Systems and methods for generating notifications based on extracted data from documents using a machine learning algorithm. The method includes receiving a document uploaded by a user including data corresponding to the user. The method further includes receiving an indication from the user granting permission to extract the data from the document. The method also includes extracting the data from the document using optical character recognition. The method further includes determining a document type based on the extracted data and document format data using a machine learning algorithm. The method also includes verifying the extracted data based on the determined document type and user data. The method also includes identifying a triggering event corresponding to the document based on the extracted data and the user data. The method further includes generating a notification corresponding to the triggering event.

An information obtaining method includes obtaining an image of a design printed on a printing target. The design includes an identifier corresponding to specific information. The information obtaining method further includes identifying the identifier included in the design based on the image; and obtaining the specific information corresponding to the identifier based on the identifier.

An information obtaining method includes obtaining an image of a design printed on a printing target. The design includes an identifier corresponding to specific information. The information obtaining method further includes identifying the identifier included in the design based on the image; and obtaining the specific information corresponding to the identifier based on the identifier.

Systems and methods of generating synthetic training data for machine learning models. First, line items in source documents such as bills, invoices, and or receipts are identified and labeled. The identification and labeling generate labeled documents. Then, in the labeled documents, the line items are augmented by adding, deleting, and or swapping line items to generate synthetic training documents. An addition operation randomly selects one or more line items and adds the selected line item(s) to the same labeled document or another labeled document. A deletion operation randomly deletes one or more line items. A swapping operation randomly swaps line items in a single labeled document or across different labeled documents. These operations can generate synthetic labeled documents of any length, which form synthetic training data for training the machine learning models.

Systems and methods of generating synthetic training data for machine learning models. First, line items in source documents such as bills, invoices, and or receipts are identified and labeled. The identification and labeling generate labeled documents. Then, in the labeled documents, the line items are augmented by adding, deleting, and or swapping line items to generate synthetic training documents. An addition operation randomly selects one or more line items and adds the selected line item(s) to the same labeled document or another labeled document. A deletion operation randomly deletes one or more line items. A swapping operation randomly swaps line items in a single labeled document or across different labeled documents. These operations can generate synthetic labeled documents of any length, which form synthetic training data for training the machine learning models.

A method for outputting a drawing reference number description regarding a patent drawing reference number according to an embodiment of the present disclosure may include recognizing a size of a patent drawing and a position of a drawing reference number included in the patent drawing and acquiring a relative position coordinate of the drawing reference number in the patent drawing; setting a relative position coordinate of the drawing reference number description corresponding to the drawing reference number based on the acquired relative position coordinate; and outputting the drawing reference number description on the set relative position coordinate, thereby outputting such that the drawing reference number description corresponds to the drawing reference number.

A method for outputting a drawing reference number description regarding a patent drawing reference number according to an embodiment of the present disclosure may include recognizing a size of a patent drawing and a position of a drawing reference number included in the patent drawing and acquiring a relative position coordinate of the drawing reference number in the patent drawing; setting a relative position coordinate of the drawing reference number description corresponding to the drawing reference number based on the acquired relative position coordinate; and outputting the drawing reference number description on the set relative position coordinate, thereby outputting such that the drawing reference number description corresponds to the drawing reference number.

The present disclosure relates to a handwritten text recognition method, including: acquiring an information sequence including a plurality of track points of handwritten text, wherein information on each track point comprises its abscissa, writing time and writing state value; dividing the plurality of track points into a plurality of strokes according to the writing state value of each track point, the writing state value including a first value representative of stroke pen-up and a second value representative of stroke pen-down, respectively; calculating a first segmentation threshold of the handwritten text; determining a first text segmentation point according to a result of comparison between an absolute value of a difference between abscissas of a start track point of one stroke and an end track point of its previous stroke and the first segmentation threshold; and performing text segmentation according to the first text segmentation point to obtain a text segmentation result.

The present disclosure relates to a handwritten text recognition method, including: acquiring an information sequence including a plurality of track points of handwritten text, wherein information on each track point comprises its abscissa, writing time and writing state value; dividing the plurality of track points into a plurality of strokes according to the writing state value of each track point, the writing state value including a first value representative of stroke pen-up and a second value representative of stroke pen-down, respectively; calculating a first segmentation threshold of the handwritten text; determining a first text segmentation point according to a result of comparison between an absolute value of a difference between abscissas of a start track point of one stroke and an end track point of its previous stroke and the first segmentation threshold; and performing text segmentation according to the first text segmentation point to obtain a text segmentation result.

Systems and methods for distributed ledger-based check verification are disclosed. In one embodiment, a method may include a bank backend computer program: (1) receiving, from a computer application executed by an electronic device, an image of a presented check as part of an electronic check deposit process; (2) performing optical character recognition on the image of the presented check; (3) generating a text file based on the optical character recognition; (4) querying a distributed ledger in a distributed ledger network to determine whether the presented check has been presented or cleared before; (5) determining that the presented check has not been presented or cleared before; (6) processing the presented check for deposit; and (7) writing the text file for the presented check to the distributed ledger.