Data mining using /spl Mscr/ℒ/spl Cscr/++ a machine learning library in C++
Ron Kohavi, Daniel A. Sommerfield, James Dougherty · Proceedings - International Conference on Tools with Artificial Intelligence, TAI · 2005
Data mining algorithms including machine learning, statistical analysis, and pattern recognition techniques can greatly improve our understanding of data warehouses that are now becoming more widespread. In this paper, we focus on classification algorithms and review the need for multiple classification algorithms. We describe a system called /spl Mscr//spl Lscr//spl Cscr/++ which was designed to help choose the appropriate classification algorithm for a given dataset by making it easy to compare the utility of different algorithms on a specific dataset of interest. /spl Mscr//spl Lscr//spl Cscr/++ not only provides a work-bench for such comparisons, but also provides a library of C++ classes to aid in the development of new algorithms, especially hybrid algorithms and multi-strategy algorithms. Such algorithms are generally hard to code from scratch. We discuss design issues, interfaces to other programs, and visualization of the resulting classifiers.
Cited by 145 works. doi:10.1109/tai.1996.560457
Authors
References
- Computers and Intractability: A Guide to the Theory of NP-Completeness
- Learning Internal Representations by Error Propagation
- C4.5: Programs for Machine Learning
- Induction of Decision Trees
- Bagging Predictors
- Pattern Classification and Scene Analysis
- A Theory of the Learnable
- Introduction to the theory of neural computation
- Stacked generalization
- Neural Networks and the Bias/Variance Dilemma