This document discusses using the Dempster-Shafer theory to detect poultry diseases. It uses five major symptoms - depression, combs, wattle, bluish face region, swollen face region, narrowness of eyes, and balance disorders. It calculates the basic probability assignment for each symptom and condition. Through combining the probabilities, it determines that avian influenza has the highest probability based on the reported symptoms. The implementation demonstrates the consultation process and identifies avian influenza as the likely disease.
Poultry Disease Detection using Dempster-Shafer Theory
1. Poultry Diseases Detection using
Dempster-Shafer Theory
Andino Maseleno, Md. Mahmud Hasan
Department of Computer Science, Faculty of Science, Universiti Brunei Darussalam
Jalan Tungku Link, Gadong BE 1410, Negara Brunei Darussalam
E-mail: andinomaseleno@yahoo.com, mahmud.hasan@ubd.edu.bn
ABSTRACT
Based on Cumulative Number of Confirmed Human Cases of Avian Influenza Reported to World
Health Organization (WHO) in the 2011 from 15 countries, Indonesia has the largest number death
because Avian Influenza which 146 deaths. In this research, we are using Dempster-Shafer Theory
for detecting poultry diseases and displaying the result of detection process. In this paper we use
five symptoms as major symptoms which include depression, combs, wattle, bluish face region,
swollen face region, narrowness of eyes, and balance disorders. Dempster-Shafer theory to
quantify the degree of belief, our approach uses Dempster-Shafer theory to combine beliefs under
conditions of uncertainty and ignorance, and allows quantitative measurement of the belief and
plausibility in our identification result. The result reveal that Dempster-Shafer theory has
successfully identified the existence of poultry diseases.
Keywords: poultry diseases; detection; Dempster-Shafer theory
INTRODUCTION rate in developing countries compared with
The demand for chicken meat has been more developed countries [3].
increasing because it has become cheaper The remainder is organized as follows. The
relative to other meats The term poultry refers Dempster-Shafer theory is briefly reviewed in
to domesticated fowl raised for meat or eggs. Section 2. The experimental results are
The poultry industry is dominated by the presented in Section 3, and final remarks are
chicken companies, development of poultry concluded in Section 4.
population and poultry industry is very rapidly
threatened by the presence of poultry disease. DEMPSTER-SHAFER THEORY
Disease is defined as a departure from health, The Dempster-Shafer theory was first
and includes any condition that impairs normal introduced by Dempster [4] and then extended
body functions. Disease results from a by shafer [5], but the kind of reasoning the
combination of indirect causes that reduce theory uses can be found as far back as the
resistance or predispose an animal to catching seventeenth century. This theory is actually an
a disease, as well as the direct causes that extension to classic probabilistic uncertainty
produce the disease. modeling. Whereas the Bayesian theory
Avian influenza virus, which has been requires probabilities for each question of
limited to poultry, now has spread to migrating interest, belief functions allow us to base
birds and has emerged in mammals and degrees of belief for on question on
among the human population. It presents a probabilities for a related question.
distinct threat of a pandemic for which the The consultation process begins with
World Health Organization and other selecting symptoms. If there are symptoms
organizations are making preparations [1]. In then will calculate, The Dempster-Shafer
2005, the World Health Assembly urged its theory provides a rule to combine evidences
Member States to develop national from independent observers and into a single
preparedness plans for pandemic influenza [2]. and more informative hint. Evidence theory is
Developing countries face particular planning based on belief function and plausible
and other challenges with pandemic reasoning. First of all, we must define a frame
preparedness as there may be a higher death of discernment, indicated by the sign Θ . The
Θ
sign 2 indicates the set composed of all the
2. subset generated by the frame of discernment. (3)
For a hypothesis set, denoted by A,
m(A)→[0,1]. Where
m( )=0
(4)
(1)
∅ is the sign of an empty set. The function m is
the basic probability assignment. Dempster's Flowchart of poultry diseases detection shown
rule of combination combines two independent in Figure 1. Table 1 shown basic probability
sets of mass assignments. assignment of symptom from each condition.
(∅) = 0 (2)
Table 1. Basic probability assignment of symptom
A. Symptom 1
Start
Depression is a symptom of Avian
Influenza (AI), Newcastle Disease (ND), Fowl
Cholera (FC), Infectious Bronchitis respiratory
Yes
form (IBRespi), Infectious Bronchitis
Input
Symptoms m1 (B) m2 (C) reproduction form (IBRepro), and Swollen
Head Syndrome (SHS). The measures of
uncertainty, taken collectively are known in
No Dempster Shafer Theory terminology as a
``basic probability assignment'' (bpa). Hence
m (A) = 1 we have a bpa, say m 1 of 0.7 given to the focal
element {AI, ND, FC, IBRespi, IBRepro, SHS}
in example, m1({AI, ND, FC, IBRespi, IBRepro,
SHS}) = 0.7, since we know nothing about the
End
remaining probability it is allocated to the
whole of the frame of the discernment in
Fig 1. Flowchart of Poultry Diseases Detection example, m1({AI, ND, FC, IBRespi, IBRepro,
SHS}) = 0.3, so:
The following will be shown implementation m1{AI, ND, FC, IBRespi, IBRepro, SHS} = 0.7
of the Dempster-Shafer theory to diagnose m1{Θ} = 1 - 0.7 = 0.3
poultry diseases.
Symptoms: B. Symptom 2
1. Depression Combs, wattle, bluish face region are
2. Combs, wattle, bluish face region symptoms of Avian Influenza with a bpa of 0.9,
3. Swollen face region so that:
4. Narrowness of eyes m2{AI} = 0.9
5. Balance disorders m2 {Θ} = 1 – 0.9 = 0.1
We combining symptom 1, and symptom 2
as seen in the table 2.
3. Table 2. Combination of symptom 1 and symptom 2 0.01 0.01
m7 (SHS) = 0.17
1 (0.81 0.07)
0.09
m7 (AI) = 0.75
1 (0.81 0.07)
0.63 0.27 0.01
m3 (AI) = 0.9 m7 (AI, ND, FC) = 0.08
1 0 1 (0.81 0.07)
m3 (AI, ND, FC, IBRespi, IBRepro, SHS) = m7 (AI, ND, FC, IBRespi, IBRepro, SHS) =
0.07
0.07 0.001
1 0 0.01
1 (0.81 0.07)
0.03
m3 (Θ) = 0.03 0.001
1 0 m7 (Θ) = 0.01
1 (0.81 0.07)
C. Symptom 3
Swollen face region is a symptom of Avian E. Symptom 5
Influenza, Newcastle Disease, Fowl Cholera Balance disorders is a symptom of
with a bpa of 0.8, so that Newcastle Diseases and Swollen Head
m4 {AI, ND, FC} = 0.8 Syndrome with a bpa of 0.6, so that:
m4 (Θ) = 1 – 0.8= 0.2 m8 {ND,SHS} = 0.6
We combining symptom 1, symptom 2, and m8 {Θ} = 1 - 0.6 = 0.4
symptom 3 as seen in the table 3. We combining symptom 1, symptom 2,
symptom 3, symptom 4 and symptom 5 as
Table 3. Combination of symptom 1, symptom 2, seen in the table 5.
and symptom 3
Table 5. Combination of Symptom 1, Symptom 2,
Symptom 3, Symptom 4 and Symptom 5
0.72 0.18
m5 (AI) = 0.9
1 0
0.06 0.02
m5 (AI, ND, FC) = 0.08 m9 (SHS) =0.1 0.07 0.31
1 0 1 0.45
m5 (AI, ND, FC, IBRespi, IBRepro, SHS) = 0.3
m9 (AI) = 0.55
0.01 1 0.45
0.01
1 0 m9 (ND) =
0.05
0.1
0.01 1 0.45
m5 (Ө) = 0.01 m9 (ND, SHS) =0.01 0.01 0.04
1 0 1 0.45
m9 (AI, ND, FC) = 0.03 0.05
D. Symptom 4 1 0.45
0.004
Narrowness of eyes is a symptom of m9 (AI, ND, FC, IBRespi, IBRepro, SHS) = 0.01
1 0.45
Swollen Head Syndrome with a bpa of 0.9, so
that: m9 (Θ) = 0.004 0.01
1 0.45
m6 (SHS) = 0.9
m6 (Θ) = 1 – 0.9 = 0.1
We combining symptom 1, symptom 2,
symptom 3, and symptom 4 as seen in the The most highly bpa value is the m9 (AI)
table 4. that is equal to 0.555 which means the
possibility of a temporary diseases with
Table 4. Combination of symptom 1, symptom 2, symptoms of depression, comb, wattle, bluish
symptom 3 and symptom 4 face region, swollen region face, narrowness
of eyes, and balance disorders is the Avian
influenza. Table 6 and figure 2 are shown final
result of the basic probability assignment and
graphic of identification from each condition.
4. Table 6. Final result of basic probability assignment
Fig 2. Poultry diseases in each condition
Figure 2 shows the highest basic consultation result can be seen in figure 3
probability assignment and the possibility of a and figure 4.
temporary diseases for each condition. The
highest bpa value for condition 1 and condition
2 are Avian Influenza diseases, condition 3
and condition 4 are Swollen Head Syndrome
diseases, condition 1 is Avian Influenza,
Newcastle Disease, Fowl Cholera, Infectious
Bronchitis respiratory form, Infectious
Bronchitis reproduction form, and Swollen
Head Syndrome. Condition 1, condition 2,
condition 4, and condition 5 need serious
treatment because has basic probability
assignment more than 0.5.
IMPLEMENTATION
The following will be shown the
working process of system in diagnosing a
case. The consultation process begins with Fig 3. Symptoms Selection
selecting the location and symptoms found
on the list of symptoms. In the cases tested,
a known symptom is depression, combs,
wattle, bluish face region, swollen face
region, narrowness of eyes, balance
disorders. The consultation process and
5. [2]. W. H. Assembly. Strengthening
pandemic-influenza preparedness and
response. Resolution WHA58.5, 2005.
[3]. E. Azziz-Baumgartner, N. Smith, R.
González-Alvarez, National pandemic
influenza preparedness planning.
Influenza and Other Respiratory Viruses
(2009) 189 – 196
[4]. A. P. Dempster, A Generalization of
Bayesian inference, Journal of the Royal
Statistical Society, 1968) 205 – 247
[5]. G. Shafer, A Mathematical Theory of
Evidence (Princeton University Press,
New Jersey, 1976)
Fig 9. Consultation Result
CONCLUSION
Detection of poultry diseases can be
performed using Dempster-Shafer Theory. In
this paper we use five symptoms as major
symptoms which include depression, combs,
wattle, bluish face region, swollen face region,
narrowness of eyes, and balance disorders.
The simplest possible method for using
probabilities to quantify the uncertainty in a
database is that of attaching a probability to
every member of a relation, and to use these
values to provide the probability that a
particular value is the correct answer to a
particular query. An expert in providing
knowledge is uncertain in the form of rules with
the possibility, the rules are probability value.
The knowledge is uncertain in the collection of
basic events can be directly used to draw
conclusions in simple cases, however, in many
cases the various events associated with each
other. Reasoning under uncertainty that used
some of mathematical expressions, gave them
a different interpretation: each piece of
evidence (finding) may support a subset
containing several hypotheses. This is a
generalization of the pure probabilistic
framework in which every finding corresponds
to a value of a variable (a single hypothesis).
The result reveal that Dempster-Shafer theory
has successfully identified the existence of
poultry diseases.
REFERENCES
[1]. L. D Sims, J. Domenech, C. Benigno, S.
Kahn, A. Kamata, J. Lubrouth, V. Martin,
and P. Roeder, Origin and evolution of
highly pathogenic H5N1 avian influenza in
Asia. The Veterinary Record (2005) 159
– 164.