Clustering method and apparatus using ranking-based network embedding
AssigneeElectronics and Telecommunications Research Institute
InventorsYeo Chan Yoon, Hyeon Jin Kim, Soo Myung Park, Hyung Keun Jee
Life of this patent
Estimating the term…
Abstract
A clustering method includes configuring a network with clustering target objects; collecting significance of the clustering target objects; performing network embedding for outputting a set of vectors representing neighboring objects of the clustering target objects constituting the network using a neural network; and performing clustering on the clustering target objects using the set of vectors and information on each of the clustering target objects, wherein the neural network is trained so that neighboring objects having high significance are output with higher probability.
Claims14
A clustering method comprising:
- configuring a network with clustering target objects;
- collecting significance of each of the clustering target objects;
- outputting a vector of neighboring clustering target objects of each of the clustering target objects constituting the network using a neural network;
- and
- performing clustering on each of the clustering target objects using similarity and significance of each of the clustering target objects and the neighboring clustering target objects belonging to the vector of the neighboring clustering target objects of each of the clustering target objects,
- wherein the neural network is trained so that neighboring clustering target objects having high significance are output with higher probability,
- wherein the network is defined as G=<V, E, W>, where V are the clustering target objects belonging to the network, E are edges representing binary relations of pairs of the clustering target objects belonging to the network, and W:E→+ are real value weight mappings for edges e (e ∈ E).
A clustering apparatus comprising at least one processor and a memory storing at least one instruction executable by the at least one processor, wherein when executed by the at least one processor, the at least one instruction is configured the at least one processor to:
- configure a network with clustering target objects;
- collect significance of each of the clustering target objects;
- output a vector of neighboring clustering target objects of each of the clustering target objects constituting the network using a neural network;
- and
- perform clustering on each of the clustering target objects using similarity and significance of each of the clustering target objects and the neighboring clustering target objects belonging to the vector of the neighboring clustering target objects of each of the clustering target objects,
- wherein the neural network is trained so that neighboring clustering target objects having high significance are output with higher probability,
- wherein the network is defined as G=<V, E, W>, where V are the clustering target objects belonging to the network, E are edges representing binary relations of pairs of the clustering target objects belonging to the network, and W:E→+ are real value weight mappings for edges e (e ∈ E).
Dependencies are inferred from the claim text.