Machine reasoning as a service
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for responding to a query. In some implementations, a computer obtains a query. The computer determines a meaning for each term in the query. The computer determines user data for the user that submitted the query. The computer identifies one or more ontologies based on the meanings for at least some of the terms. The computer identifies a knowledge graph based on the identified ontologies and the user data. The computer generates a response to the query by traversing a path of the identified knowledge graph to identify items in the knowledge graph based on the determined meaning for each of the terms. The computer generates path data that represents the path taken by the computer through the identified knowledge graph. The computer provides the generated response and the path data to the client device.
1. A computer-implemented method performed by one or more computers comprising:
accessing, by the one or more computers, a knowledge graph based on data indicative of a user associated with a client device from which a query is received;
traversing, by the one or more computers, a plurality of nodes and a plurality of edges in the accessed knowledge graph using data representing the received query, wherein the traversing comprises traversing a path including (i) a subset of the nodes that are associated with words corresponding to one or more words of the query and (ii) a subset of the edges that are associated with words corresponding to one or more words of the query;
generating, by the one or more computers, a response to the query using data identified from the traversal of the accessed knowledge graph, wherein the generating comprises generating results of the traversed path, the results comprise (i) one or more nodes in the accessed knowledge graph where the traversal ended, (ii) data indicative of confidences of correspondences between the words of the query and the words associated with the nodes of the subset of nodes or the edges of the subset of edges; and
providing, by the one or more computers and to the client device, the generated response.
2. The computer-implemented method of claim 1 , wherein the received query comprises text requesting information for a medical topic and the data indicative of the user comprises identification information of the user.
3. The computer-implemented method of claim 2 , wherein the identification information of the user comprises at least one of (i) a current location of the client device, (ii) a current job location of the user, and (iii) a current job position of the user.
4. The computer-implemented method of claim 2 , wherein accessing the knowledge graph based on the data indicative of the user comprises:
selecting, by the one or more computers, a subset of ontologies from one or more ontologies of an ontology knowledge base and at least some of the information for the medical topic of the query, the ontology knowledge base storing a plurality of ontologies, each ontology in the plurality of ontologies containing historical medical data stored in a structural manner,
wherein the one or more ontologies are accessed using the data indicative of the user.
5. The computer-implemented method of claim 4 , further comprising:
identifying, by the one or more computers, a particular knowledge graph from a plurality of knowledge graphs based on data selected in the subset of ontologies, wherein the nodes of each knowledge graph represent corresponding entities and wherein the edges of each knowledge graph represent relationships between the entities; and
providing, by the one or more computers, the identified knowledge graph to an inference engine for traversal,
wherein accessing the knowledge graph based on data indicative of the user comprises accessing the identified knowledge graph.
6. The computer-implemented method of claim 2 , wherein providing the generated response to the client device comprises providing, by the one or more computers and to the client device, (i) graphical illustrations representing the traversed path taken by the one or more computers through the accessed knowledge graph and (ii) the results of the traversed path.
7. A system comprising:
one or more computers and one or more storage devices storing instructions that are configured, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
accessing, by the one or more computers, a knowledge graph based on data indicative of a user associated with a client device from which a query is received;
traversing, by the one or more computers, a plurality of nodes and a plurality of edges in the accessed knowledge graph using data representing the received query, wherein the traversing comprises traversing a path including (i) a subset of the nodes that are associated with words corresponding to one or more words of the query and (ii) a subset of the edges that are associated with words corresponding to one or more words of the query;
generating, by the one or more computers, a response to the query using data identified from the traversal of the accessed knowledge graph, wherein the generating comprises generating results of the traversed path, the results comprise (i) one or more nodes in the accessed knowledge graph where the traversal ended, (ii) data indicative of confidences of correspondences between the words of the query and the words associated with the nodes of the subset of nodes or the edges of the subset of edges; and
providing, by the one or more computers and to the client device, the generated response.
8. The system of claim 7 , wherein the received query comprises text requesting information for a medical topic and the data indicative of the user comprises identification information of the user.
9. The system of claim 8 , wherein the identification information of the user comprises at least one of (i) a current location of the client device, (ii) a current job location of the user, and (iii) a current job position of the user.
10. The system of claim 8 , wherein accessing the knowledge graph based on the data indicative of the user comprises:
selecting, by the one or more computers, a subset of ontologies from one or more ontologies of an ontology knowledge base and at least some of the information for the medical topic of the query, the ontology knowledge base storing a plurality of ontologies, each ontology in the plurality of ontologies containing historical medical data stored in a structural manner,
wherein the one or more ontologies are accessed using the data indicative of the user.
11. The system of claim 10 , further comprising:
identifying, by the one or more computers, a particular knowledge graph from a plurality of knowledge graphs based on data selected in the subset of ontologies, wherein the nodes of each knowledge graph represent corresponding entities and wherein the edges of each knowledge graph represent relationships between the entities; and
providing, by the one or more computers, the identified knowledge graph to an inference engine for traversal,
wherein accessing the knowledge graph based on data indicative of the user comprises accessing the identified knowledge graph.
12. The system of claim 7 , wherein providing the generated response to the client device comprises providing, by the one or more computers and to the client device, (i) graphical illustrations representing the traversed path taken by the one or more computers through the accessed knowledge graph and (ii) the results of the traversed path.
13. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:
accessing, by the one or more computers, a knowledge graph based on data indicative of a user associated with a client device from which a query is received;
traversing, by the one or more computers, a plurality of nodes and a plurality of edges in the accessed knowledge graph using data representing the received query, wherein the traversing comprises traversing a path including (i) a subset of the nodes that are associated with words corresponding to one or more words of the query and (ii) a subset of the edges that are associated with words corresponding to one or more words of the query;
generating, by the one or more computers, a response to the query using data identified from the traversal of the accessed knowledge graph, wherein the generating comprises generating results of the traversed path, the results comprise (i) one or more nodes in the accessed knowledge graph where the traversal ended, (ii) data indicative of confidences of correspondences between the words of the query and the words associated with the nodes of the subset of nodes or the edges of the subset of edges; and
providing, by the one or more computers and to the client device, the generated response.
14. The non-transitory computer-readable medium of claim 13 , wherein the received query comprises text requesting information for a medical topic and the data indicative of the user comprises identification information of the user.
15. The non-transitory computer-readable medium of claim 14 , wherein the identification information of the user comprises at least one of (i) a current location of the client device, (ii) a current job location of the user, and (iii) a current job position of the user.
16. The non-transitory computer-readable medium of claim 14 , wherein accessing the knowledge graph based on the data indicative of the user comprises:
selecting, by the one or more computers, a subset of ontologies from one or more ontologies of an ontology knowledge base and at least some of the information for the medical topic of the query, the ontology knowledge base storing a plurality of ontologies, each ontology in the plurality of ontologies containing historical medical data stored in a structural manner,
wherein the one or more ontologies are accessed using the data indicative of the user.