Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: A computer-implemented method for automatic template selection for extracting data from an input electronic document is provided. The method includes receiving a first set of candidate templates and an input electronic document. For each candidate template, a template similarity ratio value is calculated that represents a similarity of the candidate template to the input electronic document. The first set of candidate templates are ranked according to the template similarity ratios and then matched to the input electronic document resulting in generating a normalized similarity score for each particular candidate from among the candidate templates. Differences in normalized similarity scores of successive pairs of the candidate templates is determined and a breaking point is established. A second set of candidate templates is formed by selecting candidate templates that are ranked above the breaking point. Data from the input electronic document is extracted using the second set of candidate templates.

Abstract: A computer-implemented method for automatic template selection for extracting data from an input electronic document is provided. The method includes receiving a first set of candidate templates and an input electronic document. For each candidate template, a template similarity ratio value is calculated that represents a similarity of the candidate template to the input electronic document. The first set of candidate templates are ranked according to the template similarity ratios and then matched to the input electronic document resulting in generating a normalized similarity score for each particular candidate from among the candidate templates. Differences in normalized similarity scores of successive pairs of the candidate templates is determined and a breaking point is established. A second set of candidate templates is formed by selecting candidate templates that are ranked above the breaking point. Data from the input electronic document is extracted using the second set of candidate templates.

Abstract: A computer-implemented method for automatic template selection for extracting data from an input electronic document is provided. The method includes receiving a first set of candidate templates and an input electronic document. For each candidate template, a template similarity ratio value is calculated that represents a similarity of the candidate template to the input electronic document. The first set of candidate templates are ranked according to the template similarity ratios and then matched to the input electronic document resulting in generating a normalized similarity score for each particular candidate from among the candidate templates. Differences in normalized similarity scores of successive pairs of the candidate templates is determined and a breaking point is established. A second set of candidate templates is formed by selecting candidate templates that are ranked above the breaking point. Data from the input electronic document is extracted using the second set of candidate templates.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: Described herein is a computer-implemented method for automatic extraction of canonical data from an electronic document. The method comprises classifying a first text rectangle in an electronic document as a label and a second text rectangle as a value using a first machine learning algorithm. A first probability score of a likelihood of the first text rectangle corresponding to a first canonical category is determined using a second machine learning algorithm. A second probability score of a likelihood of the second text rectangle corresponding to a first canonical category is determined using a third machine learning algorithm. A relative spatial position of the second text rectangle relative to the first text rectangle is calculated. Based on the relative spatial position, the first probability score, and the second probability score, the first text rectangle, and the second text rectangle are classified into the first canonical category.

Abstract: Embodiments of the disclosed technologies provide solutions for automatically reading digital electronic documents that contain tables and correctly extracting table data, rows and columns from the documents with high accuracy and high throughput. Embodiments are capable of converting a table portion of a read-only document to a searchable, editable data record using text rectangle (TR)-level numerical data that indicates probabilities of TRs belonging to canonicals and at least one convolutional neural network (CNN) that processes the TR-level numerical data to produce table-level numerical data.

Abstract: A computer-implemented method for automatic template selection for extracting data from an input electronic document is provided. The method includes receiving a first set of candidate templates and an input electronic document. For each candidate template, a template similarity ratio value is calculated that represents a similarity of the candidate template to the input electronic document. The first set of candidate templates are ranked according to the template similarity ratios and then matched to the input electronic document resulting in generating a normalized similarity score for each particular candidate from among the candidate templates. Differences in normalized similarity scores of successive pairs of the candidate templates is determined and a breaking point is established. A second set of candidate templates is formed by selecting candidate templates that are ranked above the breaking point. Data from the input electronic document is extracted using the second set of candidate templates.

Abstract: A computer-implemented method comprises defining a set of canonical features for a document type and a plurality of attributes for a canonical feature; identifying a set of text rectangles from an electronic document; obtaining a comparison set of reference document codifications, one of which comprising a plurality of canonical feature codifications, one of which comprising one or more attribute values for one or more of the plurality of attributes of one of the set of canonical features as the one canonical feature appears in the one reference document; for each current canonical feature of the set of canonical features: selecting a set of canonical feature codifications from the comparison set and identifying a match between one of the set of text rectangles and one of the set of canonical feature codifications; for each of the set of text rectangles, selecting one of the matching canonical feature codifications.

Abstract: Described herein is a computer implemented method for processing an electronic document. The method comprises accessing a comparison set of reference document codifications, each reference document codification in the comparison set comprising a plurality of canonical feature codifications. Each canonical feature codification in each reference document codification in the comparison set is processed by determining whether the electronic document has one or more text rectangles in a potential position of the canonical feature and, in response determining that the electronic document has one or more text rectangles in a potential position of the canonical feature, recording a preliminary association between the or each text rectangle and the canonical feature. For each text rectangle preliminarily associated with one or more canonical features, a final canonical feature assignment is determined for the text rectangle based on the one or more preliminarily associated canonical features.