In some implementations, a device may receive, via a non-textual image, presentation information indicating a set of items, an amount associated with each item included in the set of items, and a total amount associated with an exchange. The device may insert code into a document object model (DOM) of the GUI based on the presentation information to display, via the GUI, one or more user selection options that enable a selection of an item, from the set of items, and a selection of a portion of an amount associated with a selected item. The device may receive, via the GUI, a user selection that indicates one or more selected items and a portion of an amount for the selected item. The device may transmit, to a server device, credential information associated with an exchange medium to cause a transfer associated with the sub-exchange to be completed.

Disclosed herein are system, method, and computer program product embodiments for receive, via a client application executing on a client device, a request for a dynamically rendered formatted document based on a subset of display fields. The method includes selecting a template from a plurality of templates based at least partially on the subset of display fields, converting the display fields of the template based on the subset of display fields, transforming the display fields to integrate data into the template, and generating a formatted document or a formatted fillable document for display.

Disclosed herein are system, method, and computer program product embodiments for receive, via a client application executing on a client device, a request for a dynamically rendered formatted document based on a subset of display fields. The method includes selecting a template from a plurality of templates based at least partially on the subset of display fields, converting the display fields of the template based on the subset of display fields, transforming the display fields to integrate data into the template, and generating a formatted document or a formatted fillable document for display.

Systems, methods, devices, and non-transitory, computer-readable storage media are disclosed for matching a service requester with a service provider via a taxonomy based directed graph. The method includes: receiving a keyword associated with a service; accessing a directed graph including a root node and nodes connected by edges, each node having a title; identifying a second node of the directed graph for each of service providers, each second node having a title matching a skill of a respective service provider; determining a distance between the first node and each second node along the directed graph; and ranking the service providers based at least in part on the distance between the first node and each second node. Systems, methods, devices, and non-transitory, computer-readable storage media are further disclosed for determining and storing a quality score for the revised linguistic content.

Embodiments are directed at processing language content by a method of bi-directional conversion between language content with additional information to and from documents, using a readable text and a text grammar. A method combines additional information with the language content using punctuation idioms. The combined language content and additional information remains readable by one ordinarily skilled in the art of reading and also remains allowable according to a text grammar; that is embodiments are rigorous and may be declarative. The document is compliant with a format drawn from a set which comprises SGML, XML, TEI, HTML, DOC, DOCX, ODX, PDF and XPS. The document is publishable in a medium drawn from a set which comprises a book, a magazine, a journal, a newspaper, an article and a web page. A computer-readable memory device and a computing device are also claimed.

A method for generating presentation slides with distilled content including receiving one or more data files as source material for slide generation, obtaining content from the one or more data files for a slide of a slide presentation, identifying a layout template for the slide based on the content, and distilling the content into distilled content to generate a presentation visualization item based on the distilled content. The distilled content may include a subset of the content. The method may also include generating the slide based on the presentation visualization item and the layout template.

A non-transitory processor-readable medium stores instructions to be executed by a processor. The instructions cause the processor to receive a first trained machine learning model that generates a transcription based on a document. The instructions cause the processor to execute the first trained machine learning model and a second trained machine learning model to generate a refined transcription based on the transcription. The instructions cause the processor to execute a quality assurance program to generate a transcription score based on the document and the transcription. The instructions cause the processor to execute the quality assurance program to generate a refined transcription score based on the refined transcription and at least one of the document or the transcription. The at least one refined transcription score indicates an automation performance better than an automation performance for the at least one transcription score.

A non-transitory processor-readable medium stores instructions to be executed by a processor. The instructions cause the processor to receive a first trained machine learning model that generates a transcription based on a document. The instructions cause the processor to execute the first trained machine learning model and a second trained machine learning model to generate a refined transcription based on the transcription. The instructions cause the processor to execute a quality assurance program to generate a transcription score based on the document and the transcription. The instructions cause the processor to execute the quality assurance program to generate a refined transcription score based on the refined transcription and at least one of the document or the transcription. The at least one refined transcription score indicates an automation performance better than an automation performance for the at least one transcription score.

According to principles described herein, unsupervised statistical models, semi-supervised data models, and HITL methods are combined to create a text normalization system that is both robust and trainable with a minimum of human intervention. This system can be applied to data from multiple sources to standardize text for insertion into knowledge bases, machine learning model training and evaluation corpora, and analysis tools and databases

Techniques are disclosed relating to determining locations of child elements in a user interface (UI), automatically updating the locations of the child elements in the UI, and generating an updated UI with the updated locations. A computer system may receive information specifying content of the UI and render an image of the UI based on the received information. Computer vision algorithms may be applied to the rendered of the UI to determine locations of elements in the UI. The locations of the elements may be updated to position the elements in selected locations relative to each other. The updated locations of the elements as determined by the computer vision algorithms may be implemented to generate updating information specifying content for the updated UI.