Semantic entity search using vector space
A method may include storing a vector space representation of a set of concepts in a semantic ontology, wherein locations of the set of concepts in the vector space are based on weights of a neural network trained using triples from the semantic ontology; presenting a semantic search user interface including: a text input portion to receive a search query from an input device; a suggested search portion; and a search results portion; receiving, from a user computing device, the search query; converting, using at least one processor, the search query into a vector; computing, using the at least one processor, a set of possible concepts related to the search query based on a distance between the vector and other concepts in the vector space representation; and presenting the set of possible concepts related to the search query in the suggested search portion of the semantic search user interface.
1. A method comprising:
presenting a semantic search user interface, the semantic search user interface including:
a text input portion to receive a search query;
a suggested search portion; and
a search results portion;
receiving, from a user computing device, the search query;
computing a set of semantically close concepts related to the search query based on weights of a neural network trained using triples from a semantic ontology;
presenting the set of semantically close concepts related to the search query in the suggested search portion of the semantic search user interface;
receiving, from the user computing device, a selection of a concept in the set of semantically close concepts;
in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept in the semantic ontology; and
presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.
2. The method of claim 1 , wherein computing the set of semantically close concepts related to the search query based on the weights of the neural network trained using the triples from the semantic ontology includes:
executing a lexical correction operation to the search query that results in a modified search query; and
using the modified search query as an input to the neural network.
3. The method of claim 2 , wherein the lexical correction operation is a spell check operation.
4. The method of claim 3 , further comprising:
determining that the modified search query is a direct match to a concept in the semantic ontology; and
based on the determining, presenting the concept in the semantic ontology with the set of semantically close concepts.
5. The method of claim 1 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting representations of levels of data assets data associated with the concept in the semantic ontology.
6. The method of claim 1 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting the concept in the semantic ontology as a node on a graph linked to other nodes, the other nodes based on the concepts semantically linked to the concept in the semantic ontology.
7. The method of claim 1 , wherein computing the set of semantically close concepts related to the search query based on the weights of the neural network trained using the triples from the semantic ontology includes:
converting the search query to a vector.
8. A system comprising:
a processing unit; and
a storage device comprising instructions, which when executed by the processing unit, configure the processing unit to perform operations comprising:
presenting a semantic search user interface, the semantic search user interface including:
a text input portion to receive a search query;
a suggested search portion; and
a search results portion;
receiving, from a user computing device, the search query;
computing a set of semantically close concepts related to the search query based on weights of a neural network trained using triples from a semantic ontology;
presenting the set of semantically close concepts related to the search query in the suggested search portion of the semantic search user interface;
receiving, from the user computing device, a selection of a concept in the set of semantically close concepts;
in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept in the semantic ontology; and
presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.
9. The system of claim 8 , wherein computing the set of semantically close concepts related to the search query based on the weights of the neural network trained using the triples from the semantic ontology includes:
executing a lexical correction operation to the search query that results in a modified search query; and
using the modified search query as an input to the neural network.
10. The system of claim 9 , wherein the lexical correction operation is a spell check operation.
11. The system of claim 10 , wherein the instructions, which when executed by the processing unit, further configure the processing unit to perform operations comprising:
determining that the modified search query is a direct match to a concept in the semantic ontology; and
based on the determining, presenting the concept in the semantic ontology with the set of semantically close concepts.
12. The system of claim 8 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting representations of levels of data assets data associated with the concept in the semantic ontology.
13. The system of claim 8 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting the concept in the semantic ontology as a node on a graph linked to other nodes, the other nodes based on the concepts semantically linked to the concept in the semantic ontology.
14. The system of claim 8 , wherein computing the set of semantically close concepts related to the search query based on the weights of the neural network trained using the triples from the semantic ontology includes:
converting the search query to a vector.
15. A non-transitory computer-readable medium comprising instructions, which when executed by a processing unit, configure the processing unit to perform operations comprising:
presenting a semantic search user interface, the semantic search user interface including:
a text input portion to receive a search query;
a suggested search portion; and
a search results portion;
receiving, from a user computing device, the search query;
computing a set of semantically close concepts related to the search query based on weights of a neural network trained using triples from a semantic ontology;
presenting the set of semantically close concepts related to the search query in the suggested search portion of the semantic search user interface;
receiving, from the user computing device, a selection of a concept in the set of semantically close concepts;
in response to receiving the selection, executing a query to a graph database to retrieve concepts semantically linked to the concept in the semantic ontology; and
presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface.
16. The non-transitory computer-readable medium of claim 15 , wherein computing the set of semantically close concepts related to the search query based on the weights of the neural network trained using the triples from the semantic ontology includes:
executing a lexical correction operation to the search query that results in a modified search query; and
using the modified search query as an input to the neural network.
17. The non-transitory computer-readable medium of claim 16 , wherein the lexical correction operation is a spell check operation.
18. The non-transitory computer-readable medium of claim 17 , wherein the instructions, which when executed by the processing unit, further configure the processing unit to perform operations comprising:
determining that the modified search query is a direct match to a concept in the semantic ontology; and
based on the determining, presenting the concept in the semantic ontology with the set of semantically close concepts.
19. The non-transitory computer-readable medium of claim 15 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting representations of levels of data assets data associated with the concept in the semantic ontology.
20. The non-transitory computer-readable medium of claim 15 , wherein presenting the concepts semantically linked to the concept in the search results portion of the semantic search user interface includes:
presenting the concept in the semantic ontology as a node on a graph linked to other nodes, the other nodes based on the concepts semantically linked to the concept in the semantic ontology.