Collective Classification

Abstract

Class: Is used to name a collection of network nodes that reasonably might be grouped together. Commonly encountered classes have simple unique textual descriptions - labels Classification: Means assigning network nodes into predefined classes - giving them class labels (labeling). Having data about nodes that are already labeled (known nodes), we can derive some knowledge - train a classifier. A trained classifier is able to map (classify) the features of unknown nodes to classes - assign the class labels. For instance, we would like to allocate humans to those with and without disease based on their symptoms and illnesses of their parents. The common classification tasks accomplish either binary or multiclass classification. It means that a given data instance may be assigned with one of two or one of many classes, respectively Features (Attributes):….