Information processing device, information processing method, and program

Abstract

In an information processing device (500), a training means (511) trains a learning model (600) using graph data that has a plurality of nodes corresponding respectively to a plurality of items of content, and to which attribute data indicating attributes of each of the plurality of nodes is assigned, and relationship data indicating known relationships between nodes that are linked in the graph data. An analysis means (512) uses the trained learning model (600) to perform analysis for identifying optimal content for a keyword entered by a user. A display information generation means (513) generates a graph for showing analysis results obtained from said analysis together with the basis thereof, and also generates display information in which each node constituting the basis in the graph is assigned an icon corresponding to each attribute of the node. The information processing device (500) can be used for decision making relating to healthcare of the user.

Claims

1. A learning means for training a learning model using graph data having multiple nodes corresponding to multiple content items, each of which is assigned attribute data indicating the attributes of each of the multiple nodes, and relationship data which is data indicating known relationships between linked nodes in the graph data, An analysis means that uses the learned model on which the aforementioned learning has been performed to analyze and identify the content best suited to the keywords entered by the user, A display information generation means generates a graph to show the analysis results obtained from the above analysis along with the supporting evidence, and generates display information in which icons corresponding to the individual attributes of each node constituting the supporting evidence are assigned to each node in the graph. An information processing device having

2. The information processing apparatus according to claim 1, wherein the display information generation means acquires, based on the relationship data, data of the portion of the graph data in which the relationship between the analysis result and the basis is established as example data.

3. The information processing apparatus according to claim 2, wherein the display information generation means obtains a classification result by classifying each node included in the example data based on the relationships between nodes and the attributes of the nodes, and generates the graph by transforming the example data so that each node belonging to a group of nodes included in the classification result is arranged in an overlapping state.

4. The information processing apparatus according to claim 3, wherein the display information generation means generates the display information to which icons corresponding to the individual attributes of each node placed at the forefront in the graph are attached.

5. The information processing apparatus according to claim 1, wherein the learning means trains the learning model such that unknown relationships between unlinked nodes in the graph data are derived.

6. The aforementioned learning model is configured as a machine learning model, The information processing apparatus according to claim 1, wherein the aforementioned keyword is a string of characters related to healthcare.

7. A computer-based information processing method, The learning model is trained using graph data which has multiple nodes corresponding to multiple content items, each of which is assigned attribute data indicating the attributes of each of those multiple nodes, and relationship data which indicates the known relationships between the linked nodes in the graph data. Using the aforementioned trained model, an analysis is performed to identify the most suitable content for the keywords entered by the user. An information processing method that generates a graph to show the analysis results obtained from the above analysis along with the supporting evidence, and generates display information in which icons corresponding to the individual attributes of each node constituting the supporting evidence are assigned to each node in the graph.

8. The learning model is trained using graph data which has multiple nodes corresponding to multiple content items, each of which is assigned attribute data indicating the attributes of each of those multiple nodes, and relationship data which indicates the known relationships between the linked nodes in the graph data. Using the aforementioned learning model, an analysis is performed to identify the most suitable content for the keywords entered by the user. A program that causes a computer to generate a graph showing the analysis results obtained from the aforementioned analysis along with the supporting evidence, and to generate display information in the graph by assigning icons corresponding to the individual attributes of each node that constitute the supporting evidence.

Images