Software as a Service (SaaS) based Machine Learning for Digital Image Recognition

Abstract

Nowadays, the machine learning method and algorithms are varied with different capabilities and tasks. It is almost impossible to understand the algorithm in detail or to determine which method is appropriate for certain applications. For these reasons, the application system operates in cloud system according to Software as a Service (SaaS) method is proposed; therefore the system is accessible for multiusers with open source data mining. Wide range of algorithms in Waikato Environment for Knowledge Analysis (WEKA) machine learning are considered, for instance support vector machine (SVM), K-Nearest Neighbor (KNN), Naïve Bayes, C4.5 Decision Tree, Logistic Regression and Random Forest methods. The application facilitates the image recognition researcher of using SaaS method, due to the flexibility in purpose of research, such as search in algorithm analysis, optimal training results in digital image recognition and the implementation of application system. In addition, the system application can be accessed anytime without installation process, but through web browsing systems.