Daily rainfall prediction using clonal selection algorithm

• Main Authors: Noor Rodi, Nur Syazwani, Ismail , Amelia Ritahani, Abdul Malik, Marlinda
• Format: Conference or Workshop Item
• Language: English; English
• Published: 2012
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://irep.iium.edu.my/28575/1/Fullpaper_ICWR.pdf; http://irep.iium.edu.my/28575/2/AcceptanceLetter-ICWR-shazwani.pdf; http://irep.iium.edu.my/28575/

Daily rainfall prediction is important in water resources management in order to estimate long term forecasting for future usage. This paper utilizes Artificial Immune System (AIS) algorithm for daily rainfall prediction where rainfall is used as input nodes. AIS have the abilities of self-organizing, memory, recognition, adaptive and learning, inspire from the immune system. Immune system is a system in our body which consists of a complex set of cells and molecules that protect our bodies against infection. There are three mains algorithm in AIS which are Clonal Selection Algorithm (CSA), Immune Network Algorithm and Negative Selection Algorithm. This study focuses on development of daily rainfall prediction model using CSA adapted from clonal selection theory in human being and animal immune system. The common use of Clonal Selection Algorithm is to randomly generate the antibodies in the population. From the experiment, the results obtained more than 90% accuracy by using 100, 500 and 1000 numbers of detector which is promising that CSA can be a good predictor algorithm.