Non-transitory computer readable medium, information processing apparatus, and information processing method
A non-transitory computer readable medium storing a program causing a computer to execute a process for information processing, the process including: obtaining input data that represents attributes of each subject among subjects; generating a bipartite network which includes the subjects and the attributes included in the input data as nodes and in which a node that corresponds to each subject among the subjects are connected to nodes that correspond to the attributes of the subject via links; and performing clustering of a group of the nodes by performing iterative calculations of a stochastic process of transitions between the nodes via the links in the bipartite network.
1. A non-transitory computer readable medium storing a program causing a computer to execute a process for information processing, the process comprising:
obtaining input data that represents attributes of each subject among subjects;
generating a bipartite network which includes the subjects and the attributes included in the input data as nodes and in which a node that corresponds to each subject among the subjects is connected to a plurality of nodes that correspond to the attributes of the subject via links, such that a node of an attribute is not directly linked to another node of an attribute and is indirectly linked to the another node via a node of a common subject;
performing clustering of a group of the nodes by performing iterative calculations of a stochastic process of transitions between the nodes via the links in the bipartite network, wherein the iterative calculations of the stochastic process include calculating a probability value of transitions from one node of an attribute to another node of an attribute via links in the bipartite network;
after the clustering is completed, adding a new attribute feature to each subject based on a result of the clustering; and
analyzing the subjects using the new attribute added to each subject.
2. The non-transitory computer readable medium according to claim 1 , the process further comprising
generating, for each subject among the subjects, subject cluster feature data that represents features of the subject from a clustering viewpoint by using a result of the performed clustering.
3. The non-transitory computer readable medium according to claim 2 , the process further comprising
generating, for each attribute among the attributes, attribute cluster feature data that represents features of the attribute from a clustering viewpoint by using a result of the performed clustering.
4. The non-transitory computer readable medium according to claim 3 , the process further comprising
generating, in a case where data that represents attributes of a new subject not included in the input data is input, feature data that represents features of the new subject from a clustering viewpoint by using the attribute cluster feature data of the attributes generated by using the result of the clustering performed for the input data.
5. The non-transitory computer readable medium according to claim 4 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
6. The non-transitory computer readable medium according to claim 4 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
7. The non-transitory computer readable medium according to claim 3 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
8. The non-transitory computer readable medium according to claim 3 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
9. The non-transitory computer readable medium according to claim 2 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
10. The non-transitory computer readable medium according to claim 2 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
11. The non-transitory computer readable medium according to claim 1 , the process further comprising
generating, for each attribute among the attributes, attribute cluster feature data that represents features of the attribute from a clustering viewpoint by using a result of the performed clustering.
12. The non-transitory computer readable medium according to claim 11 , the process further comprising
generating, in a case where data that represents attributes of a new subject not included in the input data is input, feature data that represents features of the new subject from a clustering viewpoint by using the attribute cluster feature data of the attributes generated by using the result of the clustering performed for the input data.
13. The non-transitory computer readable medium according to claim 12 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
14. The non-transitory computer readable medium according to claim 12 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
15. The non-transitory computer readable medium according to claim 11 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
16. The non-transitory computer readable medium according to claim 11 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
17. The non-transitory computer readable medium according to claim 1 , wherein
in the performing of clustering, initial values of the nodes in the stochastic process are set so as to satisfy a condition that a sum total of initial values of a group of nodes that correspond to the subjects is substantially equal to a sum total of initial values of nodes that correspond to the attributes, and the set initial values of the nodes are used to perform the iterative calculations.
18. The non-transitory computer readable medium according to claim 1 , wherein
in the performing of clustering, as the iterative calculations of the stochastic process, calculations are performed by using a continuous-time model without using a discrete-time model.
19. The non-transitory computer readable medium according to claim 1 , wherein
the subjects include documents, and
the attributes include words appearing in the documents.
20. An information processing apparatus comprising:
a processor that is programmed to:
obtain input data that represents attributes of each subject among subjects;
generate a bipartite network which includes the subjects and the attributes included in the input data as nodes and in which a node that corresponds to each subject among the subjects is connected to a plurality of nodes that correspond to the attributes of the subject via links, such that a node of an attribute is not directly linked to another node of an attribute and is indirectly linked to the another node via a node of a common subject;
perform clustering of a group of the nodes by performing iterative calculations of a stochastic process of transitions between the nodes via the links in the bipartite network, wherein the iterative calculations of the stochastic process include calculating a probability value of transitions from one node of an attribute to another node of an attribute via links in the bipartite network;
after the clustering is completed, add a new attribute feature to each subject based on a result of the clustering; and
analyze the subjects using the new attribute added to each subject.
21. An information processing method comprising:
obtaining input data that represents attributes of each subject among subjects;
generating a bipartite network which includes the subjects and the attributes included in the input data as nodes and in which a node that corresponds to each subject among the subjects is connected to a plurality of nodes that correspond to the attributes of the subject via links, such that a node of an attribute is not directly linked to another node of an attribute and is indirectly linked to the another node via a node of a common subject;
performing clustering of a group of the nodes by performing iterative calculations of a stochastic process of transitions between the nodes via the links in the bipartite network, wherein the iterative calculations of the stochastic process include calculating a probability value of transitions from one node of an attribute to another node of an attribute via links in the bipartite network;
after the clustering is completed, adding a new attribute feature to each subject based on a result of the clustering; and
analyzing the subjects using the new attribute added to each subject.