PROGRESSIVE LEARNING - UNIVERSAL CLASSIFIER
To integrate the progressive learning technique with the universal classifier, thereby achieving human-learning-inspired progressively learning universally generic classifier. The resulting new classifier can be used for binary, multi-class and multi-label classification problems with dynamic introduction of new classes.
Identification of classification type
Estimating the number of target labels corresponding to each input sample
Identifying each of the associated target labels.
Binary: Diabetes, Ionosphere
Multi-class: Balance-scale, Satellite image, Digits
Multi-label: Scene, Yeast, Corel5k, Medical
Dataset specifications are given below:
The algorithm, code and datasets will be uploaded soon.
The experimental results will be updated soon.