Perceptron: A Simple Classification Method

In real life, high accuracy and quick access to information for certain topic or object is essential. In order to achieve those goals, things have to be organized or classified into classes. Moreover, it would be better if machine could work for human to deal with classification problems. The purpose of this study is to teach the perceptron how to perceive or classify some of the basic things in the real world similarly to human such as images recognition and document classification etc.  The perceptron first has  to go through the process of training by using the labeled data given by the outside sources. After the perceptron  is trained,  it will get the values of weights in vector form, and those values will be ultilized by the program to classify the unknown data. The accuracy of data classification depends on how well the program is trained, and it also depends on how complex the classification problem is. The perceptron is known for solving simple classification problems with good accuracy. One of the interesting feature of perceptron is that it can combine with many other perceptrons to create a more complex network to deal with more challenging classification problems. Last but not least, the program can work on some classification tasks like human.