US-12620249-B2 - Methods, systems, articles of manufacture and apparatus to improve tagging accuracy

Abstract

Methods, apparatus, systems, and articles of manufacture are disclosed to improve tagging accuracy. An example apparatus includes at least one memory, machine readable instructions, and processor circuitry to execute the machine readable instructions to at least search a first row of a document to identify a first row that includes a first type of entity, search the first row of the document to identify a second type of entity that is missing, search the first row of the document to identify a first integer value, and associate the first row with a product corresponding to the first integer value.

Inventors

• Jose Javier Yebes Torres
• Sricharan Amarnath
• Roberto Arroyo

Assignees

• NIELSEN CONSUMER LLC

Dates

Publication Date

20260505

Application Date

20220909

Priority Date

20220131

Claims (10)

1. 1 . At least one non-transitory machine readable medium comprising instructions to cause at least one programmable circuit to at least: perform Optical Character Recognition (OCR) on an image of a document to generate OCR data; search the OCR data to identify a first row that includes product description information; search the OCR data to identify at least one of a product code of a product identified by the product description information, a product quantity of the product, or a price of the product that is missing from the first row; and identifying the at least one of the missing product code, the missing product quantity or the missing price by at least one of: identifying a maximum integer value in the OCR data as the missing product code product code; identifying a minimum integer value in the OCR data as the product quantity; or identifying a decimal value in the OCR data as the price; and retrain a machine learning model based on at last one of the maximum integer value, the minimum integer value, or the decimal value.
2. 2 . The at least one non-transitory machine readable medium of claim 1 , wherein the product code is a first product code, the product quantity is a first product quantity, the price is a first price, and the document is a first document, wherein the instructions cause the at least one programmable circuit to retrain the model to automatically detect a second product code, a second product quantity, and a second price in a second document.
3. 3 . The at least one non-transitory machine readable medium of claim 1 , wherein the model includes processing key information extraction from documents using improved graph learning convolutional networks.
4. 4 . The at least one non-transitory machine readable medium of claim 1 , wherein the model is to transform key information extraction into a sequence tagging based on layout of the document.
5. 5 . The at least one non-transitory machine readable medium of claim 1 , wherein the document is physical and the OCR data is digital.
6. 6 . An apparatus comprising: at least one memory; machine readable instructions; and at least one programmable circuit to execute the machine readable instructions to at least: perform Optical Character Recognition (OCR) on an image of a document to generate OCR data; search the OCR data to identify a first row that includes product description information; search the OCR data to identify at least one of a product code, a product quantity, or a price that is missing from the first row; search the OCR data to identify integer values; identify the maximum integer value of the integer values; identify the minimum integer value of the integer values; associate, when the maximum integer value is different than the minimum integer value, the maximum integer value with the product code; associate, when the minimum integer value is different than the maximum integer value, the minimum integer value with the product quantity; search the OCR data for a decimal value; associate the decimal value with the price; and retrain a machine learning model based on at least one of the product code, the product quantity, or the price.
7. 7 . The apparatus of claim 6 , wherein the product code is a first product code, the product quantity is a first product quantity, the price is a first price, and the document is a first document, and wherein at least one of the at least one programmable circuit is to retrain the model to automatically detect a second product code, a second product quantity, and a second price in a second document.
8. 8 . The apparatus of claim 6 , wherein the model includes processing key information extraction from documents using improved graph learning convolutional networks.
9. 9 . The apparatus of claim 6 , wherein the model is to transform key information extraction into a sequence tagging based on layout of the document.
10. 10 . The apparatus of claim 6 , wherein the document is physical and the OCR data is digital.

Description

RELATED APPLICATION This patent claims the benefit of Indian Provisional Patent Application No. 202211005270, which was filed on Jan. 31, 2022. Indian Provisional Patent Application No. 202211005270 is hereby incorporated herein by reference in its entirety. Priority to Indian Provisional Patent Application No. 202211005270 is hereby claimed. FIELD OF THE DISCLOSURE This disclosure relates generally to image recognition and, more particularly, to methods, systems, articles of manufacture and apparatus to improve tagging accuracy. BACKGROUND In recent years, optical character recognition (OCR) has been employed to extract text from images. In some examples, OCR techniques exhibit erroneous results. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 illustrates example receipts and invoices to be processed in a manner consistent with this disclosure. FIG. 2 illustrates example receipts and invoices that do not include errors. FIG. 3 illustrates example extractions using post-processing heuristics. FIG. 4 illustrates example predictions that have been trained and include errors. FIG. 5 illustrates example product line groupings. FIG. 6 illustrates example product code corrections based on application of heuristics. FIG. 7 illustrates example product quantity corrections based on application of heuristics. FIG. 8A illustrates example product price corrections based on application of heuristics. FIG. 8B illustrates accuracy results based on application of examples disclosed herein. FIG. 9 illustrates a block diagram of an example system structured to improve tagging in purchase documents. FIG. 10 is a block diagram of example heuristics circuitry to improve entity tagging in purchase documents. FIGS. 11 and 12 are flowcharts representative of example machine readable instructions and/or example operations that may be executed by example processor circuitry to implement the example heuristics circuitry of FIG. 10. FIG. 13 is a block diagram of an example processing platform including processor circuitry structured to execute the example machine readable instructions and/or the example operations of FIGS. 11 and 12 to implement the heuristics circuitry of FIG. 10. FIG. 14 is a block diagram of an example implementation of the processor circuitry of FIG. 13. FIG. 15 is a block diagram of another example implementation of the processor circuitry of FIG. 13. FIG. 16 is a block diagram of an example software distribution platform (e.g., one or more servers) to distribute software (e.g., software corresponding to the example machine readable instructions of FIGS. 11 and 12) to client devices associated with end users and/or consumers (e.g., for license, sale, and/or use), retailers (e.g., for sale, re-sale, license, and/or sub-license), and/or original equipment manufacturers (OEMs) (e.g., for inclusion in products to be distributed to, for example, retailers and/or to other end users such as direct buy customers). In general, the same reference numbers will be used throughout the drawing(s) and accompanying written description to refer to the same or like parts. The figures are not to scale. As used herein, the phrase “in communication,” including variations thereof, encompasses direct communication and/or indirect communication through one or more intermediary components, and does not require direct physical (e.g., wired) communication and/or constant communication, but rather additionally includes selective communication at periodic intervals, scheduled intervals, aperiodic intervals, and/or one-time events. As used herein, “processor circuitry” is defined to include (i) one or more special purpose electrical circuits structured to perform specific operation(s) and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors), and/or (ii) one or more general purpose semiconductor-based electrical circuits programmable with instructions to perform specific operations and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of processor circuitry include programmable microprocessors, Field Programmable Gate Arrays (FPGAs) that may instantiate instructions, Central Processor Units (CPUs), Graphics Processor Units (GPUs), Digital Signal Processors (DSPs), XPUs, or microcontrollers and integrated circuits such as Application Specific Integrated Circuits (ASICs). For example, an XPU may be implemented by a heterogeneous computing system including multiple types of processor circuitry (e.g., one or more FPGAs, one or more CPUs, one or more GPUS, one or more DSPs, etc., and/or a combination thereof) and application programming interface(s) (API(s)) that may assign computing task(s) to whichever one(s) of the multiple types of processor circuitry is/are best suited to execute the computing task(s). DETAILED DESCRIPTION Text processing and understanding is a valuable asset for varied Artificial In