Is Perceptron a data-structure or an algorithm or both?

Question

I found Perceptron in the discussions of (overlapping of disciplines is acknowledged):

1. Neural Network
2. Machine Learning
3. Data Mining, Pattern Recognition
4. Genetic Algorithm

In the discussion of Neural Network, it is defined as:

Perceptron is a single layer neural network.

In the discussion of Machine Learning, it is defined as:

In machine learning, the Perceptron is an algorithm for supervised learning of binary classifiers.

In the discussion of Data Mining, it is defined as:

One of the older approaches to this problem in the machine learning literature is called the perceptron algorithm, and was invented by Frank Rosenblatt in 1956.

In the discussion of Evolutionary ALgorithm, it is defined as:

???

So, my questions are,

1. Is Perceptron a data structure or an algorithm or both?
2. Can you prescribe a book where I can learn about Perceptrons from Data Mining point of view?
3. Can you prescribe a book where I can learn about Perceptrons from Evolutionary Algorithm point of view?

Answer 1

1. A perceptron is simply a single-layer neural network. When we refer to neural networks more broadly, we are referring to multi-layer perceptrons because these typically have more than one hidden layer. It is an algorithm that is used specifically for training of binary classifiers.

2. You might find the following of use: "Support Vector Machines and Perceptrons: Learning, Optimization, Classification, and Application to Social Networks", by Murty and Raghava. The book focuses on application of SVMs and perceptrons to social networks, but given this area relies significantly on data mining to conduct analysis, there might be some good examples here.

3. "Neural Networks and Genome Informatics, Volume 1 (Methods in Computational Biology and Biochemistry)" by Wu and McLarty describes how to apply neural networks to the field of genome informatics, and may be quite informative in this regard.

Disclaimer: I have no affiliation with the authors of the titles I cited, or any other relationship therein. I have not read the titles personally, but anticipate that they might be useful for the information you are seeking.