Supervised Graph Node Prediction Task
(Redirected from Supervised Graph Node Classification)
Jump to navigation
Jump to search
A Supervised Graph Node Prediction Task is a data-driven graph node-based classification task that is a Graph Node-based Classification Task (with labeled graph nodes).
- AKA: Supervised Graph Node Linking, Supervised Graph Node Prediction.
- Context:
- Input: Labeled Link Training Record.
- It can be solved by a Supervised Link Prediction System (that implements a Supervised Link Prediction algorithm).
- It can be solved by a Supervised Graph Node Classification System (that implements a Supervised Graph Node Classification Algorithm).
- It can range from being a Fully-Supervised Link Prediction Task to being a Semi-Supervised Link Prediction Task.
- It can range from being a Supervised Directed Link Prediction Task to being a Supervised Undirected Link Prediction Task.
- Example(s):
- Counter-Example(s):
- See: Supervised Graph Classification Task.
References
2010
- (Aggarwal & Wang, 2010) ⇒ Charu C. Aggarwal, and Haixun Wang. (2010b). “Graph Data Management and Mining: A Survey of Algorithms and Applications.” In: (Aggarwal & Wang, 2010a) doi:10.1007/978-1-4419-6045-0_2
- Label Propagation. A subset of nodes in a graph are labeled. The task is to learn a model from the labeled nodes and use the model to classify the unlabeled nodes.
- Label Propagation. The concept of label or belief propagation [174, 209, 210] is a fundamental technique which is used in order to leverage graph structure in the context of classification in a number of relational domains. The scenario of label propagation [44] occurs in many applications. As an example, social network analysis is being used as a mean for targeted marketing.