy in systems; eGovernment may serves as second example. The part of
interrelations and interactions between subsystems. Probably power in given social system transferred to population
such pictures may help in classifications and ranking of through eGovernment is the third example. But just the task
eGovernment projects and necessary cost evaluation. For of such blocks modelling is very complex (but possible in
principle for all pillars and components). For describing one
presumable approach for general modelling here we will or liquidity participant, while the negative values mean that
concentrate mainly on human - related tasks. the participant j is either insider who work against the
information he has in order to hide himself, or a participant
III. SHORT DESCRIPTION OF ASSOCIATIVE MEMORY who is likely to be wrong in his judgment. The reputation
APPROACH FOR SOME SOCIAL PROBLEMS variables cij form a matrix
First of all we stress some problems related to population
participants at eGovernment: 1) formation of public opinion C = {cij }i, j =1,..., N ()
on some issue by electronic system; 2) voting on some
question through eGovernment; 3) expanding of
that we call the matrix of reputation. The approach cij
eGovernment system; 4) evaluation of power distribution
valuation will be discussed later at the end of this section.
between population and administration. Below we propose
for illustration the development of methodology the first As one of the basic characteristics of the system we
problem. Remark that in this paper we intend only to illustrate introduce the concept of a vector field of influence
the background of methodology on the base of simplest
examples.
sj
A. General ideas F = { f i }i =1,, N : f i = cij , cii = 0 ()
j Mj
We present here briefly the core idea of the approach and
the rough draft of the model that we are going to develop in
the research. The proposed model does not pretend to be full where fi means the integral influence of opinions of all other
and is intended only to demonstrate the basic ideas presented participants on i participant. The intuition behind this formula
here. is the following. The ratio si/Mj represents the opinion
intentions of participant j at the current step. It shows the
As the first example we consider the simplified problem number of opinion participant j is planning to support or
when all individual are involved in eGovernment system. reject as a percentage of what his actual power is. The
Lets all individuals pose personal opinion through electronic product cij×sj/Mj is the information about intentions of
networks and received some revised information through participant j filtered through the matrix of reputation. Thus,
networks. Remark that the type and volume of information is the sum (2) represents all the available to participant i
different. The first is the case of fully open process when all information about the actions of other participants, and since
individuals know the opinion of all involved participants. The it is filtered through the matrix of reputation, it is meaningful
second case is the backward distribution for all participants and trustworthy to him. We would like to note here, that all
only the integral results (for example average opinion – say the other information, participant i might have, is already
the percents of supporting individuals or the power of support incorporated in his initial intensions si.
of some issue).
Obviously, the best strategy for rational individual will be
In order to make easier understanding of the method and to adjust his own initial intentions to the filtered information
to simplify the initial formulas, we consider the idealized about others. Speaking formally, we say that every participant
society. The opinion development consists of discrete steps, is associated with the information utility function, which he is
at which the actual exchange of opinion take place. Within trying to maximize during the decision-making process. It is
each step we identify the sub steps, which describe the done by correlating the decision of individual i with the
dynamic bidding and asking or decision-making processes for corresponding value of the field of influence fi.
every individual. The society consists of N homogeneous
participants (in future developments the homogeneous Thus, we may formulate the evolution equation describing
assumption obviously should be removed). the opinion dynamics (of course it is the simplest possible
example of dynamics):
With every participant we associate the state variable
siS={0,1,2,…,Mi}, where si represents the number of
shares that participant i is planning to strength (if si>0) or to si (t + 1) =
weak (if si<0) opinion, and Mi is the maximum allowed si + 1, if f i (t ) 0 and si (t ) M i ,
()
volume, which represents the power of opinion of participant = si − 1, if f i (t ) 0 and si (t ) − M i ,
iis able to accept. si
otherwise .
With every pair of participants i and j we associate the
variable cijR – the integral value of reputation that The initial conditions for this dynamic equation are the
participant j has from the point of view of participant i. This intentions of each individual to support opinion at the
value measures the degree of how well informed; participant j beginning of the opinion forming step. They are formed under
is in the eyes of the participant i. The large positive values of the influence of the sources outside the system, and represent
cij mean that, in the opinion of participant i, participant j is an the participant’s forecast of how well the particular opinion
informed (news, insider) participant, the values close to zero distribution will be doing.
can mean that the participant j is an uninformed (noise, nice)
Given the initial conditions for si and known values of The reputation matrix in the described above model
influence matrix, we may calculate the dynamics of the remains invariable during the supporting/rejection or
opinion patterns. Such dynamics is expected to be beneficial decision-making steps. Obviously, it should change at each
for each participant, since it leads to the maximal utilization evolution step, since participants analyze their own
of the filtered, and therefore useful, information available to performance as well as the performance of other participants
him. and society as a whole. Therefore, each individual might
assign different coefficients to the corresponding elements of
Obviously, the system consists of protagonists with the matrix of reputation, which will be enforced at the next
different and frequently antagonistic goals. Thus, the actions evolution step.
beneficial for a particular participant do not necessarily
benefit the others. Moreover, each participant acts from his Thus, the reputation matrix plays one of the major roles in
own interests and generally, if somebody wins, someone the proposed model, and the applicability of the model
loses. However, all these egoistic individuals comprise the depends, to a great extent, on the correctness and accuracy of
system we consider. Therefore, from the system point of view the reputation coefficients. The numeric values for the entries
the question is, whether the defined above dynamics of every of the matrix of reputation are not readily available. However,
participant leads to a meaningful evolution of the whole one of the advantages of the given approach is that it uses
system, or is this just a disordered, chaotic motion? The already proved and experimentally tested algorithms for the
answer can be found using the analogy with the physical identification of the matrix C via the prior observations of the
systems. opinion patterns. This algorithm has the form of the well-
known rule from the pattern recognition theory of associative
As the variable summarizing the evolution of the system, memory models [7]. Its brief idea can be outlined as follows.
we introduce the concept of ‘energy’E, which characterizes
the impact all the participants have had on each other in Suppose we have recorded information about opinion
making their supporting/rejection decisions: patterns Zk, k=1,…,K, where Zk={si} at the time moment k, K
is the number of observations, i=1,…,N, N – number of
participants. Then the matrix of reputation C can be evaluated
E = − f i s i as
i
Thus, at any given point in time, ‘energy’ E characterizes sik s jk
the state of the society. Naturally, we are interested in the C = {cij }, cij = , cii = 0 ()
M M
evolution of the opinion patterns leading to a state that has the k i j
property of stability. By analogy with the physical systems,
we will call the state of the system stable if the ‘energy’ E has Of course such model correspond more to the case of
a local minimum in this point. As we will see, the system will opinion formation in parliaments, administrative councils,
tend to minimize its energy during the evolution process. To and cyberspace networks. But a lot of improvements of model
show this, we will first formulate and prove the following can be proposed. Here we describe some of most evident.
statement. Anyway more realistic is situation that only F(%) of
Statement 1. Under the law of evolution (3) the system population is involved in egovernance processes. Then the
evolves to a local minimum of energy E. frames of the model are the same but for all population only
opinions of Ne e-participants are known. This allows further
After energy reaches the local minimum, due to (A1) any developments. At first the opinion of this Ne participants
change of the state of the system will increase the energy, serves as the information for other part on society by mass-
which is impossible because of (A2). Thus, si(t+1)=si(t), i, media, social relations etc. Such information serves also as
and the system will retain its stable state until some external some kind of social questionnaires (with the same difficulties
forces are applied. Such stable state can be thought as and problems). As such the date of e-participants opinion may
equilibrium, at which opinion pattern takes place. It simply serve as the database for other models and approaches. At
means that all the participants have reached their decisions second the changes in reputations C={cij} can be introduced.
having maximized their own information utility functions. Such changes in reputations may have different reasons –
Since we are assuming that all the external information the internal and external. Internal changes have internal process
participants might have is represented by their initial of evolution as the source. External changes may have the
intentions, evolution occurs. Thus, maximization of mass-media influence, straggle of political parties, and
individuals’ information utility functions leads to the education system as the main reasons. Remark that special
minimum of energy of the system and, therefore, to its dynamical equations may be derived for evolution of C={cij}
coordinated movement during the decision-making step. during time flow [7].
The next evolution step begins with the new initial Presumable variety of matrix of reputation properties may
conditions, which contain the new information participants follow to a lot of different effects (which we cannot describe
have been able to obtain. here because the lack of space). We only remark here the
possibility of periodic solutions for slightly non-symmetrical
matrix of reputation and chaotic behaviour of public opinion
in the case of sufficiently non-symmetric reputation matrix. а) Multiplicative
Also the abrupt transition between quasi-stable stats of
opinion during time in case of non-constant matrix of vi (t ) = ( Di (t )) g i (t ) ()
reputation C={cij}.
−k Di (t )
B. Accounting the internal structures of eGoverment where for example ( Di (t )) = e . In simplest evident
participants variant we may take:
The next step in development of proposed models is to
account the internal structure of participants (we named such N
participants as ‘intellectual’).
Di (t ) = S ij (t ) − S Rj (t ) ()
Let us consider the idealized market as the collection of N j =1
intellectual participants. We will consider the process with
discrete time steps. Each participant should to do decision
b) Additive vi (t ) = g i (t ) + f i ( Di (t )) , where
(change of state) at each time step in dependence of all
participants’ states. f i ( Di (t )) – some influence function. The simplest
Participant’s state is described by the variable example is:
Si(t)S={0,1,2,…,Mi}, which corresponds to the amount
of the recourse (opinion, information, materials and so on),
R (S j S j )
N R i
which may be gain ( if Si(t) < 0) or collect (if Si(t) > 0) by i f ( Di (t )) = Cij ()
individual (participant). Here Mi is the maximal volume of its j =1 Mj
resource (its potential). Interaction of individuals in
organization is described by influence matrix C={cij},
In this model vector vi(t) represent the understanding by i
j=1,…,N, cij[0,1] where cij – influence coefficient of j
participant on the tendencies in system: If vi(t) > 0, then the
individual on i. The influence matrix C may reflect the
authority power in organization. In simplest model we take tendency is to increase the recourse, if vi(t) 0, then the
Cij=0, i=1,…,N. stability is the main tendency, if vi(t) < 0, then the tendency is
to reduce the resources.
So the collection QR(t)=({SRl(t)},{CRlj}), i,j=1,…,N
represents the real state at moment t. Let us consider also One of the most usable forms of activation function F in
Qi(t)=({Sil(t)},{Cilj}), i,j,l=1,…,N as ideal pattern of situation such type models are:
from the i participant point of view. Then we can calculate
the difference between real and ideal patterns of situation:
S iR (t + 1) =
G (t ) S iR
Di (t ) = Q i (t ) − Q R (t ) () S R (t ) + 1 if v (t ) and S iR (t ) M i ,
i i
Mi
()
R G (t ) S iR
We suppose that the dynamics of i participant depends on ( ) − 1 if ( ) and S iR (t ) − M i ,
the difference Di(t) and on the mean influence field by other i
S t v i t
M i
participants. We accept the influence field G(t)={gi(t)}, I 0 othervise ,
=1,…,N as:
N S Rj (t ) where
g i (t ) = C ijR ()
j =1 Mj
N
The term SRj(t)/Mj in (6) corresponds to the activity of j gi2 (t )
participant at the moment t. The term CRij(SRj(t)/Mj) i =1
G (t ) = ()
corresponds to activity with reputation accounting. In general N
case the dynamical law for participant takes the form (F some
law for participant’s reaction, named frequently activation Remark that very interesting development of proposed
function): models consist in introduction time dependence of
connections by some dynamical laws. The models described
here correspond to the constant bonds.
SiR (t + 1) = F (vi (t )) ()
where the argument vi(t) may takes the form:
IV. RESERCH TASKS AND PROBLEMS TO BE SOLVED multi-valued solution existing in case of individuals which
Proposed approach allows developing the software and can anticipate the future [8].
trying to understand some properties of society and
particularly eGovernment. Here we describe some examples V. CONCLUSION
of computer experiments with the models (5)–(12) which Thus in proposed paper we consider the approach for
accounting the internal structure of participants and non- system analysis and modeling which implement some
constant in time reputation of participants (Figure 4). properties of real society and eGovernment. The main
distinctive features are the accounting of internal properties of
participants. As the authors envisage, the modeling principles,
described in section 3 can lead to the formulation and solution
of the following problems:
1. Development of models of opinion patterns for the
specific real problems.
2. Investigation of the control and security problems of
eGovernment on the base of proposed approach.
3. Introducing and investigation different indexes of
eGovernment operating, especially of power of e-participants
community.
4. Numerical simulation of specific local eGovernment
problems.
5. Analysis of the eGovernment spreading in society
on the base of proposed methodology.
6. Forming proposition for building general tasks
computing systems of investigation and managing
eGovernment with accounting all aspects remarked above.
7. Proposed approach allows re-formulate the
Fig. 4. Example of opinion formation modeling
problems of cyber security of networks and more generally
The horizontal axe corresponds to the steps of evolution security of society.
of opinion formation. The vertical axe represents the
REFERENCES
intentions of different participants. The left picture
correspond to stabilization of intentions of participants. The
right-side picture corresponds to the case of society with [1] A. Makarenko System Analysis, Foresees and Management of E-
Services Impacts on Informational Societies. Proc. 4th Eastern Europ.
changeable reputations during evolution. eGov Days, Prague, Czech Republic. 2006. 6 p.
The right picture illustrates the possibilities of oscillations [2] A. Makarenko Toward the building some methodic of understanding
of the opinion. The oscillations are intrinsic for society with and improvement of e-Government Proc. 6th Eastern Europ. eGov
Days, Prague, Czech Republic 2008. 5 p.
asymmetrical reputation of participants. Moreover the society
with mostly asymmetrically informed participants may have [3] A. Makarenko, B. Goldengorin, D. Krushinskiy, N. Smelianec
Modeling of Large-Scale crowd’s traffic for e_Government and
chaotic behavior. Other very interesting phenomenon is the decision-making. Proc. 5th Eastern Europ. eGov Days, Prague, Czech
possibilities of sudden changes of stable opinion patterns in Republic. 2007. p. 5
the case of variable reputation of participants. It may [4] A. Makarenko, A., E. Samorodov, Z. Klestova, Sustainable
correspond to real phenomena in the society. Also it may Development and eGovernment. Sustainability of What, Why and
correlate with phenomena of punctuated equilibrium in How. Proc. 8th Eastern Europ. eGov Days, Prague, Czech Republic.
biology. 2010. p. 5
[5] A. Makarenko New Neuronet Models of Global Socio- Economical
Of course till now our computational investigations are Processes. In 'Gaming /Simulation for Policy Development and
model with artificial date and further investigations will be Organisational Change' (J.Geurts, C.Joldersma, E.Roelofs eds),
Tillburg Univ. Press. 1998. P. 133–138,
interesting. But just now some prospective issues may be
discussed. [6] A. Makarenko, Sustainable Development and Risk Evaluation:
Challenges and Possible new Methodologies, In. Risk Science and
First of all proposed internal representation may be Sustainability: Science for Reduction of Risk and Sustainable
Development of Society, eds. T.Beer, A.Izmail- Zade, Kluwer AP,
considered as some correlate to ontology of participant. Also Dordrecht.2003. P. 87–100.
it may be interesting for considering classical problem of
[7] S. Haykin Neural Networks: Comprehensive Fundations. — N.Y.:
reputation. At second the approach reminiscent usual multi- MacMillan.. 1994. 697 p.
agent approach. The description of participant remember [8] A. Makarenko Anticipatory participants, scenarios approach in
participant with special representation of the internal and decision- making and some quantum – mechanical analogies. Int. J. of
external worlds by network structure. Also the prospective Computing Anticipatory Systems. 2004. Vol. 15. P. 217–225.
feature in the approach is the associative memory in proposed
models. Remark that recently we had found the possibility of