Work income visualization and optimization platform
Provided are systems and methods that rely on machine learning to recommend employment opportunities. In one example, a method may include identifying, via execution of a first machine learning model, income data and spending data of a user, identifying, via execution of a second machine learning model, skill attributes of the user, determining, via execution of a third machine learning model, a recommended job for the user, where the determining comprises inputting the outputs from the first and second machine learning models into the third machine learning model, and displaying, via a user interface, a description of the recommended job.
1 . A computing system comprising:
a processor configured to:
store baseline data collected from a plurality of different data sources in a storage device, the baseline data comprising baseline employment data of a plurality of users;
ingest data records of the plurality of users from the plurality of different data sources;
determine that an event has occurred for a user based on a change to baseline data of the user that is identified from the ingested additional data records;
identify, by a machine learning service in response to the determination of the occurrence of the event, an execution sequence of one or more machine learning models of a software application to be executed and specific data of the baseline data to be input into the one or more machine learning models, the execution sequence specifying input and output relationships between the one or more machine learning models;
in response to the identification, execute the one or more machine learning models in accordance with the execution sequence using the specific data of the baseline data to identify recommendations for the user that provide an improvement for the user based on the determined event; and
display the recommendations via a user interface of the software application.
2 . The computing system of claim 1 , wherein the processor is configured to normalize income data within the ingested data records to remove sensitive details of the plurality of users, and compare the normalized income data to income attributes of the user to identify the recommendations.
3 . The computing system of claim 1 , wherein the processor is configured to ingest the data records via a plurality of application programming interfaces (APIs) of the plurality of different data sources, respectively.
4 . The computing system of claim 1 , wherein the processor is configured to fill-in a template on the user interface of the software application with image data of the recommendations for the user.
5 . The computing system of claim 1 , wherein the processor is further configured to receive an API call, and perform the identification based on one or more model identifiers included in the API call.
6 . The computing system of claim 5 , wherein the one or more machine learning models comprise an ensemble that includes at least two machine learning models that are executed in sequence to predict a recommended job opportunity for the user.
7 . The computing system of claim 5 , wherein the processor is further configured to predict, via execution of the one or more machine learning models, a recommended job opportunity for the user based on a geographical location of the user, income data of the user, and a schedule of the user.
8 . The computing system of claim 1 , wherein the processor is configured to fill in a first field of a template for a job title and a second field of the template for a geographical area based on attributes of the recommendations, and display the identified recommendations via the filled-in template.
9 . A method comprising:
storing baseline data collected from a plurality of different data sources in a storage device, the baseline data comprising baseline employment data of a plurality of users;
ingesting data records of the plurality of users from the plurality of different data sources;
determining that an event has occurred for a user from among the plurality of users based on a change to baseline data of the user that is identified from the ingested additional data records;
identifying, by a machine learning service in response to the determination of the occurrence of the event, an execution sequence of one or more machine learning models of a software application to be executed and specific data of the baseline data to be input into the one or more machine learning models, the execution sequence specifying input and output relationships between the one or more machine learning models;
in response to the identifying, executing the one or more machine learning models in accordance with the execution sequence using the specific data of the baseline data to identify recommendations for the user that provide an improvement for the user based on the determined event, and
displaying the recommendations via a user interface of the software application.
10 . The method of claim 9 , wherein the ingesting further comprises normalizing income data within the ingested data records to remove sensitive details of the plurality of users, and the comparing comprises comparing the normalized income data to income attributes of the user to identify the recommendations.
11 . The method of claim 9 , wherein the ingesting comprises ingesting the data records via a plurality of application programming interfaces (APIs) of the plurality of different data sources, respectively.
12 . The method of claim 9 , wherein the displaying comprises filling-in a template on the user interface of the software application with image data of the recommendations for the user.
13 . The method of claim 9 , wherein the method further comprises receiving an API call, and perform the identification based on one or more model identifiers included in the API call.
14 . The method of claim 13 , wherein the one or more machine learning models comprise an ensemble that includes at least two machine learning models that are executed in sequence to predict a recommended job opportunity for the user.
15 . The method of claim 13 , wherein the comparing comprises predicting, via execution of the one or more machine learning models, a recommended job opportunity for the user based on a geographical location of the user, income data of the user, and a schedule of the user.
16 . The method of claim 9 , wherein the displaying comprises filling in a first field of the template for a job title and a second field of the template for a geographical area based on attributes of the recommendations, and displaying the identified recommendations via the filled-in template.
17 . A non-transitory computer-readable medium comprising instructions which when executed by a processor cause a computer to perform a method comprising:
storing baseline data collected from a plurality of different data sources in a storage device, the baseline data comprising baseline employment data of a plurality of users;
ingesting data records of the plurality of users from the plurality of different data sources;
determining that an event has occurred for a user from among the plurality of users based on a change to baseline data of the user that is identified from the ingested additional data records;
identifying, by a machine learning service in response to the determination of the occurrence of the event, an execution sequence of one or more machine learning models of a software application to be executed and specific data of the baseline data to be input into the one or more machine learning models, the execution sequence specifying input and output relationships between the one or more machine learning models;
in response to the identifying, executing the one or more machine learning models in accordance with the execution sequence using the specific data of the baseline data to identify recommendations for the user that provide an improvement for the user based on the determined event, and
displaying the recommendations via a user interface of the software application.
18 . The non-transitory computer-readable medium of claim 17 , wherein the ingesting further comprises normalizing income data within the ingested data records to remove sensitive details of the plurality of users, and the comparing comprises comparing the normalized income data to income attributes of the user to identify the recommendations.
19 . The non-transitory computer-readable medium of claim 17 , wherein the ingesting comprises ingesting the data records via a plurality of application programming interfaces (APIs) of the plurality of different data sources, respectively.
20 . The non-transitory computer-readable medium of claim 17 , wherein the displaying comprises filling-in a template on the user interface of the software application with image data of the recommendations for the user.