Data insights using context driven lateral ai

Patent metadata

Field	Value
Publication number	US-2025036659-A1
Application number	US-202318361563-A
Country	US
Kind code	A1
Filing date	Jul 28, 2023
Priority date	Jul 28, 2023
Publication date	Jan 30, 2025
Grant date	—

Abstract

An approach is disclosed that receives a user activity from a user that is using a datastore visualization display. The display displays a first visualization pertaining to one or more datastores with the datastore visualization display provided at least in part by a context driven lateral artificial intelligence (CDLAI) engine. The received user activity is provided as input to the CDLAI to generate a second visualization that is displayed to the user at the datastore visualization display that is displayed to the user. Artificial intelligence (AI) models that are used by the CDLAI are then trained based on the received user activity. The training results in updates to the visualizations.

First claim

1 . A method, implemented by a processor coupled to a memory, comprising:
receiving, from a user, a user activity at a datastore visualization display that displays a first visualization pertaining to one or more datastores, wherein the datastore visualization display is provided at least in part by a context driven artificial intelligence engine; inputting the received user activity to the context driven artificial intelligence engine to generate a second visualization that is displayed to the user at the datastore visualization display; and training one or more artificial intelligence (AI) models that are used by the context driven artificial intelligence engine based on the received user activity, wherein the training results in one or more updates to one or more visualizations that include the first and second visualizations.

2 . The method of claim 1 further comprising:
gathering a set of data points from the datastores;
scanning a word bank for a plurality of data point names by identifying related texts in a plurality of field names included in the datastores;
retrieving one or more context attributes corresponding to one or more of the data points; and
based on the one or more data points, the related texts, and the corresponding context attributes, creating a plurality of data point contexts.

3 . The method of claim 2 further comprising:
querying a set of data from the datastores using the created data point contexts, the querying resulting in a plurality of analytics cluster functions based on the data point contexts, one or more data point types and a plurality of actual data point values corresponding to each context attribute; and
creating a new AI data model for use by the context driven artificial intelligence engine, the new AI data model being created from the running of the created the plurality of analytics cluster functions that utilizes the plurality of actual data point values that correspond to each context attribute.

4 . The method of claim 3 further comprising:
analyzing the new AI data model to identify one or more patterns corresponding to the data point contexts.

5 . The method of claim 4 further comprising:
creating a new visualization based on the identified patterns corresponding to the data point contexts; and
training one or more of the AI models based on the created new visualization.

6 . The method of claim 5 further comprising:
storing the new visualization and the updates to the visualizations in a context repository;
storing a plurality of domain specific contexts in the context repository; and
storing a plurality of user feedback contexts in the repository.

7 . The method of claim 6 wherein the user activity is a selection of an area that is a non-clickable area of the visualization display and the method further comprises:
creating a hash table of the data point contexts;
creating one or more labels corresponding to each of the data point contexts included in the hash table and including the created labels in the hash table;
plotting each data point context included in the hash table against at least one of the created analytics cluster functions, the plotting resulting in a set of entry points that are included in a multidimensional space;
displaying the multidimensional space to the user;
receiving, from the user, a selection of one of the entry points included in the displayed multidimensional space; and
responsively displaying a selected one of the visualizations from the context repository based on the entry point selected by the user.

8 . An information handling system comprising:
one or more processors; a memory coupled to at least one of the processors; and a set of instructions stored in the memory and executed by at least one of the processors to perform actions comprising:
receiving, from a user, a user activity at a datastore visualization display that displays a first visualization pertaining to one or more datastores, wherein the datastore visualization display is provided at least in part by a context driven artificial intelligence engine;
inputting the received user activity to the context driven artificial intelligence engine to generate a second visualization that is displayed to the user at the datastore visualization display; and
training one or more artificial intelligence (AI) models that are used by the context driven artificial intelligence engine based on the received user activity, wherein the training results in one or more updates to one or more visualizations that include the first and second visualizations.

9 . The information handling system of claim 8 wherein the actions further comprise:
gathering a set of data points from the datastores;
scanning a word bank for a plurality of data point names by identifying related texts in a plurality of field names included in the datastores;
retrieving one or more context attributes corresponding to one or more of the data points; and
based on the one or more data points, the related texts, and the corresponding context attributes, creating a plurality of data point contexts.

10 . The information handling system of claim 9 wherein the actions further comprise:
querying a set of data from the datastores using the created data point contexts, the querying resulting in a plurality of analytics cluster functions based on the data point contexts, one or more data point types and a plurality of actual data point values corresponding to each context attribute; and
creating a new AI data model for use by the context driven artificial intelligence engine, the new AI data model being created from the running of the created the plurality of analytics cluster functions that utilizes the plurality of actual data point values that correspond to each context attribute.

11 . The information handling system of claim 10 wherein the actions further comprise:
analyzing the new AI data model to identify one or more patterns corresponding to the data point contexts.

12 . The information handling system of claim 11 wherein the actions further comprise:
creating a new visualization based on the identified patterns corresponding to the data point contexts; and
training one or more of the AI models based on the created new visualization.

13 . The information handling system of claim 12 wherein the actions further comprise:
storing the new visualization and the updates to the visualizations in a context repository;
storing a plurality of domain specific contexts in the context repository; and
storing a plurality of user feedback contexts in the repository.

14 . The information handling system of claim 13 wherein the user activity is a selection of an area that is a non-clickable area of the visualization display and the actions further comprise:
creating a hash table of the data point contexts;
creating one or more labels corresponding to each of the data point contexts included in the hash table and including the created labels in the hash table;
plotting each data point context included in the hash table against at least one of the created analytics cluster functions, the plotting resulting in a set of entry points that are included in a multidimensional space;
displaying the multidimensional space to the user;
receiving, from the user, a selection of one of the entry points included in the displayed multidimensional space; and
responsively displaying a selected one of the visualizations from the context repository based on the entry point selected by the user.

15 . A computer program product comprising:
a computer readable storage m

Assignees

• Kyndryl Inc

Inventors

• Sharma Kalpesh
• Perumal Thevar Muniyandi
• M Nalini
• Bushetti Veeresh

Classifications

• G06F16/287 Primary

  Visualization; Browsing · CPC title

• G06F16/2255 Primary

  Hash tables · CPC title