Natural language interaction with automated machine learning systems

Patent metadata

Field	Value
Publication number	US-11989237-B2
Application number	US-201916551021-A
Country	US
Kind code	B2
Filing date	Aug 26, 2019
Priority date	Aug 26, 2019
Publication date	May 21, 2024
Grant date	May 21, 2024

Abstract

An artificial intelligence (AI) interaction method, system, and computer program product include selecting an artificial intelligence model to respond to a query to generating a response to the query using the selected artificial intelligence model, and receiving the response to the query from the selected artificial intelligence model.

First claim

What is claimed is:

1. A computer-implemented artificial intelligence (AI) interaction method, the method comprising:
mapping an input provided by a user into a configuration, the input including a query by the user and a dataset initially provided at a time of the query that includes corresponding data within the dataset to answer the query, wherein the configuration is generated from the query and the dataset received at substantially a same time;
selecting, based on the configuration, an artificial intelligence model to respond to the input based on the artificial intelligence model being capable of processing the corresponding data within the dataset to answer the query;
generating a response to the query using the selected artificial intelligence model by running the selected artificial intelligence model on the dataset with respect to the query such that the corresponding data is utilized within the selected artificial intelligence model to generate the response; and
receiving the response to the query from the selected artificial intelligence model.

2. The method of claim 1 , wherein the artificial intelligence model responds to the query based on artificial intelligence configuration files.

3. The method of claim 1 , wherein the query comprises a natural language query.

4. The method of claim 1 , wherein the query comprises a natural language query,
wherein a first model combines the dataset and the natural language query to the configuration.

5. The method of claim 4 , wherein a description for each column within a table input from the dataset is encoded using a transformer encoder to provide a feature representation for each column,
further comprising encoding the query using the transformer encoder to project the query to a same feature space of the column description, and
for each of the feature representation, concatenating the feature representation with the query feature representation to form a joint representation of the column and the query.

6. The method of claim 1 , embodied in a cloud-computing environment.

7. A computer program product for artificial intelligence (AI) interaction, the computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to perform:
mapping an input provided by a user into a configuration, the input including a query by the user and a dataset initially provided at a time of the query that includes corresponding data within the dataset to answer the query, wherein the configuration is generated from the query and the dataset received at substantially a same time;
selecting, based on the configuration, an artificial intelligence model to respond to the input based on the artificial intelligence model being capable of processing the corresponding data within the dataset to answer the query;
generating a response to the query using the selected artificial intelligence model by running the selected artificial intelligence model on the dataset with respect to the query such that the corresponding data is utilized within the selected artificial intelligence model to generate the response; and
receiving the response to the query from the selected artificial intelligence model.

8. The computer program product of claim 7 , wherein the artificial intelligence model responds to the query based on artificial intelligence configuration files.

9. The computer program product of claim 7 , wherein the query comprises a natural language query.

10. The computer program product of claim 7 ,
wherein the query comprises a natural language query, wherein a first model combines the dataset and the natural language query to the configuration.

11. The computer program product of claim 10 , wherein a description for each column within a table input from the dataset is encoded using a transformer encoder to provide a feature representation for each column,
further comprising encoding the query using the transformer encoder to project the query to a same feature space of the column description, and
for each of the feature representation, concatenating the feature representation with the query feature representation to form a joint representation of the column and the query.

12. An artificial intelligence (AI) interaction system, the system comprising:
a processor; and
a memory, the memory storing instructions to cause the processor to perform:
mapping an input provided by a user into a configuration, the input including a query by the user and a dataset initially provided at a time of the query that includes corresponding data within the dataset to answer the query, wherein the configuration is generated from the query and the dataset received at substantially a same time;
selecting, based on the configuration, an artificial intelligence model to respond to the input based on the selected artificial intelligence model being capable of processing the corresponding data within the dataset to answer the query;
generating a response to the query using the selected artificial intelligence model by running the selected artificial intelligence model on the dataset with respect to the query such that the corresponding data is utilized within the selected artificial intelligence model to generate the response; and
receiving the response to the query from the selected artificial intelligence model.

13. The system of claim 12 , wherein the artificial intelligence model responds to the query based on artificial intelligence configuration files.

14. The system of claim 12 , wherein the query comprises a natural language query.

15. The system of claim 12 ,
wherein the query comprises a natural language query,
wherein a first model combines the dataset and the natural language query to the configuration.

16. The system of claim 15 , wherein a description for each column within a table input from the dataset is encoded using a transformer encoder to provide a feature representation for each column,
further comprising encoding the query using the transformer encoder to project the query to a same feature space of the column description, and
wherein, for each of the feature representation, concatenating the feature representation with the query feature representation to form a joint representation of the column and the query.

17. The method of claim 1 , wherein the configuration is processed by a first model that is different than the artificial intelligence model.

18. The method of claim 1 , wherein, at the same time, the user provides the query and the dataset that has data within the dataset of which the artificial intelligence model runs against the dataset to find the answer to the query.

19. The method of claim 1 , wherein the selected artificial intelligence model is selected from a set of models generated by automated machine learning.

20. The method of claim 1 , further comprising outputting the configuration in an automated machine learning model format.

Assignees

• IBM

Inventors

• Wang Dakuo
• Tan Ming
• Gan Chuang
• Wang Haoyu
• Yu Mo

Classifications

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06F16/90332 Primary

  Natural language query formulation or dialogue systems · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N3/08

  Learning methods · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title