Incremental Learning: Areas and Methods – A Survey

Abstract

ABSTRACT

Incremental learning from news events

“…Chou and Chen [8] proposed an incremental probabilistic latent semantic indexing (IPLSI) algorithm for event detection and tracking, which alleviates the threshold-dependency problem and simultaneously maintains the continuity of the latent semantics. We can refer to [38,39] for a more comprehensive review of incremental learning algorithms.…”

Characteristic matrixes-based knowledge reduction in dynamic covering decision information systems

“…Many researchers [2][3][4]7,8,15,16,18,[20][21][22][23][24][25][26][27][28][33][34][35]41,42,51,52,[54][55][56] have focused on knowledge reduction of dynamic information systems by using incremental learning methods [1,[11][12][13]29,45], which handles the problem of learning new knowledge while maintaining existing knowledge. For instance, Chen et al [2][3][4] constructed approximations of sets under dynamic maintenance environments such as coarsening and refining attribute values and varying attribute sets, which provides an effective approach to attribute reduction of dynamic information systems.…”

Comparative Analysis of Different Techniques for Novel Class Detection

“…As per [8] the learning to be one that is: Capable to learn and update with every new data (labeled or unlabeled), Will use and exploit the knowledge in further learning, Will not rely on the previously learned knowledge, Will generate a new class as required and take decisions to merge or divide them as well Decision tree that provide the solution for handling novel class detection problem. ID3 is very useful learning algorithm for decision tree.C5.0 algorithm improves the performance of tree using boosting.…”