Classification
The task of monitoring an entry into one of the predefined categories.
Classification (classification) is the task of a surveillance learning (supervised learning) aimed at assigning an input into one of the predefined, final categories. In order to be trained by the model, each input needs a set of training data given with the right class tag; the model learns the relationship between the characteristics of the input and the right class, and then predicts this relationship by generalizing the new inputs it never seen. Classification can be dual or multi-class; logistics regression, decision trees, support vector machines and deep nerve networks can be realized with many different algorithm families.
Classification is one of the most common forms of application of machine learning, and it goes to us in almost any field of life: spam detection of email clients, fraud detection of banks in credit card transactions, recognition of objects in a photo of image recognition systems, determining whether medical diagnostic systems are a sign of disease in a x-ray, and a positive or negative problem of a text of emotions analysis systems. The success of the model is usually carefully evaluated with metrics such as accuracy, precision, sensitivity and F1 score, especially when classes are unbalanced.
Classification is one of the most common forms of application of machine learning, and it goes to us in almost any field of life: spam detection of email clients, fraud detection of banks in credit card transactions, recognition of objects in a photo of image recognition systems, determining whether medical diagnostic systems are a sign of disease in a x-ray, and a positive or negative problem of a text of emotions analysis systems. The success of the model is usually carefully evaluated with metrics such as accuracy, precision, sensitivity and F1 score, especially when classes are unbalanced.