새 버전 유지
Weka is a collection of machine
learning algorithms for solving
real-world data mining issues. The algorithms can
either be applied directly to a
data set or called from your own Java
code.
The application contains the tools you'll need for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also appropriate for developing new machine learning schemes.
Key features include:
- Machine learning.
- Data mining.
- Pre-processing.
- Classification.
- Regression.
- Clustering.
- Association rules.
- Attribute selection.
- Experiments.
- Workflow.
- Visualization.
Weka's collection of algorithms range from those that handle data pre-processing to modeling. It's core data mining algorithms include regression, clustering and classification.
Although Weka has a full suite of algorithms for data analysis, it has been built to handle data as single flat files. Subsequently, it does not handle multi-relational mining and sequence modeling.
Overall, Weka is a good data mining tool with a comprehensive suite of algorithms. The interface is OK, although with four to choose from, each with their own strengths, it can be awkward to choose which to work with, unless you have a thorough knowledge of the application to begin with.