Project
1. 3.4 Project Can You Believe I Just Created a Neural Network Algorithm?

Artificial Intelligence (A.I.) has been trying to emulate human intelligence in computational devices. This week, you will get first-hand experience with what that machine learning looks like and how it unfolds over time.
2. Before you get started…
Before this week’s assignment, it would be helpful to see a neural network in action. Go to ConvNetJS. Watch three different Browser Demos and jot down your observations.
As you can see from this site, the more time the computer spends learning and adapting, the more efficiently and effectively it comes up with an answer. When you teach the computer how to read handwriting, it must be exposed to various ways that letters and numbers can be written. This training comes from an artificial neural network (A.N.N.). The A.N.N. helps the computer understand human handwriting, even though handwriting is unique from person to person. One way it does this is by showing the computer lots of different handwriting samples and identifying the letters and numbers written. For example, an artificial neural network helps the Apple Watch transcribe a handwritten text into a typed text. As you can see, an A.N.N. is a powerful tool that you can use to solve complex problems in your workplace practice, evidenced by the way it can understand the diversity of handwriting.
While neural networks are powerful tools, you must teach them sound data collected on a sample of the population that you are studying. As you recall, a neural network is similar to a human brain, depicted in the image above. To determine predictions of the element in question you must follow logical protocols. Therefore, you must choose data sources wisely. Additionally, you also must make sure that the population you are studying is represented in the sample collection. For your answers to be valid, reliable, and repeatable, it is critical that you take due caution in selecting the data and sample that you are using to make inferences about the population in question.
3. Instructions
Follow the instructions below to complete this assignment:
4. Step One
Go to the Tensorflow playground. Use the + and – buttons to add and subtract nodes and neurons from the network. Also, experiment with changing the learning rate using the dropdown menu in the ribbon at the top.
Observe how the Output (on the right-hand side of the screen) changes. This will help you answer question #1 below.

5. Step Two
Respond to each of the questions below in a Word document. Your document should be two pages, excluding the title and reference page. Your answers should require additional resources. Be sure to cite those resources in APA format. Don’t forget to include your own opinions, interpretations, recommendations, and conclusions to these questions.
Before you begin, check out the rubric below to learn more about how you will be graded on this assignment.
1. Now that you have experimented with the nodes and neurons in the Tensorflow playground, what were your observations?
2. What did you learn about machine learning and neural networks when you went through the A.N.N. activities located on the ConvNetJS site?
3. Why is it important to teach an A.N.N. with sound data?
4. Why must you collect data from a representative sample of the population for the A.N.N. to help us solve problems accurately?