Machine learning is the field of study that offers computers the ability to learn without being specifically programmed to do so.

It’s a sub-area of artificial intelligence that permits computers to emerge into a self-learning mode. Thus, when faced to new data sets, these computer programs are enabled to learn, grow and develop by themselves.

Related course: Complete Machine Learning Course with Python

Machine Learning Types

Even though the concept of machine learning is used from a long time ago, the capability to autonomously conduct complex mathematical considerations to big data sets it is a recent fact that was achieved in this field a couple of years back.

Also, there are different types of machine learning, along with specific algorithms that properly work with a certain type of machine learning.

  • Supervised machine learning

    Supervised deals with simple tasks, for which recognition algorithms are employed.

    These tasks can only be done if the computer has access to an already established data of input-output pairs.

    If algorithm uses training data, it’s a supervised learning algorithm.

    iris = datasets.load_iris() 
    X =[:, :2]
    Y =
  • Unsupervised machine learning algorithms

    Seek patterns from a dataset of correct answers, being useful for exploratory analysis.

    If algorithm does not use training data, but directly runs on the data set, it’s an unsupervised algorithm

    X = np.array([[1, 2], [1, 4], [1, 0], [10, 2], [10, 4]])
    kmeans = KMeans(n_clusters=2, random_state=0).fit(X)
    kmeans.predict([[0, 0], [12, 3]])

    Frequently used algorithms for this type of learning are k-means clustering, main component analysis or autoencoders.

  • Reinforcement learning

    offers to the machine a method to measure its performance by using positive reinforcement.

    This is quite similar to the manner in which humans or animals learn certain tasks due to the fact that the machine tries different ways of solving a problem and it gets rewarded with a signal if it is successful.


Machine learning applications are vast and applicable to different domains.

  • Data security is one of the domains in which machine learning is more than useful due to the fact that malware is not a problem that will simply disappear. Usually, when it comes to identifying a malware, machine learning is employed for identifying patterns and signal abnormalities.

  • Other uses for machine learning algorithms can be observed in the field of financial trading. Patterns and predictions are those that keep stock markets in the game, which signifies that the algorithms used for machine learning aid in predicting and executing transactions.

  • Marketing personalization also uses machine learning algorithms in order to establish a targeted customer experience depending on their behaviour or location-based data.

  • What is even more interesting about machine learning algorithms, is the fact that ML can be used in health care, too.

    Many scientists and researchers established different patterns in order to train machine for detecting cancer by simply observing cell images. This can be achieved with a vast amount of great quality image data and machine learning algorithms to predict the possibility of a patient getting cancer.

Hence, machine learning is a process that continuously evolves and changes, which might turn out to be more than helpful for our everyday tasks. ML represents a tool that effectively automates the process of analytical model building and permits machines to adapt to new set-ups independently.

Download examples