We illustrate that Web searches can often be utilized to generate background text for use with text classification. This is the case because there are frequently many pages on the...
Ordinal classification is a form of multi-class classification where there is an inherent ordering between the classes, but not a meaningful numeric difference between them. Althou...
Appropriate selection of the kernel function, which implicitly defines the feature space of an algorithm, has a crucial role in the success of kernel methods. In this paper, we co...
Text classification using positive and unlabeled data refers to the problem of building text classifier using positive documents (P) of one class and unlabeled documents (U) of man...
Classical data mining algorithms implicitly assume complete access to all data, either in centralized or federated form. However, privacy and security concerns often prevent sharin...