|Title:||Application of Rule Based Advisory System to Identify Species of Pathogenic Bacteria|
|Other Titles:||In the Case of National Animal Health Diagnostic and Investigation Center (Nahdic), Sebeta, Ethiopia|
|???metadata.dc.contributor.*???:||Dr. Tibebe Beshah|
|Publisher:||Addis Ababa University|
|Abstract:||Expert Systems are computer programs that use artificial intelligence to solve problems within a specialized domain that ordinarily requires human expertise. In the clinical laboratory setting, a very important aspect of helping physicians with diagnosis of infections is the identification of microorganisms that are causing these infections. Most outstanding of these infections are infections caused by disease causing bacteria. These microorganisms are involved in virtually any type of bacterial infection in animals and humans. Their identification, although very important, is often difficult and has complex procedure. However, unlike countries with emerging and developed economies, diagnosis of diseases in countries like Ethiopia is less supported with computer based laboratory analysis. The main reason for this to happen, particularly with bacterial diseases, is the expensiveness and complicated nature of the laboratory procedures, conventional identification method is time consuming, the result can be inaccurate and further exacerbated by the limited number of qualified and experienced personnel in the domain area. This thesis addresses the above problems by providing a rule based advisory expert system that takes into consideration various clinical factors and types of tests about the microorganism, and give advice to the laboratory technicians and domain experts to facilitate the identification procedure as a possible solution. To this end, knowledge is acquired using unstructured interviews from domain experts and laboratory technicians which are selected using purposive sampling technique from National Animal Health Diagnostic and Investigation Center, Sebeta, Ethiopia and observation on the conventional method of identification in the center. Relevant documents analysis method is also followed to capture explicit knowledge. Then, the acquired knowledge is modeled using decision tree, decision table and hierarchical structure that represents concepts and procedures involved in identification of pathogenic bacteria up to species level and production rules are used to represent the domain knowledge rule based advisory system is developed using SWI Prolog editor tool. It uses backward chaining which begins with possible solutions or goals and tries to gather information that verifies the solution. In testing and evaluating the prototype system all the domain experts and laboratory technicians in the selected center are selected in order to test the performance and user acceptance of the prototype system. Thus, the overall total average performance and user acceptance of the prototype system is 87.5%. The prototype system achieves a very good performance and meets the objectives of the study. However, in order to make the system more applicable in the domain area additional study is needed like updating the rules in the knowledge base of the system automatically, incorporating a well-designed graphic user interface and a mechanism of NLP facilities like speech output and image processing|
|Description:||A Thesis Submitted to College of Natural Science of Addis Ababa University in Partial Fulfillment of the Requirement for the Degree of Master of Science in Information Science|
|Appears in Collections:||Thesis - Information Science|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.