Abstract:
Poultry diseases remain one of the major threats to poultry production. The diseases of chickens
need to be observed intensively because of its impact on the health and quality of chicken
production. Chicken disease becomes one of the problems that are very detrimental to chicken
farmers. Accurate information about handling chicken disease is still difficult to obtain. Expert
knowledge (including agricultural and medical experts), which is valuable to control and/or
treat such diseases are limited in developing countries in general and in Ethiopia in particular.
In attempt to solve this problem, knowledge based system is identified as a powerful tool with
extensive potential in alleviating agricultural and medical problems. Therefore, it is necessary to
design a knowledge base system application in detecting diseases that are experienced by
chickens based on the symptoms shown and how to handle them. This study aims at developing
knowledge base system for diagnosing, prevention and management of predominant chicken
diseases. The predominant chicken diseases selected for the purpose of this study include:
Newcastle disease, marek’s disease, infectious coryza, fowl cholera, chicken mite, coccidiosis,
aspergillosis and favus. Design Science research method was used to develop the prototype
system. To select domain experts for knowledge acquisition, purposive sampling technique was
used. The domain experts were selected from Jimma University College of Agriculture and
Veterinary Medicine and from Kito Furdisa Poultry Farm. The knowledge was acquired using
both structured and unstructured interviews from domain experts and relevant document
analysis method was also followed to capture explicit knowledge. The acquired knowledge was
modeled using decision tree that represents concepts and procedures involved in diagnosis,
prevention and management of predominant chicken disease and production rules (If-ThenAction) were used to represent the domain knowledge and Knowledge-based system was
developed using SWI Prolog editor tool. Backward chaining algorithm was used in this study. At
the end performance of the system was evaluated and produced a result of 83%. Moreover, user
acceptance of the developed system was done by visual interaction method, namely, by showing
the system to the domain experts and it was found to produce 83.4%. Thus, the average
performance of the prototype system is 83.2%. The prototype system achieves a very good
performance and meets the objectives of the study. Lastly, it is recommended that the stake
holders should have to take part in deploying the developed system.
