Data and artificial intelligence-driven automatic data model build

1 . A computer-implemented method comprising:

responsive to receiving a dataset, determining by a machine learning model of a system a relationship of a datastore, comprising a primary key and a foreign-key relationship in a relational database, by training the machine learning model wherein the machine learning model comprises an embedding model based on an attribute of the datastore;

determining by the machine learning model of the system, a rule of the dataset, comprising an indicator metric of achieving a result based on contextual data of the dataset, using the dataset and the relationship of the datastore as input into the machine learning model;

generating code by the machine learning model comprising inputting a prompt into the machine learning model, the prompt comprising the indicator metric and the primary key and the foreign-key relationship in the relational database; and

computing, by executing code generated by the machine learning model responsive to the rule and the relationship of the datastore, outputting the result for the rule of the dataset wherein the result comprises the indicator metric of the rule of the dataset in the datastore.

2 . The computer-implemented method of claim 1 , further comprising using the machine learning model to generate a subgraph from the datastore that comprises data relevant to the rule wherein computing the result is further based on the subgraph.

3 . The computer-implemented method of claim 2 , wherein generating the subgraph further comprises using the machine learning model based on contextual data of the dataset and the relationship of the contextual data and the datastore.

4 . The computer-implemented method of claim 1 , wherein computing the result further comprises merging of subgraphs wherein a first subgraph is computed by the machine learning model from the datastore based on a first indicator metric, a second subgraph of the datastore is computed by the machine learning model from the datastore based on a second indicator metric, and the first subgraph and the second subgraph are merged according to a primary key of the first subgraph and a primary key of the second subgraph.

5 . The computer-implemented method of claim 1 , wherein the embedding model comprises a feature vector of the attribute of the datastore.

6 . The computer-implemented method of claim 1 , wherein the rule of the dataset comprises a key performance indicator of the dataset.

7 . A computer program product comprising one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by a processor to cause the processor to perform operations comprising:

responsive to receiving a dataset, determining by a machine learning model of a system a relationship of a datastore, comprising a primary key and a foreign-key relationship in a relational database, by training the machine learning model wherein the machine learning model comprises an embedding model based on an attribute of the datastore;

determining by the machine learning model of the system, a rule of the dataset, comprising an indicator metric of achieving a result based on contextual data of the dataset, using the dataset and the relationship of the datastore as input into the machine learning model;

generating code by the machine learning model comprising inputting a prompt into the machine learning model, the prompt comprising the indicator metric and the primary key and the foreign-key relationship in the relational database; and

computing, by executing code generated by the machine learning model responsive to the rule and the relationship of the datastore, outputting the result for the rule of the dataset wherein the result comprises the indicator metric of the rule of the dataset in the datastore.

8 . The computer program product of claim 7 , further comprising using the machine learning model to generate a subgraph from the datastore that comprises data relevant to the rule wherein computing the result is further based on the subgraph.

9 . The computer program product of claim 8 , generating the subgraph further comprises using the machine learning model based on contextual data of the dataset and the relationship of the contextual data and the datastore.

10 . The computer program product of claim 7 , wherein computing the result further comprises merging of subgraphs wherein a first subgraph is computed by the machine learning model from the datastore based on a first indicator metric, a second subgraph of the datastore is computed by the machine learning model from the datastore based on a second indicator metric, and the first subgraph and the second subgraph are merged according to a primary key of the first subgraph and a primary key of the second subgraph.

11 . The computer program product of claim 7 , wherein the embedding model comprises a feature vector of the attribute of the datastore.

12 . The computer program product of claim 7 , wherein the rule of the dataset comprises a key performance indicator of the dataset.

13 . A computer system comprising a processor and one or more computer readable storage media, and program instructions collectively stored on the one or more computer readable storage media, the program instructions executable by the processor to cause the processor to perform operations comprising:

responsive to receiving a dataset, determining by a machine learning model of a system a relationship of a datastore, comprising a primary key and a foreign-key relationship in a relational database, by training the machine learning model wherein the machine learning model comprises an embedding model based on an attribute of the datastore;

determining by the machine learning model of the system, a rule of the dataset, comprising an indicator metric of achieving a result based on contextual data of the dataset, using the dataset and the relationship of the datastore as input into the machine learning model;

generating code by the machine learning model comprising inputting a prompt into the machine learning model, the prompt comprising the indicator metric and the primary key and the foreign-key relationship in the relational database; and

computing, by executing code generated by the machine learning model responsive to the rule and the relationship of the datastore, outputting the result for the rule of the dataset wherein the result comprises the indicator metric of the rule of the dataset in the datastore.

14 . The computer system of claim 13 , further comprising using the machine learning model to generate a subgraph from the datastore that comprises data relevant to the rule wherein computing the result is further based on the subgraph.

15 . The computer system of claim 14 , generating the subgraph further comprises using the machine learning model based on contextual data of the dataset and the relationship of the contextual data and the datastore.

16 . The computer system of claim 13 , wherein computing the result further comprises merging of subgraphs wherein a first subgraph is computed by the machine learning model from the datastore based on a first indicator metric, a second subgraph of the datastore is computed by the machine learning model from the datastore based on a second indicator metric, and the first subgraph and the second subgraph are merged according to a primary key of the first subgraph and a primary key of the second subgraph.

17 . The computer system of claim 13 , wherein the embedding model comprises a feature vector of the attribute of the datastore.