GENERATING AND MODIFYING ONTOLOGIES FOR MACHINE LEARNING MODELS
A method performed by a machine learning system that involves obtaining a first ontology that includes one or more labels. Each label is associated with a sample that includes text. The ML system is configured to use a particular label to retrieve one or more samples associated with the particular label. The method further involves receiving an identification of a label of a first ontology associated with a first machine learning model to share with a second ontology associated with a second machine learning model and sharing the label and the information with the second ontology. The method further involves training the second machine learning model using the shared information associated with the label.
1 . A method for federated learning of a machine learning model, the method comprising:
obtaining, from a first computing device, a first data training set to train a first machine learning model, the first data training set comprising a first set of entities, each entity of the first set of entities comprising one or more labels including text associated with the first set of entities, wherein each of the one or more labels represent a classification of at least a portion of the text of the entity of the first set of entities into one or more categories;
training, via a machine learning model engine, a first ontology machine learning model using the first data training set;
receiving, from a second computing device different than the first computing device, a model input for the first ontology machine learning model, the model input comprising a plurality of text samples;
inputting the model input to the first ontology machine learning model, the first ontology machine learning model configured to generate one or more labels associated with each of the plurality of text samples of the model input; and
transmitting, to the second computing device, the one or more labels associated with each of the plurality of text samples of the model input generated by the first ontology machine learning model, wherein the second computing device is restricted from accessing the first data training set to maintain a security of the first data training set of the first computing device.
2 . The method of claim 1 , wherein the first data training set is a first ontology comprising a first concept associated with a first portion of the one or more labels of the first data training set and a second concept associated with a second portion of the one or more labels of the first data training set.
3 . The method of claim 2 , wherein the first ontology further comprises at least one sub-concept associated with the first concept, a portion of the first portion of the one or more labels associated with the at least one sub-concept.
4 . The method of claim 1 , wherein the first data training set comprises a plurality of sample text and label pairs, am output of first ontology machine learning model indicating a relationship between the plurality of sample text and label pairs with a corresponding category of the one or more categories.
5 . The method of claim 1 further comprising:
storing the first data training set in a first database, the first database inaccessible by the second computing device.
6 . The method of claim 1 , wherein the first ontology machine learning model is an nth iteration of a machine learning model generated by the machine learning model engine.
7 . The method of claim 6 further comprising:
selecting, from a plurality of iterations of the machine learning model, the nth iteration of the machine learning model.
8 . The method of claim 6 further comprising:
receiving, from the second computing device, an indication of the nth iteration of the machine learning model for use in generating the one or more labels associated with each of the plurality of text samples of the model input.
9 . The method of claim 1 further comprising:
transmitting, to the first computing device, the one or more labels associated with each of the plurality of text samples of the model input generated by the first ontology machine learning model.
10 . The method of claim 1 , wherein the first ontology machine learning model further generates, in response to the model input, one or more of a marked-up version of a document annotated with the one or more labels, a classification of the document, an assessment of the document, or a recommendation associated with the document.
11 . A system for federated learning of a machine learning model, the system comprising:
a processor; and
a computer-readable medium storing instructions, wherein execution of the instructions cause the processor to:
obtain, from a first computing device, a first data training set to train a first machine learning model, the first data training set comprising a first set of entities, each entity of the first set of entities comprising one or more labels including text associated with the first set of entities, wherein each of the one or more labels represent a classification of at least a portion of the text of the entity of the first set of entities into one or more categories;
train, via a machine learning model engine, a first ontology machine learning model using the first data training set;
receive, from a second computing device different than the first computing device, a model input for the first ontology machine learning model, the model input comprising a plurality of text samples;
input the model input to the first ontology machine learning model, the first ontology machine learning model configured to generate one or more labels associated with each of the plurality of text samples of the model input; and
transmit, to the second computing device, the one or more labels associated with each of the plurality of text samples of the model input generated by the first ontology machine learning model, wherein the second computing device is restricted from accessing the first data training set to maintain a security of the first data training set of the first computing device.
12 . The system of claim 11 , wherein the first data training set is a first ontology comprising a first concept associated with a first portion of the one or more labels of the first data training set and a second concept associated with a second portion of the one or more labels of the first data training set.
13 . The system of claim 12 , wherein the first ontology further comprises at least one sub-concept associated with the first concept, a portion of the first portion of the one or more labels associated with the at least one sub-concept.
14 . The system of claim 11 , wherein the first data training set comprises a plurality of sample text and label pairs, am output of first ontology machine learning model indicating a relationship between the plurality of sample text and label pairs with a corresponding category of the one or more categories.
15 . The system of claim 11 , wherein the instructions further cause the processor to store the first data training set in a first database, the first database inaccessible by the second computing device.
16 . The system of claim 11 , wherein the first ontology machine learning model is an nth iteration of a machine learning model generated by the machine learning model engine.
17 . The system of claim 16 , wherein the instructions further cause the processor to select, from a plurality of iterations of the machine learning model, the nth iteration of the machine learning model.
18 . The system of claim 16 , wherein the instructions further cause the processor to receive, from the second computing device, an indication of the nth iteration of the machine learning model for use in generating the one or more labels associated with each of the plurality of text samples of the model input.
19 . The system of claim 11 , wherein the instructions further cause the processor to transmit, to the first computing device, the one or more labels associated with each of the plurality of text samples of the model input generated by the first ontology machine learning model.
20 . The system of claim 11 , wherein the first ontology machine learning model further generates, in response to the model input, one or more of a marked-up version of a document annotated with the one or more labels, a classification of the document, an assessment of the document, or a recommendation associated with the document.