No abstract

Hepatitis is a dreaded disease that has taken the lives of so many people over the recent past years. The research survey shows that hepatitis viral disease has five major variants referred to as Hepatitis A, B, C, D, and E. Scholars over the years have tried to find an alternative diagnostic means for hepatitis disease using artificial intelligence (AI) techniques in order to save lives. This study extensively reviewed 37 papers on AI based techniques for diagnosing core hepatitis viral disease. Results showed that Hepatitis B (30%) and C (3%) were the only types of hepatitis the AI-based techniques were used to diagnose and properly classified out of the five major types, while (67%) of the paper reviewed diagnosed hepatitis disease based on the different AI based approach but were not classified into any of the five major types. Results from the study also revealed that 18 out of the 37 papers reviewed used hybrid approach, while the remaining 19 used single AI based approach. This shows no significance in terms of technique usage in modeling intelligence into application. This study reveals furthermore a serious gap in knowledge in terms of single hepatitis type prediction or diagnosis in all the papers considered, and recommends that the future road map should be in the aspect of integrating the major hepatitis variants into a single predictive model using effective intelligent machine learning techniques in order to reduce cost of diagnosis and quick treatment of patients.

No abstract

processing and management of students' results involve many complex stages (Okebukola, 2010). As more information are made available in a variety of formats and sources, processing and management of data becomes more and more complex, which can be handled using Information System approach (Oliveira, Thomas & Espadanal, 2014). Despite the proliferation of vendor result processing and management systems, the automation level of the existing singleplatform system models prevalent in Nigerian Universities are unsatisfactory. The existing single-platform systems are the completely offline (standalone) system and the completely online (client-server) system, which can be web-based or application-based (

processing and management of students' results involve many complex stages (Okebukola, 2010). As more information are made available in a variety of formats and sources, processing and management of data becomes more and more complex, which can be handled using Information System approach (Oliveira, Thomas & Espadanal, 2014). Despite the proliferation of vendor result processing and management systems, the automation level of the existing singleplatform system models prevalent in Nigerian Universities are unsatisfactory. The existing single-platform systems are the completely offline (standalone) system and the completely online (client-server) system, which can be web-based or application-based (