How to Use Massive Online Analysis in Your Code
Massive Online Analysis is a data mining, open-source framework that exploits algorithms to generate a continuous stream of useful, informative data. The framework is written in Java. You integrate MOA into code to create real and synthetic data streams. Advanced end users, developers, scientists and researchers running a Java Swing application to study, compare, evaluate and apply a measure of artificial intelligence to projects use MOA. The framework has built-in classification and clustering capabilities for research purposes as well. While the individual approach to deploying MOA varies, a few essential pieces help you use it effectively.
Other People Are Reading
Instructions
-
-
1
Compile a suitable Java Swing application. Use of MOA requires the latest release of Java SE platform. Install a full Java Development Kit (JDK) because it's necessary for compiling compatible programs. The platform you choose to run MOA must be equipped with a suitable Java virtual machine and well-developed support libraries.
-
2
Modify an existing program or create one from scratch in an application such as NetBeans, which is a free, open-source Integrated Development Environment recommended by Oracle, or a favored alternate IDE. Watch out for syntax errors as you build or modify code. Go back and fix code that concludes with the output statement, "Build Failed." Getting the most from MOA depends on creating a reliable foundation for the framework.
-
-
3
Choose between relying on a graphical user interface or command line interface. Both options are supported by MOA. Since MOA can define probability and use a sigmoid function, select the type of interface that suits the goals of your advanced end users.
-
4
Incorporate support for MOA's data stream environment into your code. Data will be traveling at high-speed to produce an algorithm -- therefore your code should be efficient and precise. Eliminate extraneous operations that impact the speed of the application or its graphic interface. Create smooth running code and use simple strategies, such as writing new classifiers to extend the functioning of MOA, for example.
-
1
