Granted Patent B2

US 12,230,046 · App. 18/492,910 · Granted Feb 18, 2025

Inventors: Shivam Agrahari (Bansi, IN); Dakshayani Singaraju (Tirupati, IN); Veresh Jain (Bangalore, IN)

Assignee: Oracle Financial Services Software Limited

G06V30/412G06F16/8373G06F40/103G06F40/143G06V10/82G06V30/16G06V30/19147G06V30/413

Abstract

Claims (89)

1. A computer-implemented method comprising:

extracting, by a table extraction model from an input document image, a plurality of tables as a plurality of objects, the plurality of tables including a plurality of cells, respectively;

outputting, by the table extraction model, the plurality of tables in an extensible Markup Language (XML) format and outputting cell level information of the plurality of cells;

filtering the plurality of tables based on a plurality of keywords and a condition comprising one from among AND and OR, to reduce a number of the plurality of tables, the filtering further comprising:

for each of the plurality of keywords, traversing each of the plurality of tables along a row direction and a column direction using one or more sub-keywords, each of the one or more sub-keywords being composed of one or more words of the keyword,

based on the traversing:

for each of the plurality of tables, determining whether a table comprises a first keyword from the plurality of keywords and determining whether the table comprises a second keyword from the plurality of keywords,

for the condition being AND, based on the table containing the first keyword and the second keyword, outputting the table, as a filtered table among filtered tables, and, based on the table not containing the first keyword and the second keyword, not outputting the table, and

for the condition being OR, based on the table containing the first keyword or the second keyword, outputting the table, as a filtered table among the filtered tables;

outputting the filtered tables that include the at least one from among the first keyword and the second keyword based on the condition, among the plurality of tables;

executing a query on the filtered tables, the query including one or more keys composed based on the cell level information associated with the filtered tables;

identifying one or more cells among the plurality of cells based on the one or more keys; and

providing target content that corresponds to the one or more cells, wherein the target content is a response to the query.

2. The computer-implemented method of claim 1 , further comprising:

obtaining a set of conditions, the set of conditions comprising AND and OR, prior to the filtering; and

obtaining the condition for the filtering from the set of conditions.

3. The computer-implemented method of claim 1 , wherein the filtering further comprises:

initializing a vector comprising an array having a number of elements equal to a number of the plurality of keywords;

based on the table containing the first keyword, generating a value of 1 as a first element of the array;

based on the table containing the second keyword, generating a value of 1 as a second element of the array;

based on the condition being AND, outputting the table if the first element and the second element have the value of 1; and

based on the condition being OR, outputting the table if at least one from among the first element and the second element has the value of 1.

4. The computer-implemented method of claim 1 , wherein the executing the query further comprises:

obtaining the cell level information associated with each of the filtered tables, the cell level information comprising, for each of the filtered tables, coordinates of cells and table architecture; and

deriving grammar and rules specific to each of the filtered tables, using the coordinates of cells and the table architecture.

5. The computer-implemented method of claim 4 , wherein the executing the query further comprises creating the query based on the grammar and rules specific to each of the filtered tables.

6. The computer-implemented method of claim 5 , wherein:

at least one of the filtered tables is a nested table, and

the query comprises at least one composite key that expresses parent-child relationship of a header and at least one sub-header present in the nested table.

7. A computer system comprising:

one or more data processors; and

one or more non-transitory computer-readable storage media storing instructions that, when executed by the one or more data processors, cause the one or more data processors to perform a method including:

extracting, by using a table extraction model from an input document image, a plurality of tables as a plurality of objects, the plurality of tables including a plurality of cells, respectively;

receiving, as an output the table extraction model, the plurality of tables in an extensible Markup Language (XML) format and outputting cell level information of the plurality of cells;

filtering the plurality of tables based on a plurality of keywords and a condition comprising one from among AND and OR, to reduce a number of the plurality of tables, the filtering further including:

for each of the plurality of keywords, traversing each of the plurality of tables along a row direction and a column direction using one or more sub-keywords, each of the one or more sub-keywords being composed of one or more words of the keyword,

based on the traversing:

for each of the plurality of tables, determining whether a table comprises a first keyword from the plurality of keywords and determining whether the table comprises a second keyword from the plurality of keywords,

for the condition being AND, based on the table containing the first keyword and the second keyword, outputting the table, as a filtered table among filtered tables, and, based on the table not containing the first keyword and the second keyword, not outputting the table, and

for the condition being OR, based on the table containing the first keyword or the second keyword, outputting the table, as a filtered table among the filtered tables;

outputting the filtered tables that include the at least one from among the first keyword and the second keyword based on the condition, among the plurality of tables;

executing a query on the filtered tables, the query including one or more keys composed based on the cell level information associated with the filtered tables;

identifying one or more cells among the plurality of cells based on the one or more keys; and

providing target content that corresponds to the one or more cells, wherein the target content is a response to the query.

8. The computer system of claim 7 , wherein the method further includes:

obtaining a set of conditions, the set of conditions comprising AND and OR, prior to the filtering; and

obtaining the condition for the filtering from the set of conditions.

9. The computer system of claim 8 , wherein the filtering further includes:

initializing a vector comprising an array having a number of elements equal to a number of the plurality of keywords;

based on the table containing the first keyword, generating a value of 1 as a first element of the array;

based on the table containing the second keyword, generating a value of 1 as a second element of the array;

based on the condition being AND, outputting the table if the first element and the second element have the value of 1; and

based on the condition being OR, outputting the table if at least one from among the first element and the second element has the value of 1.

10. The computer system of claim 7 , wherein the executing the query further includes:

obtaining the cell level information associated with each of the filtered tables, the cell level information comprising, for each of the filtered tables, coordinates of cells and table architecture; and

deriving grammar and rules specific to each of the filtered tables, using the coordinates of cells and the table architecture.

11. The computer system of claim 10 , wherein the executing the query further includes creating the query based on the grammar and rules specific to each of the filtered tables.

12. The computer system of claim 11 , wherein:

at least one of the filtered tables is a nested table, and

the query comprises at least one composite key that expresses parent-child relationship of a header and at least one sub-header present in the nested table.

13. A computer-program product tangibly embodied in one or more non-transitory computer-readable storage media including instructions configured to cause one or more data processors to perform a method including:

extracting, by using a table extraction model from an input document image, a plurality of tables as a plurality of objects, the plurality of tables including a plurality of cells, respectively;

receiving, as an output the table extraction model, the plurality of tables in an extensible Markup Language (XML) format and outputting cell level information of the plurality of cells;

filtering the plurality of tables based on a plurality of keywords and a condition comprising one from among AND and OR, to reduce a number of the plurality of tables, the filtering further including:

for each of the plurality of keywords, traversing each of the plurality of tables along a row direction and a column direction using one or more sub-keywords, each of the one or more sub-keywords being composed of one or more words of the keyword,

based on the traversing:

for each of the plurality of tables, determining whether a table comprises a first keyword from the plurality of keywords and determining whether the table comprises a second keyword from the plurality of keywords,

for the condition being AND, based on the table containing the first keyword and the second keyword, outputting the table, as a filtered table among filtered tables, and, based on the table not containing the first keyword and the second keyword, not outputting the table, and

for the condition being OR, based on the table containing the first keyword or the second keyword, outputting the table, as a filtered table among the filtered tables;

outputting the filtered tables that include the at least one from among the first keyword and the second keyword based on the condition, among the plurality of tables;

executing a query on the filtered tables, the query including one or more keys composed based on the cell level information associated with the filtered tables;

identifying one or more cells among the plurality of cells based on the one or more keys; and

providing target content that corresponds to the one or more cells, wherein the target content is a response to the query.

14. The computer-program product of claim 13 , wherein the method further includes:

obtaining a set of conditions, the set of conditions comprising AND and OR, prior to the filtering; and

obtaining the condition for the filtering from the set of conditions.

15. The computer-program product of claim 13 , wherein the filtering further includes:

initializing a vector comprising an array having a number of elements equal to a number of the plurality of keywords;

based on the table containing the first keyword, generating a value of 1 as a first element of the array;

based on the table containing the second keyword, generating a value of 1 as a second element of the array;

based on the condition being AND, outputting the table if the first element and the second element have the value of 1; and

based on the condition being OR, outputting the table if at least one from among the first element and the second element has the value of 1.

16. The computer-program product of claim 13 , wherein the executing the query further includes:

obtaining the cell level information associated with each of the filtered tables, the cell level information comprising, for each of the filtered tables, coordinates of cells and table architecture; and

deriving grammar and rules specific to each of the filtered tables, using the coordinates of cells and the table architecture.

17. The computer-program product of claim 16 , wherein:

the executing the query further includes creating the query based on the grammar and rules specific to each of the filtered tables,

at least one of the filtered tables is a nested table, and

the query comprises at least one composite key that expresses parent-child relationship of a header and at least one sub-header present in the nested table.

Assignments (1)

Continuity (2)

References Cited (43)