How Complex Is Your Classification Problem?

@article{Lorena2019HowCI,
  title={How Complex Is Your Classification Problem?},
  author={Ana Carolina Lorena and L. P. F. Garcia and Jens Lehmann and M. D. Souto and T. Ho},
  journal={ACM Computing Surveys (CSUR)},
  year={2019},
  volume={52},
  pages={1 - 34}
}
Characteristics extracted from the training datasets of classification problems have proven to be effective predictors in a number of meta-analyses. Among them, measures of classification complexity can be used to estimate the difficulty in separating the data points into their expected classes. Descriptors of the spatial distribution of the data and estimates of the shape and size of the decision boundary are among the known measures for this characterization. This information can support the… Expand
A Many-Objective optimization Approach for Complexity-based Data set Generation
Redundancy and Complexity Metrics for Big Data Classification: Towards Smart Data
Revisiting Feature Selection with Data Complexity
Data Separability for Neural Network Classifiers and the Development of a Separability Index
A Literature Survey and Empirical Study of Meta-Learning for Classifier Selection
MFE: Towards reproducible meta-feature extraction
...
1
2
3
4
...

References

SHOWING 1-10 OF 24 REFERENCES
An instance level analysis of data complexity
Using Complexity Measures to Evolve Synthetic Classification Datasets
A meta-learning framework for pattern classification by means of data complexity measures
Domains of competence of the semi-naive Bayesian network classifiers
Data Characterization for Effective Prototype Selection
A Set of Complexity Measures Designed for Applying Meta-Learning to Instance Selection
Selecting Discrete and Continuous Features Based on Neighborhood Decision Error Minimization
Dynamic ensemble selection VS K-NN: Why and when dynamic selection obtains higher classification performance?
Measurement of data complexity for classification problems with unbalanced data
...
1
2
3
...