Solving Poverty using Ontology
Zarmeen Nasim and Imran Khan
Institute of Business Administration, Karachi, Pakistan
Keywords:
Ontology, Problem Solving.
Abstract:
This paper presents an ontology-based approach to address poverty. Poverty has been one of the serious
societal problems that the world is facing in recent times. The approach has been modeled in such a way that
it can be used to solve any other wicked problem of the society as well including corruption, bad governance,
traffic management, poor education and many others. The causes of the said problem were modeled at different
level of granularity. The proposed model also incorporates various different ways of addressing poverty by
addressing possible causes of poverty. The presented scheme of using an ontology to address the wicked
problem has the inferencing capability also to infer the indirect causes of poverty.
1 INTRODUCTION
The World Bank Organization describes poverty in
this way:
”Poverty is hunger. Poverty is lack of shelter.
Poverty is being sick and not being able to see a doc-
tor. Poverty is not having access to school and not
knowing how to read. Poverty is not having a job, is
fear for the future, living one day at a time.
Poverty has many faces, changing from place to
place and across time, and has been described in
many ways. Most often, poverty is a situation peo-
ple want to escape. So poverty is a call to action
for the poor and the wealthy alike a call to change
the world so that many more may have enough to eat,
adequate shelter, access to education and health, pro-
tection from violence, and a voice in what happens in
their communities.
The problem of Poverty is well known across the
globe. Every country is facing this issue. However,
the extent to which poverty prevails in the society is
different in different parts of the globe. There is no
one ultimate cause of poverty. The cause of poverty
varies from location to location. Living a poor life
in USA is different from the life of poverty in a third
world country such as Pakistan or Bangladesh.
We believe that in order to address the issue of
poverty, we need to model poverty as a system of
causes and possible solutions to these causes. This
paper describes our systematic approach based on On-
tology, developed to address the causes of poverty.
The causes which have been considered in this paper
are based on our observation of livelihood in Pakistan.
There are different measures to capture the
poverty in a country. The most commonly used mea-
sure is a Multidimensional Poverty Index (MPI). The
multidimensional poverty index is used to measure
the severe deprivations that the individual experiences
with respect to the living standard, education and
health services. Multiple variables are used to cap-
ture the quality of education, health and standard of
living in a country.
According to the survey report on Multidimen-
sional poverty published by the ministry of planning,
development and reform in Pakistan
1
, the multidi-
mensional poverty index of Pakistan is 0.197. The
MPI showed that the 38.8% of the population of Pak-
istan is poor. The statistics mentioned in the report
showed that the proportion of poor people living in ur-
ban areas is significantly lower than the ratio of poor
people living in rural areas of Pakistan. Education de-
privation in Pakistan made the greatest contribution
to national poverty, followed by living standards and
health services deprivation.
The main contributions of this research are as fol-
lows:
1. The development of an ontology aiming to model
wicked problems.
2. The instantiation of the developed ontology to
capture the global issue of Poverty.
1
http://www.pk.undp.org/content/pakistan/en/home/libr
ary/hiv aids/Multidimensional-Poverty-in-Pakistan.html
Nasim, Z. and Khan, I.
Solving Poverty using Ontology.
DOI: 10.5220/0006944102710278
In Proceedings of the 10th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2018) - Volume 2: KEOD, pages 271-278
ISBN: 978-989-758-330-8
Copyright © 2018 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
271
This paper presents the ontological solution to ad-
dress the issue of poverty. The ways presented in
this paper are further discussed in the Discussion sec-
tion using the leverage points proposed by Donella H.
Meadows (Meadows et al., 1997). Donella H. Mead-
ows discussed leverage points to intervene in a sys-
tem. Nine different leverage points were presented
which are as follows:
1. Numbers (Subsidies, Taxes, Stan-
dards):Numbers were considered as the least
important leverage point.
2. Material Stocks and Flows: The stocks and
flows effect the system operations.
3. Regulating Negative Feedback Loops: Negative
feedback loops are an important leverage point.
Systems’ performance can be improved by regu-
lating negative feedback loops.
4. Driving Positive Feedback Loops: Positive
feedback loops determine development and de-
structions in systems.
5. Information Flows: The information flow in the
system can be considered as another important
leverage point for the policy makers and company
strategist. It is important to identify the missing
information correctly at the right time.
6. The Rules of the System: Rules of the system
includes incentives and constraints applied to the
system. If someone wants to understand the mal-
functions of systems, then he should try to identify
where in the system some rule is broken.
7. The Power of Self-organization: Self-
organization refers to the change in any aspect of
a system lower on this list, such as addition or
removal of the new physical structure, adding or
deleting negative or positive loops or information
flows or rules.
8. The Goals of the System: The goal of the system
is an important leverage point. As everything on
this list is forced to conform to system goals.
9. The Mindset or Paradigm out of which
the Goals, Rules, Feedback Structure Arise:
Paradigms are the bases of systems. Goals, in-
formation flows, feedbacks, stocks, flows emerge
from the paradigm. People who manage to in-
tervene in systems at the level of paradigm hit a
leverage point that completely changes systems.
The rest of the paper is organized as follows. Sec-
tion 2 describes the related work. Section 3 defines
poverty as the wicked societal problem and the char-
acteristics of the wicked problem. Section 4 presents
the methodology. Section 5 includes the discussion
and finally, Section 6 concludes this paper.
2 RELATED WORK
The literature review section is subdivided into two
subsections. In the first subsection, we described the
application of ontology. In the second subsection, the
approaches proposed for addressing poverty are de-
scribed.
2.1 Ontology
The use of ontology to model complex systems
has been widely studied in literature. This sec-
tion presents the brief overview of the work done
on ontology by various researchers. L.C Van Rui-
jven (Van Ruijven, 2013) discussed the interoperabil-
ity issues faced in capital facilities industry due to
the highly distributed nature of such industries. It
was highlighted that the lack of proper communica-
tion and inadequate exchange of information within
projects were the main reason of project failure. An
ontology-based approach was proposed for managing
various industrial processes. The proposed ontology
is currently used in major infrastructure projects in
the Netherlands. The ontology was based on the sim-
plification of ISO 15926. The implementation of the
presented ontology was done using the information
management tool Relatics.
H.J Happel (Happel and Seedorf, 2006) discussed
various applications of ontologies in software engi-
neering. The paper presented a framework for clas-
sifying usage of ontologies in software engineering.
The concrete approaches for using ontology in the
software engineering life cycle was discussed. The
use of ontology in each phase of development cycle
improves the communication and information man-
agement. It enables developers to share a common
vocabulary for communication.
Eric C. Honour (Honour and Valerdi, 2006) dis-
cussed the need for ontology in designing systems.
An ongoing project Systems Engineering Return on
Investment (SE-ROI) aims to collect empirical infor-
mation to understand how systems engineering meth-
ods relate to program success (cost, schedule, and
technical terms). The purpose of the project is to con-
vert ontology into a methodology for measure return
on investment of software engineering projects. The
paper highlighted the fact that one of the greatest chal-
lenges in quantifying systems engineering is the lack
of a shared conceptualization. The challenge can be
overcome through the use of well-defined ontology
for software systems.
Jonathan DiLeo et al. (DiLeo et al., 2006) ex-
tended the approach developed earlier for analyzing,
designing, and developing multiagent systems, called
KEOD 2018 - 10th International Conference on Knowledge Engineering and Ontology Development
272
Multiagent Systems Engineering (MaSE). The pre-
sented approach extends MaSE to include the use
of ontologies for information domain specification.
The authors highlighted the fact that though many
methodologies have been developed for engineering
multi-agent systems, however, these approaches do
not sufficiently address the need of information do-
main system. The extensions allow the designer to
specify information flow by using objects from the
ontology as parameters in agent conversations. The
ontologies were used to specify the information do-
main of a multiagent system. The agents in the sys-
tem interact by passing messages and these messages
frequently involve passing parameters.
Humberto L. O. Dalpra et al. (Dalpra et al., 2015)
proposed an approach based on ontology for the soft-
ware process improvement. The paper discussed that
the process improvement can be achieved by analyz-
ing process data execution with an approach called
PROV-Process, which uses a layer for storing pro-
cess provenance and an ontology based on PROV-O.
The use of ontology enables software engineering to
detect the activities that influenced the generation of
other activities. Furthermore, we can identify agents
that could be associated with the solution of the de-
ployment task, considering that they already handled
the artifacts involved in this task in any other execu-
tion of the process.
Kohli et. al(Kohli et al., 2012) presented the on-
tology of slums for classifying images. They pro-
posed an ontological framework to analyze and clas-
sify slums. The ontology contained well-defined
terms that can be used to understand the relationship
between image and slum characteristics.
2.2 Poverty
Rodr
´
ıguez-Pose et al. (Rodr
´
ıguez-Pose and Hardy,
2015) addressed the problem of poverty using rural
development methodology. The political, social and
economic changes to rural areas impact the rural de-
velopment. Their research suggested that the agricul-
tural and non-agricultural development methods can
be used to alleviate the problem of poverty.
Yang Zhou et. al (Zhou et al., 2018) studied the
problem of poverty in China. The research studied
the impact of land policy innovations in alleviating
poverty from China. Another study on rural poverty
in China was conducted by Yansui Liu (Liu et al.,
2017). The study examined the spatial distribution
characteristics of poverty in rural China.
Abdul Hameed et. al (Hameed et al., 2016) pre-
sented their study on poverty in Pakistan. According
to this study, 59% population of Pakistan is poor. The
author suggested that the rural poverty can be elim-
inated through education, improved living standards,
and income growth.
Sajid Pervez (Pervez, 2016) described the role of
education in addressing poverty. Their results showed
that the education has a significant impact on improv-
ing the employment status of people which helps in
reducing poverty from the country.
Alvarez et. al (Alvarez et al., 2015) discussed
five major ways of poverty alleviation which includes
micro-finance, social entrepreneurship, foreign aid,
the base of the pyramid initiatives and the property
rights. Each of these measures was critically reviewed
and limitations were identified. Despite these ve ma-
jors, their paper concluded that international industri-
alization can have a far more promising impact on the
poverty alleviation.
3 POVERTY - WICKED
PROBLEM
Rittel et al. (Rittel and Webber, 1973) discussed cer-
tain characteristics of wicked problem. The paper dis-
cussed that most of the real-life problems are wicked
problems, unlike the problems we study in various
domains such as mathematics, physics and science
and engineering. Wicked problems are the problems
which are poorly defined. The paper explained vari-
ous characteristics of such problems which are as fol-
lows:
i The fundamental property of wicked problems is
that there is no complete formulation possible for
such problems. We are unable to propose a for-
mulation that covers all the variables or factors
impacting a wicked problem.
ii Wicked problems have no stopping rule because
there is no criteria for sufficiently understanding
the underlying problem. Usually, people stopped
working on such problems not because the work
gets completed but because either the resources
get exhausted or the solution obtained so far is
sufficient for the time being.
iii There are no right or wrong solutions to wicked
problems, rather we compare the quality of var-
ious different solutions of wicked problems with
each other to decide which one to adopt.
iv Another property of wicked problem is that there
is no immediate or ultimate test of a solution
to wicked problems. Once implemented, conse-
quences can be seen over the period of time.
v With wicked problems planning, every solution
which gets implemented have a consequence on
Solving Poverty using Ontology
273
the society. A cost is involved; therefore, a criti-
cal analysis of the solution and the possible con-
sequences is required before the solution gets im-
plemented.
vi Wicked problems don’t have a clearly defined list
of all possible solutions
vii Every wicked problem has a distinguishing char-
acteristic which implies that the solution to one
wicked problem cannot be applied to another
problem.
viii A wicked problem can be considered as a cause
of another wicked problem. We can consider
a wicked problem as a network of subproblems
which also possess the attributes of a wicked
problem.
ix Wicked problems can be explained using numer-
ous ways. The explanation determines the nature
of the proposed solution.
In the light of attributes of a wicked problem dis-
cussed above, the issue of poverty can be modeled as
a wicked problem. We are unable to formalize all the
social variables causing poverty. The reason of this
incomplete formalization is due to the cultural depen-
dency of poverty on various social attributes. This
leads to consider poverty as a wicked problem. More-
over, there can be various ways to address poverty.
However, there is no perfect solution so far to com-
plete eradicate poverty. We can model poverty as a
network of other subproblems. This will be discussed
in detail in the next section.
4 METHODOLOGY
This section describes the ontology-based approach
of modeling poverty. Though the approach discussed
here is implemented the problem of Poverty, however,
it can be adopted to address any wicked problem.
Figure 1 describes the root causes of poverty. It
was identified that the root causes also have various
sub-causes. Unemployment has various sub-causes
expanded on level 3 and level 4 as shown in Figure
2.
After the identification of various causes and sub-
causes of Poverty, ontology of Poverty was con-
structed. Protege was used to construct the ontology.
4.1 Classes
Following classes are defined in the proposed ontol-
ogy.
Figure 1: Level 1 and Level 2 causes of Poverty.
Figure 2: Causes network of Unemployment.
1. Problem: The class problem represents the con-
cept of any wicked problem.
2. Solution: The class solution represents the con-
cept of the solutions to a wicked problem.
4.2 Instances
4.2.1 Instances of Problem
Following instances are defined for class Problem in
the proposed ontology.
i Absence of Vocational Training Institutes: It
KEOD 2018 - 10th International Conference on Knowledge Engineering and Ontology Development
274
Figure 3: Ontology.
has been observed that there is a lack of voca-
tional training institutes in Pakistan. The lack of
vocational training institutes leads to lack of job
opportunities.
ii Economic Situation: The economic situation
refers to the economic growth of the country. Eco-
nomic growth of the country greatly influences
the poverty ratio. In the proposed ontology, we
considered economic situation as a direct cause
of poverty.
iii Export/Import Imbalance: The import/export
imbalance occurs when the ratio of export is sig-
nificantly less than the ratio of import. This is
considered as a sub-cause of economic situation.
If export to import ratio is good then the economic
growth will rise, which eventually help us in re-
ducing poverty.
iv Improper Hiring: Improper hiring mechanism
of teachers at primary/secondary level of educa-
tion degrades the quality of education. Due to
the low quality education, youth face difficulty
in ceasing potential job opportunities. This leads
to increase in unemployment ratio which rise the
poverty ratio in a country.
v Improper Planning of Human Resource: Un-
like developed countries, there is no proper plan-
ning department for managing large population
in under developed countries. Due to the un-
planned growth in population, poverty level rises
in a country.
vi Incapable Authorities: Due to the incapability
of authorities, the natural resources are not uti-
lized properly for the creation of new job oppor-
tunities. This causes increase in poverty.
vii Incapable Teachers: It has been observed that
the incapable teachers are one of the cause of sub
standard education in Pakistan.
viii Lack of Government Support: At the govern-
ment level, there is no support for the startups in
terms of tax relaxation and funding. This causes
lack of startups in Pakistan.
ix Lack of Job Opportunities: One of the major
cause of unemployment is the lack of job oppor-
tunities.
x Lack of Startups: Lack of job opportunities is
due to the lack of startups. The increase in num-
ber of startups will provide job opportunities to
the youth which will help in addressing unem-
ployment.
xi Large Population: The unplanned growth in
population is one of the major cause of poverty.
xii Law and Order Situation: Law and order situ-
ation has the direct impact on the business of the
country. If the situation is good, then investors
will be willing to invest in small businesses/ star-
tups.
xiii Poor planning of Natural Resources: Natural
resources are the wealth of any country. The
proper utilization of these resources can bring
prosperity to the country which ultimately helps
in reducing the poverty from the country.
xiv Poverty: Poverty is one of the major problem of
the third world countries such as Pakistan.
xv Substandard Education: Substandard education
refers to the poor quality of education being pro-
vided to the youth of the country. It is one of the
major cause of unemployment in the country.
xvi Unaffordable Education: The high quality ed-
ucation is too expensive for the common man of
the country.
xvii Unemployment: Unemployment is one of the di-
rect cause of poverty.
Solving Poverty using Ontology
275
4.2.2 Instances of Solution
Following instances are defined for class Solution in
the proposed ontology.
i Awareness: Making general public aware of the
new job opportunities, vocational and develop-
ment programs will help in addressing poverty.
ii Support of Donors: The problem of lack of star-
tups due to funding issues can be addressed if the
financial support can be attained from the donor
organizations.
iii Government Support: If government provides
support to small startups in terms of taxation then
this will encourage people to start small business
which will ultimately help in reducing unemploy-
ment.
iv Industries: To solve the problem of ex-
port/import imbalance, new industries need to be
established.
v Personality Development and Training: If we
have more capable teachers then the quality of
education can be improved. If teachers are pro-
vided with personality development and training
programs, they can grow professionally.
vi Proper Planning: Proper planning for utilizing
natural resources for the prosperity of the country
will solve the problem of poor planning of natural
resources.
vii Vocational Training: Vocational training insti-
tutes can help in developing skills in the youth so
that the can find job according to their skill set.
This will help in reducing unemployment in the
country.
4.3 Object Properties
To connect problems with solutions, following object
properties were defined in the proposed ontology.
1. Caused by: The domain of caused by is the in-
stance of problem and the range is also an instance
of problem. For e.g poverty is caused by unem-
ployment
2. Solved by: The domain of solved by is the in-
stance of problem and the range is an instance
of solution. For e.g lack of startup is solved by
donors support
3. Cause: This is the inverse of caused by property.
For e.g unemployment cause poverty
4.4 Data Properties
Data properties are the attributes of the classes. Fol-
lowing data properties were defined in the proposed
ontology.
1. Description: This defines the description of prob-
lem. The domain is the instance of problem and
range is the string.
2. Level: It is used to represent the level of the prob-
lem. The domain is the instance of problem and
range is the integer. For e.g Poverty is assigned
level 0 and all the primary causes of poverty are
assigned level 1.
Figure 3 presents proposed ontology of the poverty.
Figure 4: Reasoning through proposed Ontology.
4.5 Reasoning
Reasoning is one of the very powerful feature of on-
tology. Reasoning enabled us to understand the rela-
tionship between various causes of poverty at differ-
ent level of granularities. As described in Section IV
that the symptoms of wicked problem are related to
each other, with reasoning we were able to identify
these relationships.
Figure 4 shows that the proposed methodology in-
ferred that the cause Improper Hiring also causes Sub-
standard Education, Unemployment and Poverty. The
inferencing was based on the transitivity of cause and
caused by properties. The instance Unemployment is
the upper level Level1 cause of Poverty where as Sub-
standard Education was defined as the cause of Un-
employment. Further, Improper Hiring is related to
Substandard Education through cause property.
The reasoning engine inferred that since Sub-
standard Education cause Unemployment and Un-
employment cause Poverty therefore, the instance
Improper Hiring is related to Unemployment and
Poverty through cause relationship.
KEOD 2018 - 10th International Conference on Knowledge Engineering and Ontology Development
276
5 DISCUSSION
This section discussed the findings of our research
work as follows
1. The modeling of a wicked problem poverty help
us in understanding the various different causes
of the root problem and the solutions. Also we
were able to model the relationship between these
causes and solutions at different level of granular-
ities.
2. The ontology helps us in understanding the holis-
tic view of the problem. The reasoning capability
of the proposed ontology also help us in under-
standing how the causes are indirectly related to
each other. For example, causes with level num-
ber 3 such as Improper Hiring mechanism are in-
directly impacting the causes with level less than
3.
To the best of our knowledge, this is the first attempt
of modeling poverty using ontology as a modeling
tool. As discussed in the literature review section,
people have performed economic studies on the issue
of poverty.
The leverage points presented in Section I can be
used to evaluate the effectiveness of various different
solutions proposed in this paper for poverty. Below is
the description of each of the leverage point that can
be applied on the solutions proposed above.
1. Numbers: By increasing the number of industries
and vocational training institutes, we can address
unemployment which will ultimately help in re-
ducing the poverty.
2. Information Flows: This is the responsibility of
policy makers to collect population statistics re-
garding poverty such as ratio of population having
access to clean water, food, shelter and education.
3. The Goals of the System: In order to reduce
poverty in a systematic manner, short term and
long term goals can be established to evaluate the
performance of working bodies.
4. The Mindset: The mindset of people can be
improved by spreading awareness regarding the
poverty and the efforts made to eradicate it. Per-
sonality development and training of teachers can
also help in improving the quality of education.
By improving the quality of education, we can re-
duce poverty.
The leverage points discussed above will help in
improving the quality of proposed solutions and this
will ultimately help in the eradication of poverty from
the society.
6 CONCLUSION
This research discusses the ontology-based model of
poverty where we have shown the relationship be-
tween poverty and its causes. The causes were ar-
ranged at the different level of granularity. The pro-
posed ontology also incorporates the ways through
which we can address the issue poverty.
This paper presented the ontological approach to
model the problem of poverty. The proposed ap-
proach can help us in modeling any wicked problem.
The benefit of using ontology-based approach over
other traditional modeling approaches such as Uni-
fied Modeling Language(UML) and Entity relation-
ship diagram (ERD) is the reasoning capability of on-
tology. Through reasoning, we can infer useful in-
sights on the ontology-based model. Reasoning en-
abled us to understand the interrelationships between
various causes of poverty. This understanding can
help us to prioritize our solutions while addressing
these causes. Furthermore, the solutions proposed for
poverty alleviation were analyzed using the leverage
points proposed by Meadows (Meadows et al., 1997).
In the future, the work can be extended further to en-
rich the ontology with more details by incorporating
other relationships.
REFERENCES
Alvarez, S. A., Barney, J. B., and Newman, A. M. B. (2015).
The poverty problem and the industrialization solu-
tion. Asia Pacific Journal of Management, 32(1):23–
37.
Dalpra, H. L., Costa, G. C. B., Sirqueira, T. F., Braga, R. M.,
Campos, F., Werner, C. M. L., and David, J. M. N.
(2015). Using ontology and data provenance to im-
prove software processes. In ONTOBRAS.
DiLeo, J., Jacobs, T., and DeLoach, S. (2006). Integrat-
ing ontologies into multiagent systems engineering.
Technical report, Air Univ Maxwell AFB Al Center
for Aerospace Doctrine Research and Education.
Hameed, A., Karim, S., et al. (2016). Multidimensional
poverty mapping for rural pakistan.
Happel, H.-J. and Seedorf, S. (2006). Applications of on-
tologies in software engineering. In Proc. of Work-
shop on Sematic Web Enabled Software Engineer-
ing”(SWESE) on the ISWC, pages 5–9. Citeseer.
Honour, E. C. and Valerdi, R. (2006). Advancing an ontol-
ogy for systems engineering to allow consistent mea-
surement. Technical report.
Kohli, D., Sliuzas, R., Kerle, N., and Stein, A. (2012).
An ontology of slums for image-based classifica-
tion. Computers, Environment and Urban Systems,
36(2):154–163.
Solving Poverty using Ontology
277
Liu, Y., Liu, J., and Zhou, Y. (2017). Spatio-temporal pat-
terns of rural poverty in china and targeted poverty al-
leviation strategies. Journal of Rural Studies, 52:66–
75.
Meadows, D. et al. (1997). Places to intervene in a system.
Whole Earth, 91(1):78–84.
Noy, N. F., McGuinness, D. L., et al. (2001). Ontology
development 101: A guide to creating your first ontol-
ogy.
Pervez, S. (2016). Role of education in poverty elimina-
tion in pakistan with special reference of south pun-
jab. International Journal of Innovation and Applied
Studies, 17(1):70.
Rittel, H. W. and Webber, M. M. (1973). Dilemmas in a
general theory of planning. Policy sciences, 4(2):155–
169.
Rodr
´
ıguez-Pose, A. and Hardy, D. (2015). Addressing
poverty and inequality in the rural economy from a
global perspective. Applied Geography, 61:11–23.
Smith, J. (1998). The Book. The publishing company, Lon-
don, 2nd edition.
Van Ruijven, L. (2013). Ontology for systems engineering.
Procedia Computer Science, 16:383–392.
Zhou, Y., Guo, Y., Liu, Y., Wu, W., and Li, Y. (2018). Tar-
geted poverty alleviation and land policy innovation:
Some practice and policy implications from china.
Land Use Policy.
KEOD 2018 - 10th International Conference on Knowledge Engineering and Ontology Development
278