With all the development in technology sometimes it is hard to grasp concepts such as synthesis ai machine learning. To better understand what machine learning is, the datasets for machine learning and how it works, we need a simple example. Machine learning is usually split into two; they have supervised learning and unsupervised learning. See the following explanations for simple examples of each of these approaches.
Examples of the application of supervised learning. Imagine we are going to separate apples from bananas. And we have 2 fruit baskets, namely fruit basket A and fruit basket B. In supervised learning, the fruits in basket A have been labeled, namely Bananas and Apples. Furthermore, the model is trained to distinguish between Apples and Bananas based on their characteristics, for example, based on color, shape, and weight. After the best model is obtained, this model will be applied to fruit basket B, where the fruits in it have not been labeled. As a result, the fruits in basket B will then be labeled according to the model’s predictions, whether they are bananas or apples.
Examples of applying unsupervised learning. Meanwhile, in unsupervised learning, we classify baskets where the fruits have not been labeled. For example, we apply a clustering algorithm such as K-means clustering to fruit basket B. We only need to determine the number of classes (or in this case how many types of fruit) are in our basket. If we determine there are 2 classes, then the algorithm will separate the fruits in the basket into two clusters, for example, Cluster X and Cluster Y.
Our next task is to label each cluster. For example, Cluster X is an Apple, and Cluster Y is a Banana. Face recognition technology or face detection is an example of the use of machine learning that is commonly used. Facebook uses it to detect the faces of our friends in the photos we upload to make it easier for us to tag them in photos. Apple also uses it as one of the most secure iPhone unlock methods. Using this technology, the Xiaomi cellphone camera can also guess the gender and age of the face in front of it.