AIVIS

By: Justin L.
Year: 2021
School: Troy High
Grade: 10
Science Teacher: Kevin Jones

AIVIS is an online application that allows for easy creation of an Artificial Neural Network.

Artificial neural networks (ANNs), usually simply called neural networks (NNs), are computing systems inspired by the biological neural networks that constitute animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain.

The user launches the application via a web browser, and will then be greeted by the home screen, where the user can click the “get started” button in order to proceed into the network editor page. Once on the editor page, the user will be greeted by a configurable multi-layer perceptron model neural network, a network configuration bar, as well as a multi-purpose display-bar. The user can input their input and label dataset as a CSV file format. They can also modify the configurable network by options present on the network configuration bar, or scroll on the layers to change the numbers of perceptrons present in each layer. The user can then train the model, assuming they have a proper network and have input in all the training parameters, and also test it to view the result of training. A help function is also provided in case the user does not know how to use the program.

This project eliminated the need of programming or mathematical calculation while creating deep learning models through the use of a web browser interface.

Analysis and Results:
I had to change the original prototype to include a more compact output as well as a smaller network size to be able to handled by a web browser