INFORMATION PROCESSING APPARATUS AND NON-TRANSITORY COMPUTER READABLE MEDIUM

1 . An information processing apparatus comprising:

an information collecting unit that collects information about one or more users and information about one or more documents; and

a processor that receives, for processing, the information collected by the information collecting unit and that is configured to,

through execution of a program,

generate a bipartite network in which one or more nodes corresponding to the one or more users are linked to one or more nodes corresponding to the one or more documents,

generate property information including user property information of the one or more users and document property information of the one or more documents,

generate a network with property information by combining the bipartite network with the property information, and

select a recommendation document for a target user by using the network with property information.

2 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to

extract a community from the network with property information, the community indicating an aggregation having dense connection of links, and

select the recommendation document for the target user by using the extracted community.

3 . The information processing apparatus according to claim 1 ,

wherein the user property information includes a domain knowledge keyword of the one or more users, and

wherein the document property information includes at least one of an appearing word, a category tag, and distributed representation, the distributed representation being obtained in a deep learning model.

4 . The information processing apparatus according to claim 2 ,

wherein the user property information includes a domain knowledge keyword of the one or more users, and

wherein the document property information includes at least one of an appearing word, a category tag, and distributed representation, the distributed representation being obtained in a deep learning model.

5 . The information processing apparatus according to claim 3 ,

wherein the processor is configured to

generate the bipartite network as an N×N matrix, N representing a node count of the one or more users plus a node count of the one or more documents,

generate the property information as N×h1 vectors, h1 representing a dimension count of a vector, and

generate the network with property information by combining the N×N matrix with the N×h1 vectors.

6 . The information processing apparatus according to claim 4 ,

wherein the processor is configured to

generate the bipartite network as an N×N matrix, N representing a node count of the one or more users plus a node count of the one or more documents,

generate the property information as N×h1 vectors, h1 representing a dimension count of a vector, and

generate the network with property information by combining the N×N matrix with the N×h1 vectors.

7 . The information processing apparatus according to claim 5 ,

wherein the processor is configured to preprocess the network with property information by using graph convolution networks.

8 . The information processing apparatus according to claim 6 ,

wherein the processor is configured to preprocess the network with property information by using graph convolution networks.

9 . The information processing apparatus according to claim 2 ,

wherein the processor is configured to

calculate one or more first recommendation scores between the target user and one or more recommendation document candidates by using the community, and

select a recommendation document candidate as the recommendation document, the selected recommendation document candidate having a first recommendation score which is relatively high.

10 . The information processing apparatus according to claim 9 ,

wherein the processor is configured to

calculate degrees of similarity between a new document and existing documents, the new document being not included in the one or more documents collected by the information collecting unit, the existing documents being collected by the information collecting unit,

extract a plurality of the existing documents as existing document candidates, the extracted existing documents having the relatively large degrees of similarity,

calculate second recommendation scores between the target user and the existing document candidates, and

calculate a first recommendation score between the target user and the new document by using the second recommendation scores.

11 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

12 . The information processing apparatus according to claim 2 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

13 . The information processing apparatus according to claim 3 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

14 . The information processing apparatus according to claim 4 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

15 . The information processing apparatus according to claim 5 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

16 . The information processing apparatus according to claim 6 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

17 . The information processing apparatus according to claim 7 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

18 . The information processing apparatus according to claim 8 ,

wherein the processor is configured to

feed view information back to the network with property information, the view information indicating whether or not the target user has viewed the recommendation document.

19 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to

form the network with property information by using a weight in accordance with an elapsed time of each piece of the information collected by the information collecting unit.

20 . A non-transitory computer readable medium storing a program causing a computer to execute a process comprising:

collecting information about one or more users and information about one or more documents;

by using the collected information, generating a bipartite network in which one or more nodes corresponding to the one or more users are linked to one or more nodes corresponding to the one or more documents;

generating property information including user property information of the one or more users and document property information of the one or more documents;

generating a network with property information by combining the bipartite network with the property information; and

selecting a recommendation document for a target user by using the network with property information.