US 12,608,534 B2Granted

Managing data transformation tools using generative artificial intelligence

AssigneeInternational Business Machines Corporation

InventorAhamed Jalaldeen Shahul Hamid

Life of this patent

Estimating the term…

Loading patent…

Abstract

A computer implemented method manages data transformation tools. A processor set selects a foundation model to retrieve content of respective user manuals for a number of data transformation tools. The processor set generates a large language model from the foundation model based on the content of user manuals for the number of data transformation tools and a number of scripts for automation bots for the number of data transformation tools. The processor set generates a number of new automation bots for each data transformation tool in the number of data transformation tools using the large language model. The processor set performs a number of operations associated with data transformation maps in the number of data transformation tools using the number of new automaton bots.

Claims17

Dependent claims

A computer implemented method for managing data transformation tools, the computer implemented method comprising:

selecting, by a processor set, a foundation model to process respective user manuals for a number of data transformation tools, wherein the number of data transformation tools use different data formats to store information;
generating, by the processor set, a large language model from the foundation model based on user manuals for the number of data transformation tools and a number of scripts for automation bots for the number of data transformation tools, wherein the automation bots are software robots for automating operations for the number of data transformation tools, and wherein the large language model is an artificial intelligence model that performs natural language processing to generate instructions for automation bots based on the user manuals and the number of scripts for the automation bots;
generating, by the processor set, a number of new automation bots for each data transformation tool in the number of data transformation tools using the large language model, wherein each new automation bot is configured to perform an operation in data transformation tools associated with each new automation bot;
selecting, by the processor set, a first data transformation tool and a second data transformation tool from the number of data transformation tools;
retrieving, by the processor set, a first data transformation map from the first data transformation tool using a first automation bot for the first data transformation tool, wherein the first data transformation map is a pre-existing data transformation map, and wherein the first automation bot is an automation bot from the number of new automation bots;
creating, by the processor set, a first mapping specification based on the first data transformation map using the first automation bot for the first data transformation tool;
retrieving, by the processor set, the first mapping specification using a second automation bot for the second data transformation tool, wherein the second automation bot is an automation bot from the number of new automation bots;
and
creating, by the processor set using the second automation bot for the second data transformation tool, a new data transformation map in the second data transformation tool based on the first mapping specification.

6 more elements

A computer system comprising:

a processor set;
a set of one or more computer-readable storage media;
and
program instructions, collectively stored in the set of one or more storage media, for causing the processor set to perform the following computer operations:
select a foundation model to process respective user manuals for a number of data transformation tools, wherein the number of data transformation tools use different data formats to store information;
generate a large language model from the foundation model based on user manuals for the number of data transformation tools and a number of scripts for automation bots for the number of data transformation tools, wherein the automation bots are software robots for automating operations for the number of data transformation tools, and wherein the large language model is an artificial intelligence model that perform natural language processing to generate instructions for automation bots based on the user manuals and the number of scripts for the automation bots;
generate a number of new automation bots for each data transformation tool in the number of data transformation tools using the large language model, wherein each new automation bot is configured to perform an operation in data transformation tools associated with each new automation bot;
and
select a first data transformation tool and a second data transformation tool from the number of data transformation tools;
retrieve a first data transformation map from the first data transformation tool using a first automation bot for the first data transformation tool, wherein the first data transformation map is a pre-existing data transformation map, and wherein the first automation bot is an automation bot from the number of new automation bots;
creating, by the processor set, a first mapping specification based on the first data transformation map using the first automation bot for the first data transformation tool;
retrieve the first mapping specification using a second automation bot for the second data transformation tool, wherein the second automation bot is an automation bot from the number of new automation bots;
and
create a new data transformation map in the second data transformation tool based on the first mapping specification using the second automation bot for the second data transformation tool.

11 more elements

A computer program product for managing data transformation tools, the computer program product comprising:

a set of one or more computer-readable storage media;
program instructions, collectively stored in the set of one or more storage media, cause a processor set to perform the following computer operations:
select a foundation model to process respective user manuals for a number of data transformation tools, wherein the number of data transformation tools use different data formats to store information;
generate a large language model from the foundation model based on user manuals for the number of data transformation tools and a number of scripts for automation bots for the number of data transformation tools, wherein the automation bots are software robots for automating operations for the number of data transformation tools, and wherein the large language model is an artificial intelligence model that perform natural language processing to generate instructions for automation bots based on the user manuals and the number of scripts for the automation bots;
generate a number of new automation bots for each data transformation tool in the number of data transformation tools using the large language model, wherein each new automation bot is configured to perform an operation in data transformation tools associated with each new automation bot;
select a first data transformation tool and a second data transformation tool from the number of data transformation tools:
retrieve a first data transformation map from the first data transformation tool using a first automation bot for the first data transformation tool, wherein the first data transformation map is a pre-existing data transformation map, and wherein the first automation bot is an automation bot from the number of new automation bots;
creating, by the processor set, a first mapping specification based on the first data transformation map using the first automation bot for the first data transformation tool;
retrieve the first mapping specification using a second automation bot for the second data transformation tool, wherein the second automation bot is an automation bot from the number of new automation bots;
and
create a new data transformation map in the second data transformation tool based on the first mapping specification using the second automation bot for the second data transformation tool.

8 more elements

Dependencies are inferred from the claim text.