Relation visualizing apparatus, relation visualizing method and program

1. A relation visualizing apparatus comprising:

a memory; and

a processor configured to:

receive, as an input, filter data and dialog content data representing a content of a dialog between a plurality of persons speaking about a plurality of predetermined topic items;

store at least a part of the dialog content data in a dialog content database, wherein the dialog content database stores the at least a part of the dialog content data indexed by the plurality of predetermined topic items for data retrieval;

extract, based on the filter data, a set of filtered dialog content data from the dialog content database, wherein the filter data specifies at least a type of data stored in the dialog content database and at least a predetermined topic item of the plurality of topic items for filtering;

generate, using the set of filtered dialog content data, graph data, wherein the graph data includes nodes and edges, wherein each node represents the predetermined topic item of the plurality of predetermined topic items, each edge represents a person of the plurality of persons in the dialog, each edge connects two nodes of the plurality of nodes, and the nodes and edges collectively indicate a relationship among the plurality of predetermined topic items and the plurality of persons speaking about respective predetermined topic items of the plurality of predetermined topic items; and

display, based on the generated graph data, a graph, wherein the graph visually illustrates the relationship among the plurality of predetermined topic items and the plurality of persons speaking about respective predetermined topic items of the plurality of predetermined topic items.

2. The relation visualizing apparatus according to claim 1 , wherein

the plurality of predetermined topic items is at least a part of a predetermined topic item group of a plurality of predetermined topic item groups, and

the processor generates a Sankey diagram as a graph, wherein the graph includes a hierarchy of nodes and edges, the graph uses the predetermined topic item group as the hierarchy of nodes and edges, the plurality of predetermined topic items is at least a part of the predetermined topic item group, and the nodes in the hierarchy represent the topic item group.

3. The relation visualizing apparatus according to claim 1 , wherein the processor is further configured to:

filter the dialog content data using hosting information data and utter information data, the host information data representing information of at least one of a hosting place and a hosting date of the dialog, and the utterer information data representing attribute information of the utterer; and

generate, based on the filtered dialog content data as input, the graph.

4. The relation visualizing apparatus according to claim 3 , wherein

the hosting information data and the utterer information data are associated with the dialog content data, and

the filter operation further comprises, in a predetermined sequence:

filtering the hosting information data thereby filtering the dialog content data corresponding to the hosting information data,

filtering the utterer information data thereby filtering the dialog content data corresponding to the utterer information data, and

filtering the dialog content data.

5. A relation visualizing method executed by a computer including a memory and a processor, the relation visualizing method comprising:

inputting filter data and dialog content data representing a content of a dialog between a plurality of persons speaking about to a plurality of predetermined topic items;

storing at least a part of the dialog content data in a dialog content database, wherein the dialog content database stores the at least a part of the dialog content data indexed by the plurality of predetermined topic items for data retrieval;

extracting, based on the filter data, a set of filtered dialog content data from the dialog content database, wherein the filter data specifies at least a type of data stored in the dialog content database and at least a predetermined topic item of the plurality of topic items for filtering;

generating, using the set of filtered dialog content data, a graph data, wherein the graph data includes nodes and edges, wherein each node represents the predetermined topic item of the plurality of predetermined topic items, each edge represents a person of the plurality of persons in the dialog, each edge connects two nodes of the plurality of nodes, and the nodes and edges collectively indicate a relationship among the plurality of predetermined topic items and the plurality of persons speaking about respective predetermined topic items of the plurality of predetermined topic items; and

displaying, based on the generated graph data, a graph, wherein the graph visually illustrates the relationship among the plurality of predetermined topic items and the plurality of persons speaking about respective predetermined topic items of the plurality of predetermined topic items.

6. A non-transitory computer-readable recording medium having stored therein a program causing a computer to operate as the relation visualizing apparatus described in claim 1 .