Introduction To Machine Learning — Session 1

In this sessions, I’m going to introduce to you the idea of machine learning. We will learn the background of machine learning, we will see some shallow learning of shallow learning ML (machine learning) and go deep to peek into deep learning.

So, What is Machine Learning?

Machine learning is a type of artificial intelligence (AI) that provides computers with the ability to learn without being explicitly programmed. Machine learning focuses on the development of computer programs that can change when exposed to new data.
 — By wikipila

Basically, machine learning is when the program changed by the input it gets (the training) and not by the code that it was written. In the past, we use rule-based programming, when this happened to do that, with the rule base system the code stay static and is not change but the program stays static also it doesn’t need the “training” phase that machine learning needs.

Terminology

  • Feature — An individual measurable property of a phenomenon being observed.
  • Samples — A sample is an item to process (e.g. classify). It can be a document, a picture, a sound, a video, a row in database or CSV file, or whatever you can describe with a fixed set of quantitative traits.
  • Feature vector — An n-dimensional vector of numerical features that represent some object (a row in a CSV file).
  • Feature extraction — 
    1.
    Preparation of feature vector.
    2. Transforms the data in the high-dimensional space to a space of fewer dimensions.
  • Training/Evolution set — Set of data to discover potentially predictive relationships.

Use-Cases

  1. Spam Email Detection
  2. Language Translation
  3. Spell Checker

4. Image Search

5. Amazon Recommendations (Clustering)

6. Face Detection

7. Fraud detection

8. Decision Making (: e.g. Bank/Insurance sector)

and more, and more, and more ….

The Machine Learning “Tree”

Supervised Learning

The machine learning task of inferring a function from labeled training data

Let’s say that we have this kind of picture data, of dog and hotdogs, and each picture data has a label if it is a hotdog or not. (the picture data can be features extract from the picture like colors, shapes, paces and more)

We train our supervised model with all the input data which is labeled and the model is learning what defined hotdog best, which feature and what “level” of combinations is a hotdog and what not.

when we are done with the training phase, our model is ready to predict to us if the data is the data of a hotdog or not.

we are not sending to the model unlabeled data with the same features as the label data (without the label) and ask the machine if this is a hotdog or not?


Next Time

In session 2 we will start with unsupervised learning, Reinforcement learning, I will show you some framework for machine learning and Techniques.

Until next time, enjoy reading :) and don’t forget to claps… it is free :)

Like what you read? Give KovaCode a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.