A Comparison of Machine Learning Algorithms on Handwritten Digit Recognition
摘要
Handwritten digit recognition is a crucial issue in the study of computer vision and machine learning, with applications ranging from postal code identification in mail sorting to digitizing old documents. This paper weighs up the accomplishments and execution of numerous classifier calculations on database of manually written digits. A interesting zone of design acknowledgment is the advancement of Handwritten character recognition (HCR). For character recognition, a few techniques have been proposed. It too centers on troubles and obstacles included in extraction and acknowledgment of handwritten digits. This errand is performed utilizing calculations like SVM (Support Neural Systems), Decision Tree, Random Forest, artificial Neural Network, K-Nearest Neighbor (KNN). Indeed, so, there are sufficient inquire about and papers that portray the strategies for changing paper record content into machine-readable frame. Character recognition innovation may be vital within the close future in arrange to prepare and filter existing paper documents in arrange to set up a paperless environment. This survey paper offers a careful investigation of the handwritten character recognition field.