System and method for information extraction and retrieval for automotive repair assistance

Patent metadata

Field	Value
Publication number	US-11734267-B2
Application number	US-201816234989-A
Country	US
Kind code	B2
Filing date	Dec 28, 2018
Priority date	Dec 28, 2018
Publication date	Aug 22, 2023
Grant date	Aug 22, 2023

Abstract

A server includes a controller in communication with the server and configured to receive a query related to an issue with a vehicle, retrieve information related to the query from a database that includes structured data including data related to vehicle issues and a solution to the vehicle issues, determine whether the database includes structured data including data related to the solution responsive to the query, and output information including the solution to the issue when the determination identifies that the structured data including the solution in the database or retrieve information from a website when the determination identifies that the structured data including the solution is not found in the database, wherein the controller is further configured to retrieve the information from the website by conducting a keyword search for a first set of results and filtering the information from the website utilizing an autoencoder.

First claim

What is claimed is:

1. A server comprising:
a controller in communication with the server, the controller configured to:
receive and parse a text query related to an issue with a vehicle, wherein the controller is configured to parse the text query utilizing an autoencoder to encode, compress, and decode the text query, wherein the autoencoder includes a first encoder layer and a second encoder layer, wherein the second encoder layer includes less dimensional vectors than the first encoder layer, and the autoencoder includes a first decoder layer and a second decoder layer;
in response to the text query, retrieve information related to the text query from a database that includes structured data in a data schema related to both vehicle issues and a solution to the vehicle issues;
determine whether the database is responsive to the text query; and
in response to the determination, output information including the solution to the issue when the solution is found in the database, and when the solution is not found in the database, retrieve solution information from a website utilizing an information exchange module configured to utilize knowledge verification processing that utilizes knowledge bases associated with the database with a pre-defined scheme to assess information extracted from sources outside of the knowledge bases when information found in the knowledge bases is below a threshold, wherein the controller is further configured to retrieve the solution information from the websites by conducting a keyword search for a first set of results.

2. The server of claim 1 , wherein the autoencoder is configured to utilize an input associated with the text query, wherein the input includes the first encoding layer with a first vector dimension that includes numerical vectors.

3. The server of claim 2 , wherein the autoencoder is configured to encode the input to a short code that includes a coding vector dimension, wherein the first vector dimension includes more vectors than the coding vector dimension.

4. The server of claim 3 , wherein the autoencoder is configured to decode the short code to an output that includes on a first decoding dimension, wherein the first decoding dimension is larger than the coding vector dimension.

5. The server of claim 1 , wherein the controller is further configured to filter the information from the website utilizing an autoencoder, and utilize input from a domain expert to verify the information extracted from sources outside of the knowledge base.

6. The server of claim 5 , wherein the controller is further configured to output the filtered information retrieved from the website.

7. The server of claim 6 , wherein the controller is further configured to store the filtered information into the database.

8. The server of claim 7 , wherein the controller is further configured to store the filtered information into the database in response to an input received from a user.

9. The server of claim 8 , wherein the controller is further configured to store the filtered information into the database by converting it to the structured data.

10. A server comprising:
a controller in communication with the server, the controller configured to:
receive and parse a text query related to an issue with a vehicle;
in response to the text query, retrieve information related to the text query from a database that includes structured data including data related to vehicle issues and a solution to the vehicle issues;
determine whether the database includes structured data including data related to the solution responsive to the text query;
output information including the solution to the issue when the determination identifies that the structured data including the solution in the database and when the determination identifies that the structured data is not found in the database, retrieve information from a website when the determination identifies, wherein the controller is further configured to retrieve the information from the website by utilizing knowledge verification processing that utilizes knowledge bases associated with the database with a pre-defined scheme to assess information extracted from sources outside of the knowledge bases when information found in the knowledge bases is below a threshold and conducting a keyword search at the website for a first set of results and filtering the information from the website utilizing an autoencoder, wherein the autoencoder includes a first encoding layer and a second encoding layer and a first decoding layer and a second decoding layer, and a hidden layer, wherein the hidden layer includes a lower dimensionality than input data associated with the text query.

11. The server of claim 10 , wherein controller is configured to parse the text query by utilizing the autoencoder configured to encode the text query to the first encoding layer with a first vector dimension, and further encoding the first encoding layer to the second encoding layer with a second vector dimension, wherein the first vector dimension includes more vectors than the second vector dimension.

12. The server of claim 11 , wherein the controller is configured to encode the second encoding layer to code that includes a coding vector dimension, wherein the second vector dimension includes more vectors than the coding vector dimension.

13. The server of claim 12 , wherein the code includes less vectors than the first encoding layer and the second encoding layer.

14. The server of claim 13 , wherein the controller is further configured to decode the code to a first decoding layer includes a first decoding vector dimension, wherein the first decoding vector dimension includes more vectors than the coding vector dimension.

15. The server of claim 14 , wherein the controller is further configured to decode the first decoding layer to a second decoding layer that includes a second decoding vector dimension, wherein the second decoding vector dimension includes more vectors than the first decoding vector dimension.

16. The server of claim 10 , wherein the controller is further programmed to output the filtered information from the website.

17. A querying system, comprising:
an input device configured to allow a user to enter a text query related to automotive repairs;
a display configured to output information related to the text query;
a database that includes structured data with a data related to vehicle issues and a solution to the automotive repairs; and
a processor in communication with the input device, database, and display, wherein the processor is programmed to:
receive and parse the text query;
in response to the text query, retrieve information related to the text query from the database that includes a solution to the automotive repairs;
determine whether the database includes structured data including data related to the solution responsive to the text query; and
output information including the solution to the issue when the structured data including the solution is found in the database and when the determination identifies that the structured data is not found in the database retrieve information from a website, wherein the processor is further configured to retrieve the information from the web site by conducting a keyword search for a first set of results and filtering the information from the website utilizing an autoencoder to define a second set of results, wherein the autoencoder includes a first encoder layer and a second encoder layer, wherein the second encoder layer includes less dimensional vectors than the first encoder

Assignees

• Bosch Gmbh Robert

Inventors

• Araki Jun
• Zhao Lin
• Feng Zhe

Classifications

• G06F16/24522 Primary

  Translation of natural language queries to structured queries · CPC title

• G06F16/248

  Presentation of query results · CPC title

• G06F16/9538

  Presentation of query results · CPC title

• G06F16/334 Primary

  Query execution (filtering based on additional data G06F16/335) · CPC title

• G06F16/9035 Primary

  Filtering based on additional data, e.g. user or group profiles · CPC title