Study on Using Scenarios of Linear and Nonlinear Classifiers
Yan Tianyang · 2020
Classifier plays a crucial role in data mining by establishing a function or model based on existing data, which can then map all the data in the database to one of a given category to help with prediction. However, as the known data have various problems, such as errors in samples and how to define categories accurately, which will have a huge impact on the accuracy and temporal cost of its results, it is very important to select the optimal classifier based on the known samples. Before selecting a classifier, a detailed understanding of the advantages and disadvantages of each classifier can produce twice the result with half the effort. In this paper, classifiers are divided into linear classifiers and non-linear classifiers, and some typical linear classifiers and nonlinear classifiers are listed, comparing their advantages and disadvantages via their classification principles. This helps to select the most appropriate classifier and improved classifier.