How to classify data with a Perceptron Neural Network?

Asked

Viewed 191 times

1

I created a perceptron neural network to understand the concept. My goal would be to define what an orange is by its number of chemical elements and what an apple is by the same formula. It’s a neural network of just a layer, I would have to use a network of multi-layered to carry out this process? Could someone show a practical example? It would be nice to talk a little about.

  • It would be great if someone provided an algorithm example in practice applying this concept of AI MLP, even more because next semester I will do a TCC on exactly that rs.

  • @Pauloroberto truth! I understood the concept and put into practice, but there are many doubts still how to use this type of RN.

  • you have this answer here https://answall.com/a/82061/3082. to give an in depth in the situation if you are interested

  • Yeah, but I want to talk specifically about Perceptron. I want to know what is possible to do and what is not? Like the example I passed...

  • 1

    The perceptron is a linear binary sort. You can solve any problem with this algorithm, provided that the data sets can be classified into two (i.e., oranges and apples, for example) and linearly separable (if the data is in R n, there is a hyperplane in R (n-1) that divides the two groups). This is the answer sought?

  • Colleague @Marcus is absolutely right. That is, you can use a single perceptron to solve the problem of oranges vs apples if these "data" (represented by "chemical elements" - whatever you mean here) are linearly separable. If you want practical examples, take a look here: http://machinelearningmastery.com/implement-perceptron-algorithm-scratch-python/

  • @Luizvieira I think to pass for example the acidity of the fruits as an example.

  • 1

    Okay. Then do it like this: measure the acidity of some fruits (both oranges and apples) and Plote in a graph (can be a histogram even, with the count of the number of fruits of each type by acidity value). If with this unique feature the problem is linearly separable it will be very visible on the graph (as in the classic fish problem: https://answall.com/questions/40135/explica-algoritmo-svr/40149#40149).

  • 1

    Interesting! @Luizvieira

Show 4 more comments
No answers

Browser other questions tagged

You are not signed in. Login or sign up in order to post.