MENGENALI FUNGSI LOGIKA “AND” MELALUI PEMROGRAMAN PERCEPTRON DENGAN MATLAB

Yaya Finayani

Abstract


This paper focuses on recognizing the logic function "and" with two input variables through programming perceptron. Perceptron is a simple form of artificial neural networks, perceptron is usually used to produce a certain type of pattern is often known by the separation in a linear fashion. The first simulations, carried out the programming perceptron to recognize the logic function "and" 2 input variables with binary inputs (0 or 1) and the target output is binary (0 or 1), simulation results show to get the performance = 0 which means all the patterns recognized correctly 9 epoch iteration process is needed. The second simulation, programming perceptron to recognize the logic functions 'and' two input variables with bipolar inputs (-1 or 1), the results of simulations required four epoch iteration process. The third simulation, programming perceptron to recognize the logic function “and” 2  input variables with binary input, but the initial value of weights and the initial bias = 0, simulation result show to get to work = 0 8  epoch iteration process is required. From the simulation results concluded that the programming perceptron to recognize the logic function "and" 2 input variables using a binary input (0 or 1) a longer iteration process (9 epoch) compared with bipolar inputs (-1 or 1) with a 4 epoch, the value initial weights and biases affect the initial value of the iteration process is faster iteration process (8 epoch) for the initial value of weights and the initial value bias = 0 compared with the initial value of weights (-1.1) and the initial value bias = 1 with 9 epoch.

Keyword: perceptron, logic function, and, epoch, programming, Matlab.

Full Text: PDF

Refbacks

  • There are currently no refbacks.


Pusat Penelitian dan Pengabdian Masyarakat Politeknik Pratama Mulia Surakarta