Open AccessComparison of classifiers for different data in application of classification
Author(s) -
Huirong Gu,
Jiyuan Jiao
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1994/1/012015
Subject(s) - artificial intelligence , computer science , support vector machine , classifier (uml) , random subspace method , pattern recognition (psychology) , machine learning , random forest , contextual image classification , linear classifier , task (project management) , image (mathematics) , management , economics
The classification task is very important in many application fields, such as image recognition, speech recognition, and text classification. Machine learning and deep learning methods are used as the classifiers in their specific classification tasks. Classical machine learning classifiers, including Random Forest, XGBoost, GMM, and SVM, and deep learning classifiers including CNN and LSTM are compared in this paper to show the different computing characteristics in their specific classification tasks. The comparison results show that the CNN-based classifier performs the best in its own classification, especially the image classification. The results illustrate that the complexity of the classification task may heavily influence the performance of the classifiers. The research in this paper has a reference significance for choosing the right classifier in applying the classification task.