00000ctmaa22000003a 4500 UP-1685675941123923323 Buklod 20151015151525.0 aa rb |||1 u| ta 151015s xx d r |||| u| (iLib)UPTAC-00017720437 UPVTC eng LG 993.5 2009 C66 A43 Albao, Tyron Jan G. author. ANNAISocr artificial neutral network using artificial immune system in optical character recognition Tyron Jan G. Albao. 2009. [4], 54 leaves illustrations. Undergraduate thesis (B.S. Computer Science) -- University of the Philippines, Tacloban. The application of Artificial Neutral Networks is the most common approach to optical character recognition. Artificial neutral networks are configured to be used in pattern recognition or data classification with the use of a learning process (supervised or unsupervised) called training. Many algorithms used in neutral networks have already been proven to perform well when applied to optical character recognition. This study proposes an immune inspired algorithm, the Clonal Selection Algorithm, used as a training algorithm in neutral networks. In this approach, the input to the neural network is treated as an antigen and the set of weights of the network as the antibodies. Using a fitness function, the weights that are carried by best-fit antibody will be the optimal solution. Experiment results show that an artificial immune system is an effective training algorithm in artificial neural network. Character recognition. Artificial Neural Networks. Artificial Immune System. Clonal Selection Algorithm. Training algorithms. FI UP UPTAC UPTAC LG 993.5 2009 C66 A43 Thesis