Abstract: A computer-implemented method, apparatus and program product use the spatial locations of words identified in an unstructured document to both reconstruct lines in the unstructured document and vertically partition the unstructured document. Key/value pairs may then be generated from one or more of the reconstructed lines by using one or more words to one side of the vertical partition as keys and using one or more words to the other side of the vertical partition as values.

Abstract: A system for electronic data interchange (EDI) management includes a memory for storing the EDI document data and a machine learning model representing a set of features of EDI documents and a corresponding status. The system further includes a processor and a non-transitory computer readable medium storing instructions for: accessing an EDI file, the EDI file comprising envelope metadata for an envelope and a first EDI document; and translating the EDI file into a first translated EDI document containing the envelope metadata and a set of EDI document data extracted from the first EDI document, the first translated EDI document formatted according to a hierarchical structure comprising attributes translatable into features processable by the machine learning model to determine a status of the first EDI document.

Abstract: A processor may receive first issue data. The first issue data may be associated with input data entered by a user into a user interface on the issue submission application. The processor may analyze the first issue data. The processor may select a first set of prompted issue descriptions. The first set of prompted issue descriptions may be selected based on analyzing the first issue data. The processor may prompt the user to select a subset of the first set of prompted issue descriptions. The processor may receive from the user a selected subset of the first set of prompted issue descriptions. The processor may output an identified issue description. The identified issue description may be generated based on the selected subset of the first set of prompted issue descriptions.

Abstract: An auditory signal is received by a wearable contact lens. The auditory signal is received from a user. The receiving occurring during a listening mode. The wearable contact lens positioned adjacent to an eye of the user during the receiving. A connection is made from the wearable contact lens to a paired computing device of the user. The connection is through a personal network connection. The auditory signal is transferred from the wearable contact lens, through the personal network connection, to the paired computing device of the user.

Abstract: An Artificial Intelligence (AI)-based regulatory data processing system accesses a regulatory text corpus for training machine learning (ML) models including a topic extraction model, a feature selection model, an entity identification model and a section classification model. The regulatory text corpus includes documents pertaining to a specific domain corresponding to a received domain-specific regulatory text document. Various trained machine learning (ML) models are used to extract topics, identify entities from the new regulatory document and to classify portions of the domain-specific regulatory text document into one of a plurality of predetermined sections. The information in the new regulatory document is therefore converted into machine consumable form which can facilitate automatic execution of downstream processes such as identification of actions needed to implement the regulations and robotic process automation (RPA).

Abstract: Systems, methods, and products for auto tagging structured PDF documents that do not have accessibility tags. In one embodiment, structured PDF documents having accessibility tags are first parsed and analyzed to organize the visual components of the documents. The relationships of the identified objects to DOM elements (e.g., tags) are determined, and the objects and related DOM elements are stored in training files. The training files are used to train various classifiers. Untagged PDF documents are then parsed to identify included visual objects, and the classifiers are used to determine DOM elements that should be associated with visual objects identified in the untagged PDF documents. This information is used to construct a DOM structure corresponding to each untagged document. A new PDF is then generated corresponding to each untagged document using the generated DOM structure and visual object information.

Abstract: A computer-implemented method and system may include receiving at an e-commerce platform a change request that results in a master configuration change for rendering page data, where the master configuration change results in a change from a first master configuration to a second master configuration; determining by an override resolution engine, override data of the first master configuration that affects the second master configuration; and applying, by the override resolution engine, the determined override data to the second master configuration.

Abstract: Methods, systems, and computer-readable storage media for providing, by a machine learning (ML) platform, a first binary classifier, processing, by the first binary classifier a super-set of invoices to provide a plurality of sets of invoices based on matching pairs of invoices in the super-set of invoices, providing, by the ML platform, a second binary classifier, processing, by the second binary classifier, a bank statement and the plurality of sets of invoices to define two or more super-invoices based on aggregate features of invoices in the plurality of sets of invoices, and match the bank statement to a super-invoice of the two or more super-invoices, and outputting a match of the bank statement to the super-invoice.

Abstract: Among pieces of attribute information contained in a document, user-designated attribute information and other attribute information that may give additional value to the user-designated attribute information are accurately and efficiently presented.

Abstract: A computer-implemented method of encoding a word for use in a method of text analysis comprises receiving input text to be analysed, the input text comprising a first word which is not represented in a vocabulary set stored on a storage. The vocabulary set comprises a plurality of words and an associated word embedding vector for each word in the set. The method comprises identifying the first word as a word which is not represented in the vocabulary set and determining one or more sub-words within the first word with which to encode the first word. Each of the one or more sub-words corresponds with a word represented in the vocabulary set and having an embedding vector in the vocabulary set. The method comprises determining an encoding for the first word based on the one or more sub-words.

Abstract: A computer-implemented method, computer program product and computing system for identifying one or more portions of a website structure of a specific website; and associating the one or more portions of the website structure with one or more descriptors of the specific website to define a specific data description model corresponding to the specific website.

Abstract: A device occupation method includes a first device that obtains information related to the first device or information related to a second device, where the second device occupies a first to-be-occupied device, and the first device prepares to occupy the first to-be-occupied device, and occupying, by the first device, the first to-be-occupied device when the information matches.

Abstract: Methods, systems, and non-transitory computer readable storage media are disclosed for training a machine-learning model utilizing batchwise weighted loss functions and scaled padding based on source density. For example, the disclosed systems can determine a density of words or phrases in digital content from a digital content source that indicate an affinity towards one or more content classes. In some embodiments, the disclosed systems can use the determined source density to split digital content from the source into segments and pad the segments with padding characters based on the source density. The disclosed systems can also generate document embeddings using the padded segments and then train the machine-learning model using the document embeddings. Furthermore, the disclosed system can use batchwise weighted cross entropy loss for applying different class weightings on a per-batch basis during training of the machine-learning model.

Abstract: A device configured to retrieve a markup language file; identify, based on the markup language file, an item to be used on a web page, address information for retrieving an object for the item, and an object type corresponding to the object; retrieve an evaluation result for the object; identify the object as an object to be used based on the evaluation result; identify address information of the object; and retrieve and execute the object based on the address information.

Abstract: In one aspect, an example method includes (a) determining, via a thermal conductivity sensor of an automobile damage detection device, one or more thermal conductivities at one or more locations on an automobile; (b) transmitting, via a network interface of the automobile damage detection device, a request for anticipated thermal conductivity data from an automobile claims system, wherein the anticipated thermal conductivity data corresponds to anticipated thermal conductivities at the one or more locations on the automobile; (c) in response to transmitting the request, receiving, via the network interface from the automobile claims system, the anticipated thermal conductivity data; and (d) in response to receiving, from the automobile claims system, the anticipated thermal conductivity data, displaying, via a graphical user interface, a graphical representation of the determined one or more thermal conductivities and the anticipated thermal conductivity data.

Abstract: A voice assistant is provided whereby, in response to receipt of a voice instruction for requesting a recommended piece of content from a user, the voice assistant receives, from a system, a response including information on a recommended piece of content determined using another playback history having a feature similar to a feature of playback histories stored in association with a user ID of the user, and outputs the piece of content.

Abstract: Technologies pertaining to electronic document understanding are described herein. A document is received, wherein the document includes a section of a type. An image of the document is generated, and a candidate region is identified in the image of the document, wherein the candidate region encompasses the section. A label is assigned to the candidate region based upon text of the section, wherein the label identifies the type of the section. An electronic document understanding task is performed based upon the label assigned to the candidate region.

Abstract: A method may include determining, by a computing device and based on at least one user coherency factor, a user coherency level. The coherency level may include a predicted ability of a user to comprehend information. The method may also include determining, by the computing device and based on the user coherency level, information having a complexity that satisfies the predicted ability of the user to comprehend information. The method may further include outputting, by the computing device, at least a portion of the information.

Abstract: Techniques and solutions are described for determining association rules using a machine learning technique on a subset of data to which the association rules might apply, and from which they can be determined. In particular, association rules are determined by tracking changes to attribute values of data objects having a type. The changed attribute value can be used as a consequent in an association rule. Values of other attributes of data objects having the changed attribute value can be used as antecedents in association rules. Values used in antecedents can be constrained, such as by limiting values to those associated with scope attributes or values satisfying a threshold occurrence frequency. In some cases, determined association rules can be automatically implemented, such as to process input or stored data for data objects of the type.

Abstract: A document is represented as a node tree in a document processing system. Edits to a node are represented in a change record that has a one-way link to the node. A text mark has a one-way link to the change record. It deletes that link when the changes represented by the change record are reflected in the text mark. A memory management system releases the memory allocated to the change record when no other object links to it.