Poultry Diseases Expert System using Dempster-Shafer                             Theory                                   ...
Step 1                      Representation of uncertainty   Alternative Rule                        on the set of basic ev...
Start                                       Yes               Input                         Symptoms                    m1...
Knowledge Acquisition                         Knowledge Base                         Inference Engine                User ...
Table 1: Knowledge Base No.    Symptom                   Disease                         Basic Probability                ...
Table 2: Combination of Symptom 1 and Symptom 2                                              {AI}       0.9               ...
0.01  0.01m7 (SHS) =                       0.17             1  (0.81  0.07)                 0.09m7 (AI) =             ...
Table 6. Final Result No.                         Diseases                                Basic Probability               ...
5. IMPLEMENTATIONThe following will be shown the working process of expert system in diagnosing a case. Theconsultation pr...
6. CONCLUSIONIdentification of poultry diseases can be performed using Dempster-Shafer Theory. In thispaper we describe fi...
[17]. R. Yager, “Reasoning with Uncertainty for Expert Systems,” International Joint      Conference on Artificial Intelli...
Upcoming SlideShare
Loading in …5
×

Poultry Diseases Expert System using Dempster-Shafer Theory

1,378 views
1,302 views

Published on

Based on World Health Organization (WHO) fact sheet in the 2011, outbreaks of poultry diseases especially Avian Influenza in poultry may raise global public health concerns due to their effect on poultry populations, their potential to cause serious disease in people, and their pandemic potential. In this research, we built a Poultry Diseases Expert System using Dempster-Shafer Theory. In this Poultry Diseases Expert System We describe five symptoms which include depression, combs, wattle, bluish face region, swollen face region, narrowness of eyes, and balance disorders. The result of the research is that Poultry Diseases Expert System has been successfully identifying poultry diseases.

Published in: Business, Technology
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
1,378
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
27
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Poultry Diseases Expert System using Dempster-Shafer Theory

  1. 1. Poultry Diseases Expert System using Dempster-Shafer Theory 1 2 Andino Maseleno and Md. Mahmud Hasan 1,2 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 AbstractBased on World Health Organization (WHO) fact sheet in the 2011, outbreaks of poultrydiseases especially Avian Influenza in poultry may raise global public health concerns due totheir effect on poultry populations, their potential to cause serious disease in people, and theirpandemic potential. In this research, we built a Poultry Diseases Expert System usingDempster-Shafer Theory. In this Poultry Diseases Expert System We describe five symptomswhich include depression, combs, wattle, bluish face region, swollen face region, narrownessof eyes, and balance disorders. The result of the research is that Poultry Diseases ExpertSystem has been successfully identifying poultry diseases.Keywords: Poultry Diseases, Expert System, Dempster-Shafer Theory1. INTRODUCTIONThe demand for chicken meat has been increasing because it has become cheaper relative toother meats The term poultry refers to domesticated fowl raised for meat or eggs [1]. Thepoultry industry is dominated by the chicken companies, development of poultry populationand poultry industry is very rapidly threatened by the presence of avian disease. Disease isdefined as a departure from health, and includes any condition that impairs normal bodyfunctions. Disease results from a combination of indirect causes that reduce resistance orpredispose an animal to catching a disease, as well as the direct causes that produce thedisease [2]. Avian influenza virus H5N1, which has been limited to poultry, now has spread tomigrating birds and has emerged in mammals and among the human population. It presents adistinct threat of a pandemic for which the World Health Organization and other organizationsare making preparations [3]. The current strain of H5N1 virus attacked poultry and humans invarious countries in Asia which caused many deaths in humans [4]. The World HealthAssembly urged its Member States to develop national preparedness plans for pandemicinfluenza [5].2. UNCERTAINTY IN EXPERT SYSTEMThe construction of expert and other intelligent computer systems requires sophisticatedmechanism for representing and reasoning with uncertain information. At least three forms ofuncertainty can be identified as playing a significant role in these types of systems. The first ofthese possibilistic uncertainty appears in situations where the value of a variable can only benarrowed down to a set of values one of which is the actual value of the variable. The secondkind of uncertainty is related to situations in which there exists uncertainty as to satisfaction ofa predicate by an element. This is manifested by concepts which have imprecise or grayboundaries. A very powerful tool for handling this type of uncertainty which also handles thefirst type of uncertainty is the fuzzy set. The third type of uncertainty is related to situations inwhich the value of a variable assumes can be modeled by the performance of a randomexperiment [17]. Figure 1 shows the process for dealing with uncertainty in Poultry DiseasesExpert System.
  2. 2. Step 1 Representation of uncertainty Alternative Rule on the set of basic events Step 2 Step 3 Merging uncertain information Conclusion Figure 1. The process for dealing with uncertainty in Poultry Diseases Expert SystemIn step 1, an expert provides uncertain knowledge in the form of rules with the possibility.These rules are probability values. In step 2, the uncertain knowledge on a set of basic eventscan be directly used to draw conclusions in simple cases (step 3). However, in many casesthe various events associated with each other. Therefore, it is necessary to combine theinformation contained in step 1 into the global value system. In step 3, the goal is knowledge-based systems draw conclusions. It is derived from uncertain knowledge in steps 1 and 2,and is usually implemented by an inference engine. Working with the inference engine, theexpert can adjust the input that they enter in Step 1 after displaying the results in steps 2 and3.3. DEMPSTER-SHAFER THEORYThe Dempster-Shafer theory was first introduced by Dempster [6] and then extended byshafer [7], but the kind of reasoning the theory uses can be found as far back as theseventeenth century. This theory is actually an extension to classic probabilistic uncertaintymodeling. Whereas the Bayesian theory requires probabilities for each question of interest,belief functions allow us to base degrees of belief for on question on probabilities for a relatedquestion. The advantages of the Dempster-Shafer theory as follows:1. It has the ability to model information in a flexible way without requiring a probability to be assigned to each element in a set,2. It provides a convenient and simple mechanism (Dempsters combination rule) for combining two or more pieces of evidence under certain conditions.3. It can model ignorance explicitly.4. Rejection of the law of additivity for belief in disjoint propositions.Poultry Diseases Expert System using Dempster-Shafer theory in the decision supportprocess. Flowchart of Poultry Diseases Expert System shown in Figure 2.
  3. 3. Start Yes Input Symptoms m1 (B) m2 (C) No m (A) = 1 End Figure 2. Flowchart of Poultry Diseases Expert SystemThe consultation process begins with selecting the symptoms. If there are symptoms then willcalculate, The Dempster-Shafer theory provides a rule to combine evidences fromindependent observers and into a single and more informative hint. Evidence theory is basedon belief function and plausible reasoning. First of all, we must define a frame of discernment, Θindicated by the sign Θ . The sign 2 indicates the set composed of all the subset generatedby the frame of discernment. For a hypothesis set, denoted by A, m(A)→[0,1]. m (Ø) = 0 (1)∅ is the sign of an empty set. The function m is the basic probability assignment. Dempstersrule of combination combines two independent sets of mass assignments. (∅) = 0 (2) (3)Where (4)4. POULTRY DISEASES EXPERT SYSTEMPoultry Diseases Expert System has four main architectural components that are theknowledge base, the inference engine, the knowledge acquisition module, and the userinterface for input/output. Figure 3 shows architecture of Poultry Diseases Expert System.
  4. 4. Knowledge Acquisition Knowledge Base Inference Engine User Interface Figure 3. Architecture of Poultry Diseases Expert System4.1 Knowledge acquisitionIn the present work, knowledge has been obtained from two sources. We acquired textualinformation from literature such as extension booklets, reports, papers, etc., related to theavian influenza diseases. The printed material allowed became familiar with the subject and amore effective communication with the experts. Most knowledge was acquired from theexperts using conventional interviewing techniques. The interview methods permitted a betterunderstanding of the problem and its further representation. This knowledge was provided byexperts on poultry diseases. Unstructured and structured interviews were used. Theunstructured interviews were used to define the familiar tasks involved in the process ofidentification, to obtain an initial understanding of the range of complications involved, and todefine specific problems for later discussion. The questions were more or less spontaneousand notes were taken on discussion. These methods were complemented with structuredinterviews. In the structured interviews, we revised and discussed in depth familiar tasks toclarify questions.4.2 Knowledge BaseA critical aspect of building an expert system is formulating the scope of the problem andgleaning from the source expert the domain information needed to solve the problem. Thereliability of an expert system depends on the quality of knowledge contained in theknowledge base. In this research, we built a Poultry Diseases Expert System and wedescribe five symptoms which include depression, combs, wattle, bluish face region, swollenface region, narrowness of eyes, and balance disorders. Knowledge base of which input onthe testing of this expert system can be seen in table 1.
  5. 5. Table 1: Knowledge Base No. Symptom Disease Basic Probability Assignment 1 Depression Avian Influenza 0.7 Newcastle Disease Fowl Cholera Infectious Bronchitis Respiratory Form Infectious Bronchitis Reproduction Form Swollen Head Syndrome 2 Combs, Wattle, Bluish Avian Influenza 0.9 Face Region 3 Swollen Face Region Avian Influenza 0.8 Newcastle Disease Fowl Cholera 4 Narrowness of Eyes Swollen Head Syndrome 0.9 5 Balance Disorders Newcastle Diseases 0.6 Swollen Head Syndrome4.3 Inference EngineThe inference engine is charged with the role of performing and controlling inference on theknowledge base. Specific features of the inference engine depend on the knowledgerepresentation scheme used for the knowledge base. Because the most commonrepresentation scheme is production rules, this inference engine will be exemplified. Thefollowing will be shown the inference engine using Dempster-Shafer Theory to diagnosepoultry diseases.Symptoms:1. Depression2. Combs, wattle, bluish face region3. Swollen face region4. Fever5. Pain in the rashSymptom 1Depression is a symptom of Avian Influenza (AI), Newcastle Disease (ND), Fowl Cholera(FC), Infectious Bronchitis respiratory form (IBRespi), Infectious Bronchitis reproduction form(IBRepro), and Swollen Head Syndrome (SHS). The measures of uncertainty, takencollectively are known in Dempster Shafer Theory terminology as a ``basic probabilityassignment (bpa). Hence we have a bpa, say m 1 of 0.7 given to the focal element {AI, ND,FC, IBRespi, IBRepro, SHS} in example, m 1({AI, ND, FC, IBRespi, IBRepro, SHS}) = 0.7,since we know nothing about the remaining probability it is allocated to the whole of the frameof the discernment in example, m1({AI, ND, FC, IBRespi, IBRepro, SHS}) = 0.3, so:m1{AI, ND, FC, IBRespi, IBRepro, SHS} = 0.7m1{Θ} = 1 - 0.7 = 0.3Symptom 2Combs, wattle, bluish face region are symptoms of Avian Influenza with a bpa of 0.9, so that:m2{AI} = 0.9m2 {Θ} = 1 – 0.9 = 0.1We calculate the new bpa for some combinations (m3). Combination rules for the m3 can beseen in the table 2.
  6. 6. Table 2: Combination of Symptom 1 and Symptom 2 {AI} 0.9 Θ 0.1 {AI, ND, FC, IBRespi, 0.7 {AI} 0.63 {AI, ND, FC, IBRespi, 0.07 IBRepro, SHS} IBRepro, SHS} Θ 0.3 {AI} 0.27 Θ 0.03 0.63  0.27m3 (AI) =  0.9 1 0 0.07m3 (AI, ND, FC, IBRespi, IBRepro, SHS) =  0.07 1 0 0.03m3 (Θ) =  0.03 1 0Symptom 3Swollen face region is a symptom of Avian Influenza, Newcastle Disease, Fowl Cholera witha bpa of 0.8, so thatm4 {AI, ND, FC} = 0.8m4 (Θ) = 1 – 0.8= 0.2We calculate the new bpa for some combinations (m5). Combination rules for the m5 can beseen in the table 3. Table 3: Combination of Symptom 1, Symptom 2, and Symptom 3 {AI, ND, FC} 0.8 Θ 0.2 {AI} 0.9 {AI} 0.72 {AI} 0.18 {AI, ND, FC, IBRespi, 0.07 {AI, ND, FC} 0.06 {AI, ND, FC, IBRespi, 0.01 IBRepro, SHS } IBRepro, SHS } Θ 0.03 {AI, ND, FC} 0.02 Θ 0.01 0.72  0.18m5 (AI) =  0.9 1 0 0.06  0.02m5 (AI, ND, FC) =  0.08 1 0 0.01m5 (AI, ND, FC, IBRespi, IBRepro, SHS) =  0.01 1 0 0.01m5 (Ө) =  0.01 1 0Symptom 4Narrowness of eyes is a symptom of Swollen Head Syndrome with a bpa of 0.9, so that:m6 (SHS) = 0.9m6 (Θ) = 1 – 0.9 = 0.1We calculate the new bpa for some combinations (m7). Combination rules for the m7 can beseen in the table 4. Table 4: Combination of Symptom 1, Symptom 2, Symptom 3 and Symptom 4 {SHS} 0.9 Θ 0.1 {AI} 0.9 Θ 0.81 {AI} 0.09 {AI, ND, FC} 0.08 Θ 0.07 {AI, ND, FC} 0.01 {AI, ND, FC, IBRespi, 0.01 {SHS} 0.01 {AI, ND, FC, IBRespi, 0.001 IBRepro, SHS} IBRepro, SHS} Θ 0.01 {SHS} 0.01 Θ 0.001
  7. 7. 0.01  0.01m7 (SHS) =  0.17 1  (0.81  0.07) 0.09m7 (AI) =  0.75 1  (0.81  0.07) 0.01m7 (AI, ND, FC) =  0.08 1  (0.81  0.07) 0.001m7 (AI, ND, FC, IBRespi, IBRepro, SHS) =  0.01 1  (0.81  0.07) 0.001m7 (Θ) =  0.01 1  (0.81  0.07)Symptom 5Balance disorders is a symptom of Newcastle Diseases and Swollen Head Syndrome with abpa of 0.6, so that:m8 {ND,SHS} = 0.6m8 {Θ} = 1 - 0.6 = 0.4We calculate the new bpa for some combinations (m9). Combination rules for the m9 can beseen in the table 5. Table 5: Combination of Symptom 1, Symptom 2, Symptom 3, Symptom 4 and Symptom 5 {ND, SHS} 0.6 Θ 0.4{SHS} 0.17 {SHS} 0.1 {SHS} 0.07{AI} 0.75 Θ 0.45 {AI} 0.3{AI, ND, FC} 0.08 {ND} 0.05 {AI, ND, FC} 0.03{AI, ND, FC, 0.01 {ND, SHS} 0.01 {AI, ND, FC, IBRespi, IBRepro, 0.004IBRespi, IBRepro, SHS}SHS}Θ 0.01 {ND, SHS} 0.01 Θ 0.004 0.1  0.07m9 (SHS) =  0.31 1  0.45 0.3m9 (AI) =  0.55 1  0.45 0.05m9 (ND) =  0.1 1  0.45 0.01  0.01m9 (ND, SHS) =  0.04 1  0.45 0.03m9 (AI, ND, FC) =  0.05 1  0.45 0.004m9 (AI, ND, FC, IBRespi, IBRepro, SHS) =  0.01 1  0.45 0.004m9 (Θ) =  0.01 1  0.45The most highly bpa value is the m9 (AI) that is equal to 0.55 which means the possibility of atemporary diseases with symptoms of depression, comb, wattle, bluish face region, swollenregion face, narrowness of eyes, and balance disorders is the Avian influenza (H5N1). Table6 and figure 4 are shows the basic probability assignment final result and graphic ofidentification.
  8. 8. Table 6. Final Result No. Diseases Basic Probability Assignment 1 Swollen Head Syndrome 0.31 2 Avian Influenza 0.55 3 Newcastle Disease 0.1 Newcastle Disease 0.04 4 Swollen Head Syndrome 5 Avian Influenza 0.05 Newcastle Disease Fowl Cholera 6 Avian Influenza 0.01 Newcastle Disease Fowl Cholera Infectious Bronchitis respiratory form Infectious Bronchitis reproduction form Swollen Head Syndrome Figure 4. Graphic of IdentificationFigure 4 shows the graphic of identification, the highest basic probability assignment is AI(Avian Influenza) that is equal to 0.55 which show from the last calculation of Dempster-Shafer on symptom 5 which means the possibility of a temporary diseases with symptoms ofdepression, comb, wattle, bluish face region, swollen region face, narrowness of eyes, andbalance disorders is the Avian Influenza.4.4 User InterfaceA user interface is the method by which the expert system interacts with a user. The input oroutput interface defines the way in which the expert system interacts with the user and othersystems such as databases. Interfaces are graphical with screen displays, windowing, andmouse control. They receive input from the user and display output to the user The userinterface development of Poultry Diseases Expert System to begin with designing web pagesand designing a web-based applications. Designing web pages using PHP, the web pagethen connected to the MySQL database, after that designing a web-based applications whichused to access the network database. The relationship between applications with networkdatabase has shaped overall applications to manage information to be displayed.
  9. 9. 5. IMPLEMENTATIONThe following will be shown the working process of expert system in diagnosing a case. Theconsultation process begins with selecting the symptoms found on the list of symptoms. In thecases tested, a known symptoms are depression, comb, wattle, bluish-colored facade region,region of the face swollen, eyes narrowed and balance disorders. The consultation procescan be seen in Figure 5. Figure 5. Symptoms SelectionIn the case of depression, comb, wattle and region of the face bluish, region of the faceswollen, eyes narrowed and lachrymal glands swollen. The result of consultation is avianinfluenza. Figure 6 shows the result of consultation. Figure 6. The Result of Consultation
  10. 10. 6. CONCLUSIONIdentification of poultry diseases can be performed using Dempster-Shafer Theory. In thispaper we describe 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 databaseis that of attaching a probability to every member of a relation, and to use these values toprovide the probability that a particular value is the correct answer to a particular query. Anexpert in providing knowledge is uncertain in the form of rules with the possibility, the rulesare probability value. The knowledge is uncertain in the collection of basic events can bedirectly used to draw conclusions in simple cases, however, in many cases the various eventsassociated with each other. Knowledge based is to draw conclusions, it is derived fromuncertain knowledge. Reasoning under uncertainty that used some of mathematicalexpressions, gave them a different interpretation: each piece of evidence (finding) maysupport a subset containing several hypotheses. This is a generalization of the pureprobabilistic framework in which every finding corresponds to a value of a variable (a singlehypothesis). In this research, Poultry Diseases Expert System has been successfullyidentifying Poultry Diseases diseases and displaying the result of identification process Thisresearch can be an alternative in addition to direct consultation with doctor and to find outquickly poultry disease. REFERENCES [1]. B. L. Ligon, Avian Influenza Virus H5N1: “A Review of Its History and Information Regarding Its Potential to Cause the Next Pandemic,” Seminars in Pediatric Infectious Diseases, Elsevier, pp. 326 – 335, 2005. [2]. A. S. Fauci, “Emerging and Re-Emerging Infectious Diseases: Influenza as a Prototype of the Host-Pathogen Balancing Act,” Elsevier, Cell, pp. 665 -670, 2006. [3]. 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, pp. 159 – 164, 2006. [4]. W. H. Assembly, “Strengthening pandemic-influenza preparedness and response,” Resolution WHA58.5, 2005. [5]. E. Azziz-Baumgartner, N. Smith, R. González-Alvarez, “National pandemic influenza preparedness planning,” Influenza and Other Respiratory Viruses, pp. 189 – 196, 2006. [6]. Anonim, Lampung in Figures, BPS – Statistics of Lampung Province, BPS, Lampung, 2010. [7]. J. Durkin, Expert system: Catalog of applications: Intelligent Computer Systems, Inc., Akron, OH. First Edition, 1993. [8]. J. Durkin, Expert Systems: Design and Development, Prentice Hall, Englewood Cliffs, N.J., 1994. [9]. J. Giarratano, and G. Riley, Expert Systems: Principles and Programming, Fourth Edition. Boston, MA, Thomson/PWS Publishing Company, 2004. [10]. S. Russell, and P. Norvig, Artificial Intelligence: A Modern Approach, Prentice Hall, Englewood Cliffs, NJ, Second Edition. 2002. [11]. E. A. Feigenbaum, “Dendral and Meta-Dendral: Roots of Knowledge Systems and Expert System Applications,” Artificial Intelligence 59 (1-2), pp. 233 – 240, 1993. [12]. J. McDermott, “R1: an expert in the computing system domain,” In Proceedings of the National Conference on Artificial Intelligence, pp. 269–271, 1980. [13]. E. H. Shortliffe, “Computer-Based Medical Consultations: MYCIN,” Elsevier, New York, 1976. [14]. E. Turban, J. E. Aronson, Decision support systems and intelligent systems, sixth Edition (6th ed), Hongkong: Prentice International Hall. 2001.[15]. V. Akbar, T. Sulistyorini, “Expert System for Avian Influenza Diagnose using Visual Basic.NET,” Research Report, Gunadarma University, 2007.[16]. U. Ridwan, “Sistem Pakar Diagnosa Penyakit Avian Influenza pada Unggas Berbasis Web,” Research Report, IPB, 2010.
  11. 11. [17]. R. Yager, “Reasoning with Uncertainty for Expert Systems,” International Joint Conference on Artificial Intelligence, 1985.[18]. A. P. Dempster, “A Generalization of Bayesian inference,” Journal of the Royal Statistical Society, pp. 205 – 247, 1968.[19]. G. Shafer, A Mathematical Theory of Evidence, Princeton University Press, New Jersey, 1976.

×