Capturing organization specificities with embeddings in a model for a multi-tenant database system
For a multi-tenant database accessible by a plurality of separate organizations, a system is provided for capturing organization specificities in a model for the multi-tenant database. The system includes a neural network. The system is configured to: receive an organization encoding for one or more separate organizations making previous search queries into the multi-tenant database; generate a vector matrix from the organization encoding to embed organization specificities for training a model of the neural network; and using the vector matrix, train the model of the neural network for processing a present search query into the multi-tenant database. In some embodiments, the model of the neural network is global across the separate organizations accessing the database.
1. For a multi-tenant database accessible by a plurality of separate organizations, a system comprising:
a communication interface configured to receive an organization encoding for one or more separate organizations making previous search queries into the multi-tenant database;
a memory storing parameters of a neural network and a plurality of processor-executable instructions,
wherein the organization encoding is based on the previous search queries, and
wherein a total number of previous search queries made by each organization orders the organization encoding; and
one or more processors executing the plurality of processor-executable instructions to:
generate a vector matrix from the organization encoding to embed organization specificities for training a model of the neural network; and
by using the vector matrix, train the model of the neural network for predicting results in response to a present search query into the multi-tenant database.
2. The system of claim 1 , wherein the model of the neural network is global across the separate organizations accessing the multi-tenant database.
3. The system of claim 1 , wherein the organization encoding comprises a mapping of one of the separate organizations accessing the multi-tenant database to the entities most relevant to that organization.
4. The system of claim 1 , wherein, after training, the model of the neural network predicts an entity that is most relevant to a user making the present search query based on the organization to which the user belongs.
5. The system of claim 1 , wherein the model of the neural network is capable of being loaded at a time when the present query is made into the multi-tenant database.
6. For a multi-tenant database accessible by a plurality of separate organizations, a method performed by one or more processors executing machine executable code, the method comprising:
receiving, via a communication interface, an organization encoding for one or more separate organizations making previous search queries into the multi-tenant database, wherein the organization encoding is based on the previous search queries, wherein a total number of previous search queries made by each organization orders the organization encoding into feature representations;
generating, by a processor, a vector matrix from the organization encoding to embed organization specificities for training a model of a neural network; and
using the vector matrix, training the model of the neural network for predicting results in response to a present search query into the multi-tenant database.
7. The method of claim 6 , wherein the model of the neural network is global across the separate organizations accessing the multi-tenant database.
8. The method of claim 6 , wherein the organization encoding comprises a mapping of one of the separate organizations accessing the multi-tenant database to entities most relevant to that organization.
9. The method of claim 6 , wherein, after training, the model of the neural network predicts an entity that is most relevant to a user making the present search query based on the organization to which the user belongs.
10. The method of claim 6 , wherein the model of the neural network is capable of being loaded at a time when the present query is made into the multi-tenant database.
11. For a multi-tenant database accessible by a plurality of separate organizations, a non-transitory machine readable medium having stored thereon instructions for performing a method comprising machine executable code which when executed by at least one machine, causes the machine to:
receive an organization encoding for one or more separate organizations making previous search queries into the multi-tenant database, wherein the organization encoding is based on the previous search queries, wherein a total number of previous search queries made by each organization orders the organization encoding;
generate a vector matrix from the organization encoding to embed organization specificities for training a model of a neural network; and
using the vector matrix, train the model of the neural network for predicting results in response to a present search query into the multi-tenant database.
12. The non-transitory machine readable medium of claim 11 , wherein the model of the neural network is global across the separate organizations accessing the multi-tenant database.
13. The non-transitory machine readable medium of claim 11 , wherein the organization encoding comprises a mapping of one of the separate organizations accessing the multi-tenant database to the entities most relevant to that organization.
14. The non-transitory machine readable medium of claim 11 , wherein, after training, the model of the neural network predicts an entity that is most relevant to a user making the present search query based on the organization to which the user belongs.
15. The non-transitory machine readable medium of claim 11 , wherein the model of the neural network is capable of being loaded at a time when the present query is made into the multi-tenant database.