Course Lesson 1

Course:

Selection of simple algorithms for Machine Learning applications.

Lesson 1: Identify Patterns.

To remind: Machine Learning, a subset of artificial intelligence, is the eternal search for patterns. After grouping the known data, and graphing them, you only have to look for the equation (and its algorithm) that unites them (if there is any linear relationship), or that groups them (if they have common characteristics). After feeding data to your machine learning model (dataset), if you help the computer, to do this job, we will call it «supervised learning». if you let the computer do all the work (find the patterns and select the algorithm), we will call it: «Unsupervised learning». so, to «make a forecast», we just need to evaluate that function, or see if that new data belongs to one of the created groups. then, for us, the computer will make a decision, for example: «yes, or not».

Step 1: Do you know what a pattern is? Let’s see this image:

Step 2: And now, check out these:

Step 3: Now, you have a visual image of what a pattern is.

Step 4: However, do not be fooled, the human brain has evolved to «find patterns», and many times, we can be wrong:

Step 5: do you know the word “Apophenia”? Have you ever looked at a cloud and thought it looked like an animal? Or have you ever looked at the moon and thought there was a face on? I invite you to visit these links, so you can learn a little more; It would even be fun, until you realize that it can lead to false conclusions in an investigation.

https://en.wikipedia.org/wiki/Apophenia

Step 6: Ah, but this is also a pattern, more abstract, but you can see it.

Step 7: By the way, you are right, the value of «…» is 16, because the logic of this pattern is that the number is increased by twice its value.

Step 8: Or better, let’s make a graph, as we did in school:

We can even use our beloved spreadsheet:

Step 9: Generally, the data, before being processed by our Machine learning application, is grouped into patterns. “Data engineers” take care of this, even when digital images are analyzed; for example, for face recognize, these patternss (pixels), these pixels are stored in an array.. Don’t worry too much.

Only, remenber this cat:

Step 10: In the next classes, we will work more with data, imagine that these have previously been obtained from large data repositories, using Big Data techniques, and that these have been grouped into graphs, diagrams, or schemes that show how they are distributed said data, statistically, to answer a question, or research (data scientists are responsible for this). Each point is a data, and the color of this data, is only a “label”. For example:

Step 11: Better, Let’s see for now, more examples of graphic patterns, in artistic images:

Step 12: Find an image on the internet, which is also a pattern, and explain why; you can also create that image yourself, by hand, or in some drawing software or APP; If you prefer numbers (abstract thinking), you can use them, indicating why they are a pattern. Do not forget that it is mandatory that, somewhere in this image, your student identification number must be visible. In any case, you will have to take that image, digitize it and place it in a repository of free images, or on a page of your blog. Then, place the link to your image here, or try to embed it as a comment, accompanying it with your explanation. Likewise, I ask you to also respond to the comment of one of your classmates, arguing why you agree or why you don’t. to your examples. Use this blog forum to show your work.

I recommend you wait for the response of your facilitator instructor, to continue with the next lesson. Good Luck!