Abstract: There are provided systems and methods for sentence level dialogue summaries using unsupervised machine learning for keyword selection and scoring. A service provider, such as an electronic transaction processor for digital transactions, may provide live chat service channels for assistance through live agents and chatbot services. When interacting with these channels over a period of time, a user may create a dialogue with multiple different live agents and/or chatbots. These chat sessions may be asynchronous but have important and relevant information for future live agents to review. Thus, the service provider may provide a dialogue summarizer, which may summarize a dialogue automatically to a number of most relevant sentences. This may be done using an unsupervised machine learning system that utilizes different machine learning algorithms to select and rank keywords within the sentences. The sentences are then scored and highest scored sentences are output as a summary.

Abstract: In some aspects, the techniques described herein relate to a method including: obtaining, by a computing device, a set of messages, each message including message metadata and message structure, at least one message of the set of messages comprising a digital asset; clustering, by the computing device, the set of messages based on the message metadata and the message structure; classifying, by the computing device, the clusters into categories; determining, by the computing device, a digital asset structure from at least one message cluster; and generating, by the computing device, at least one mapping rule based on the digital asset structure.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating coachmarks and concise instructions based on operation descriptions for performing application operations. For example, the disclosed systems can utilize a multi-task summarization neural network to analyze an operation description and generate a coachmark and a concise instruction corresponding to the operation description. In addition, the disclosed systems can provide a coachmark and a concise instruction for display within a user interface to, directly within a client application, guide a user to perform an operation by interacting with a particular user interface element.

Abstract: A device receives image data of a contractual document that includes an offer including terms of a proposed transaction, converts the image data to text data that identifies text within the contractual document, and receives preferences information for a recipient of the offer. The device identifies key terms within the contractual document by using term identification to analyze the text. The key terms may include a first key term that identifies subject matter of the proposed transaction and other key terms that are part of the offer. The device determines term scores that correspond to likelihoods of the other key terms being favorable to the recipient by using a data model to analyze the key terms and the preferences information. The device, based on the term scores, generates and provides another device with a recommendation to be used in determining whether the accept the offer.

Abstract: Forward-looking sales activities are poorly served by existing Customer Relationship Managements systems, as the customer information those systems maintain by definition does not include the speculatively interested “leads” pursued in initial sales contacts. An Enrichment database, asynchronously loaded with corporate and optionally individual contact information in anticipation of use, provides real-time access to information in support of such first customer contacts. The Enrichment information may also be used to expedite creation and revision of CRM records.

Abstract: A system in which existing email protocols are leveraged to exchange information between two or more client devices. In some examples, a trained classifier can be leveraged to apply a label or classification to a received email. Based on the classification of the email, one or more parsers can be selected which can output data to be displayed in a dedicated portion of a graphical user interface.

Abstract: Disclosed herein are systems and methods capable of performing text exploration on large volume of corpus without prior knowledge in an accurate and efficient manner and may also provide any number of additional or alternative benefits and advantages. In particular, embodiments described herein provide a text exploration executable environment that uses unsupervised machine-learning to assist a human analyst with distilling key emerging themes from a corpus of hundreds or thousands of text files presented in a time series graphical user interface (GUI). A document may be a unit of text under analysis received from a particular data source, such as word-processing documents, paragraphs, sentences, chat sessions, speech-to-text call segments, online texts, social media postings (e.g., Tweets®), and other machine-readable text.

Abstract: A device receives image data of a contractual document that includes an offer including terms of a proposed transaction, converts the image data to text data that identifies text within the contractual document, and receives preferences information for a recipient of the offer. The device identifies key terms within the contractual document by using term identification to analyze the text. The key terms may include a first key term that identifies subject matter of the proposed transaction and other key terms that are part of the offer. The device determines term scores that correspond to likelihoods of the other key terms being favorable to the recipient by using a data model to analyze the key terms and the preferences information. The device, based on the term scores, generates and provides another device with a recommendation to be used in determining whether the accept the offer.

Abstract: A computer-implemented summary generation method includes obtaining input text; generating an initial lattice including serially coupled nodes corresponding to words within the input text; generating a node corresponding to an expression within the initial lattice; adding the generated node to the initial lattice to provide an extended lattice corresponding to the input text; calculating a generation probability of each word within the input text using a dictionary and a machine learning model (model); calculating a generation probability for each node included in the extended lattice based on a hidden state output by a cell corresponding to the node among cells in an encoder of the model and a hidden state updated by a cell in a decoder of the model; and generating an element of a summary of the input text based on the generation probability of each word and the generation probability of each node of the extended lattice.

Abstract: A method, computer program product, and a system where a processor(s) monitors, via sensor(s) configured to monitor physical behaviors of a user engaged in reading document(s) on a user interface of a client, where the monitoring captures the physical behaviors and portions of the one or more processors being read during the physical behaviors. The processor(s) analyses the physical behaviors and the captured portions to generate implicit feedback indicating amounts of time the user engaged with each portion of the one or more documents during the monitoring. The processor(s) generates a model profile for the user including relative timing measures indicating ranked levels of engagement of the user with the portions and ranked levels of importance of each portion.

Abstract: Methods, systems, and computer-readable storage media for generating document representations of documents in a set of documents based on sentence embeddings that are provided using a recurrent neural network (RNN) encoder, each document including an electronic document, generating aspect representations based on sentences included in documents of the set of documents, and comment documents in a set of comment documents, determining a first topic representation based on the document representations, determining a second topic representation based on aspect saliences with respect to the first topic representation, calculating salience scores, each salience score associated with a respective sentence, and calculated based on a set of initial salience scores, and a respective aspect salience score, and generating a summary of the set of documents based on the salience scores, the summary including one or more sentences included in documents of the set of documents.

Abstract: A method comprising receiving digital documents, a query statement, and a summary length constraint; identifying, for each of said digital documents, a sentence subset, based, at least in part, on said query statement, a modified version of said summary length constraint, and a first set of quality objectives, generating, for each of said sentence subsets, a random forest representation; iteratively (i) sampling, from each of said random forest representations, a plurality of tokens to create a corresponding candidate document summary, based, at least in part, on weights assigned to each of said tokens, (ii) assigning a quality ranking to said candidate document summary, based, at least in part, on said first set of quality objectives and a second set of quality objectives, and (iii) adjusting said weights, based, at least in part, on said quality rankings; and outputting a highest ranking said candidate document as a compressed summary.

Abstract: An interest determination system includes a data acquisition unit configured to acquire a first data group indicating a feature of each word contained in user's dialog information; a first learning unit configured to learn by using a first input and output values, the first input value being set to a data group that associates the first data group with a second data group indicating a feature of a predetermined word, the first output value being set to information indicating whether there is a user's interest; and a determination unit configured to, when a data group that associates a third data group indicating a feature of a word to be determined with the first data group has been input to the first learning unit, determine whether there is a user's interest in the word to be determined, based on an output value output from the first learning unit.

Abstract: Systems and methods are provided for displaying content on a mobile electronic communications device having a curved display surface that includes a front screen area and a plurality of edge screen areas. By way of example, the described techniques may entail determining whether the content is suitable for display on both the front screen area and the plurality of edge screen areas, displaying unsuitable content only on the front screen area and displaying suitable content on both the front screen area and the plurality of edge screen areas when the plurality of edge screen areas are unobstructed, and otherwise, displaying the suitable content on only the front screen area.

Abstract: Forward-looking sales activities are poorly served by existing Customer Relationship Managements systems, as the customer information those systems maintain by definition does not include the speculatively interested “leads” pursued in initial sales contacts. An Enrichment database, asynchronously loaded with corporate and optionally individual contact information in anticipation of use, provides real-time access to information in support of such first customer contacts. The Enrichment information may also be used to expedite creation and revision of CRM records.

Abstract: Methods, systems, and computer-readable storage media for processing, through an encoder of an attention-reconstruction-based summarization (ARS) platform, the primary document to provide a contextual vector, identifying, using reconstruction regularizers of the ARS platform, information that is absent from the one or more linking documents, and is present in the primary document at least partially based on the contextual vector, for each word of the primary document, providing, using an attention mechanism of the ARS platform, a word salience score at least partially based on the information that is absent from the one or more linking documents, for each sentence in the primary document, determining a sentence salience score based on word salience scores of words within the sentence, ranking sentences of the primary document based on sentence salience scores, and selecting two or more sentences of the primary document based on ranking to provide a summary of the primary document.

Abstract: Multimedia document summarization techniques are described. That is, given a document that includes text and a set of images, various implementations generate a summary by extracting relevant text segments in the document and relevant segments of images with constraints on the amount of text and number/size of images in the summary.

Abstract: A system for reuse of content within documents can include storage media on which a document graph is stored, and indexing software that, when executed by a processing system, directs the processing system to at least: identify content component types within a source document; and represent content components in a document graph at least connected by ‘contains’ edge to source document. A search of the document graph for reusable content can then be directed to related content components including those of a particular content component type or relevancy.

Abstract: In general, embodiments of the present invention provide systems and computer readable media for implementing a single data integration platform that supports multiple data access interfaces to a single corpus of stored dynamic data collected from multiple data sources. In embodiments, the data integration platform includes a record tables layer that stores a group of data records and supports a CRUD interface for accessing the data records; a resolution mapping layer that stores a set of entities generated by a many-to-one mapping of data records to entities using entity resolution; and an entities layer that stores resolved entities which may be accessed via either a search interface based on search criteria or a hybrid search interface that supports “get via record id” queries.

Abstract: The present disclosure relates to a response analysis system that employs a small-data training dataset to train a neural network that accurately performs domain-agnostic opinion mining. For example, in one or more embodiments, the response analysis system trains a response classification neural network using part of speech information (e.g., syntactic information) to learn and apply response classification labels for opinion text responses. In particular, the response analysis system employs part of speech information patterns without regard to word patterns to determine whether words in a text response correspond to an opinion, the target of the opinion, or neither. In addition, the trained response classification neural network has a significantly reduced learned parameter space, which decreases processing, memory requirements, and overall complexity.

Abstract: A terminal device such as a smartphone displays a first image on a display unit, and outputs a second image including the first image as a part to a display device such as a navigation device. Also, the terminal device obtains relative position information indicating a relative positional relation of the first image with respect to the second image from an external server. When a user makes a touch input to the second image displayed on the display device, the input position is transmitted from the display device to the terminal device. The terminal device obtains the conversion necessity information indicating whether or not the input position on the second image needs to be converted based on the relative position information, and converts the input position on the second image to the input position on the first image based on the relative position information when the conversion is necessary.

Abstract: A portion data structure representing a portion extracted from a formatted source document is described. A portion data structure contains a first subtree of nodes that is modeled after a second subtree of a complete hierarchical representation of the formatted source document. Explicit formatting attribute values are specified for nodes of the first subtree only where a value calculated for the formatting attribute in a node of the first subtree differs from a value calculated for the formatting attribute in the corresponding node in the second subtree at a time when the node of the first subtree descends from a reset node specifying standardized formatting attribute values. The contents of the portion data structure are usable to display the portion extracted from the formatted source document in a context other than the formatted source document.

Abstract: The present disclosure generally relates to systems and processes for emoji word sense disambiguation. In one example process, a word sequence is received. A word-level feature representation is determined for each word of the word sequence and a global semantic representation for the word sequence is determined. For a first word of the word sequence, an attention coefficient is determined based on a congruence between the word-level feature representation of the first word and the global semantic representation for the word sequence. The word-level feature representation of the first word is adjusted based on the attention coefficient. An emoji likelihood is determined based on the adjusted word-level feature representation of the first word. In accordance with the emoji likelihood satisfying one or more criteria, an emoji character corresponding to the first word is presented for display.

Abstract: A method and an invoice analytics system (IAS) for analyzing an entity's business service transactional invoice data (BSTID) and performing invoice analytics (IA) in real time are provided. The IAS extracts contract line item (CLI) data from contracts and stores the CLI data in a CLI database. The IAS transforms the CLI data into a quick view format and generates a contract quick view interface that displays the CLI data on an entity device. The IAS extracts and segments invoice line item (ILI) data from aggregated BSTID, stores the ILI data in an ILI database, and reconciles the invoices in accordance with the contracts by identifying and rectifying errors and off-contract business service items. The IAS performs IA on the ILI data for computing cost savings and generates an interactive, dynamic, and searchable IA report including graphical data representations of purchase patterns, outliers, variations, and predictive analytics for the entity.

Abstract: Provided is an operating method of a terminal. The method includes: displaying a contents list on a display of the terminal, the contents list including a plurality of items, each of the plurality of items corresponding to a respective one of a plurality of contents; selecting at least two items from among the plurality of items in the displayed contents list in response to receiving of a first input, the selected at least two items displayed distinguishably from the rest of the plurality of items that is not selected in the contents list; and displaying a first content among the plurality of selected contents in response to receiving of a second input, the first content corresponding to a first item among the selected at least two items.

Abstract: Methods, devices and systems are provided for provisioning an output device for use with a media device to produce synchronized audio and video portions of media content. An exemplary method involves a media device identifying a device type associated with the output device, obtaining synchronization settings for the device type from a server via a network, and presenting audiovisual content in accordance with the synchronization settings obtained from the server.

Abstract: A terminal device such as a smartphone displays a first image on a display unit, and outputs a second image including the first image as a part to a display device such as a navigation device. Also, the terminal device obtains relative position information indicating a relative positional relation of the first image with respect to the second image from an external server. When a user makes a touch input to the second image displayed on the display device, the input position is transmitted from the display device to the terminal device. The terminal device obtains the conversion necessity information indicating whether or not the input position on the second image needs to be converted based on the relative position information, and converts the input position on the second image to the input position on the first image based on the relative position information when the conversion is necessary.

Abstract: An apparatus comprises a processing platform configured to implement a cluster labeling system for documents comprising unstructured text data. The cluster labeling system comprises a clustering module and a visualization module. The clustering module implements a topic model generator and is configured to assign each of the documents to one or more of a plurality of clusters based at least in part on one or more topics identified from the unstructured text data using at least one topic model provided by the topic model generator. The visualization module comprises multiple view generators configured to generate respective distinct visualizations of a selected one of the clusters. The multiple view generators include at least a bigram view generator configured to provide a visualization of a plurality of term pairs from the selected cluster, and a summarization view generator configured to provide a visualization of representative term sequences from the selected cluster.

Abstract: System and methods are provided for detecting numerical text strings within a text string and converting those numerical text strings into digit strings. The digit strings may be reflected in real-time, such as when the user is typing a text message. If more than one possible format of the digit string is determined, the system may then provide a selection of the various formats for selection. Once the proper format for the digit string is determined, that digit string may replace the numerical string previously detected in the text string. The text to digit conversion and associated formatting expedites user text entry such that the user is not required to switch keyboard views, (e.g., virtual keyboards). Additionally, converting to digit strings compresses message length, as well as provide other benefits.

Abstract: A digital plaque that can display one of several stored certificates that can also retrieve updates to a displayed certificate when needed. The digital plaque is used to display a certificate and associated information. Typically the certificate displayed is one of an award, a recognition, a score card, a picture, a transcript, a document, a sales deed, a ticket, a poster, a document cover, a book cover, a patent application, a granted patent and a graduation record. A collection of such certificates is stored in memory and displayed when selected by a user. Status information and supporting documents associated with the selected certificate can be retrieved from a remote server and portions of the retrieved information can be displayed.

Abstract: A new system for, first, creating a cross-reference between user characteristics and applicable portions of platform requirement documents; and, second, displaying on a case by case basis, only portions of platform requirement documents applicable to those user characteristics, starts with a Subject Matter Expert (SME) creating, from the platform requirement documents, a list of platform: (a) core requirements; (b) conditional physical requirements; and, (c) conditional procedural requirements; then cross-referencing those core requirements, conditional physical requirements and conditional procedural requirements with applicable portions of the platform requirement document; next creating a cross-reference between user characteristics that interact with the platform and their corresponding platform conditional physical requirements and conditional procedural requirements; and, finally, creating a list of those cross-referenced user characteristics.

Abstract: Systems, computer-program products and methods for annotating electronic text documents by resolving abbreviated text of concepts defined in a controlled vocabulary. In one embodiment, a method of annotating an electronic text document includes evaluating, by a computer, the electronic text document for one or more multi-word expressions having a first word and a second word. The method further includes, if a multi-word expression is found in the electronic text document, storing an abbreviation of the multi-word expression and evaluating the electronic text document for occurrences of the abbreviation after a location of the multi-word expression. The method further includes, if an occurrence of the abbreviation is found following the multi-word expression, annotating the occurrence of the abbreviation with the multi-word expression.

Abstract: Advertisers specify the size of an ad in pixels or in physical units such as inches or millimeters. The physical square area of an ad specified in pixels will change as the physical size of the television screen or computer display screen varies. The number of pixels of an ad specified in inches or millimeters will also change as the physical size of the television screen or computer display screen varies. The present invention manages the issue of keeping the physical area size of an image constant across television display screens regardless of the physical size of the screen.

Abstract: To provide a user with an easily understandable help document, the MFP capable of executing a plurality of processes includes a process designation accepting portion to accept designation of at least one of the plurality of processes, a workflow generating portion to generate workflow definition data defining the one or more processes accepted, and a help document generating portion to generate a help document corresponding to the generated workflow. The help document generating portion includes a summary page generating portion to generate a summary page having listed thereon process names for identification of the one or more processes defined by the corresponding workflow definition data.

Abstract: One embodiment of a method for summarizing an electronic document includes splitting the electronic document into a plurality of terms, wherein each of the plurality of terms is associated with a respective length, a respective informativeness score, and a respective coherence score, automatically selecting a subset of the plurality of terms, such that an aggregate informativeness score of the subset is maximized while an aggregate length of the subset is less than or equal to a maximum length, and arranging the subset as a summary of the electronic document.

Abstract: Disclosed are methods and devices for rendering interactions between virtual and physical objects on a substantially transparent display are disclosed. In one embodiment, the method includes displaying a user-interface on a substantially transparent display of a wearable computing device. The method further includes displaying a virtual object in the view region at a focal length along a first line of sight and detecting a physical object at a physical distance along a second line of sight. The method still further includes determining that a relationship between the focal length and the physical distance is such that the virtual object and the physical object appear substantially co-located in a user-view through the view region and, responsive to the determination, initiating a collision action between the virtual object and the physical object.

Abstract: The present invention relates to a data sharing service system, and a method and a device for a data sharing service, and to the data sharing service in a multi-device environment. Particularly, the present invention enables a user to summarize information on content being used or to transmit data on specific content to terminal device of another party by using a messenger, so that the user can gather the contents of interested articles (or books) to manage the contents by compiling statistics and the user can easily notify the other party of the contents of the articles (or books) without requiring the user to write out the interested contents one by one when using the messenger.

Abstract: The subject matter disclosed herein relates to generating an abstract of an electronic document based at least in part on a utility metric associated with a group phrases as well as on a ranked order of sentences.

Abstract: A system and method for obtaining information embedded in unstructured text is provided. The method comprising generating computer-readable annotations based on the unstructured text, at least one of the computer-readable annotations comprising an indication of a linguistic feature. A pair of the computer-readable annotations may be used to generate at least one computer-readable relation between the pair. The annotations and/or relations may be stored as characteristic data structures in a database. A query comprising at least one criterion may be received. In response to the query, an information result may be generated based on at least one of the characteristic data structures stored in the database.

Abstract: A technique provides an electronic summary of source content. The technique involves performing, on the source content, a content recognition operation to electronically generate text output from the source content. The technique further involves electronically evaluating text portions of the text output based on predefined usability criteria to produce a respective set of usability properties for each text portion of the text output. The technique further involves providing, as the electronic summary of the source content, summarization output which summarizes the source content. The summarization output includes a particular text portion of the text output which is selected from the text portions of the text output based on the respective set of usability properties for each text portion of the text output.

Abstract: A computer based system and method for exporting user-selected text by allowing a user to select a plurality of text segments, selecting from a block of text, flagging each of the text segments from a plurality of text segments such that each of text segments may be retrieved later, retrieving the plurality of text segments, and then exporting the plurality of text segments as data that may be read by a software application. The selected text segments may also be assigned one or more indicia for categorizing and sorting according to user-defined criteria.

Abstract: Methods, apparatus, and terminal devices for truncating a character string are provided. A character string to be truncate is read and a byte length occupied by a current character is obtained. It is determined whether a substring, formed from a starting character of the character string to be truncated to the current character, satisfies a pre-set truncating condition. When the substring, formed from the starting character of the character string to be truncated to the current character, satisfies the pre-set truncating condition, the current character is subtracted from the substring to generate a truncated substring for displaying, according to the byte length occupied by the current character.

Abstract: A method of determining main text in a mark-up document is provided, which comprises determining a length of each paragraph in the mark-up document; and determining one or more main paragraphs of the mark-up document based upon the length of the paragraphs in the mark-up document.

Abstract: Assigning sentiment to themes is disclosed. Reputation data extracted from at least one data source is received. The reputation data includes user-authored reviews. The user-authored reviews include text and at least one rating. For a first review included in the reputation data, at least one keyword is determined using the first review's text. A sentiment is assigned for a theme associated with the keyword based at least in part on the first review's rating.

Abstract: Embodiments include notes application features including notes page preview and other features, but the embodiments are not so limited. In an embodiment, a computer-based method can be configured to generate a notes page preview based on a source notes page, wherein the notes page preview includes one or more of a title, a thumbnail, and/or a snippet generated from the source notes page. A handheld device of an embodiment includes a notes application interface configured in part to display and dynamically update one or more notes page previews, wherein each notes page preview includes a plurality of preview display areas including a title display area, a thumbnail display area, and/or a snippet display area. Other embodiments are included.

Abstract: A system and method for constructing a document redundancy graph with respect to a document set. The redundancy graph can be constructed with a node for each paragraph associated with the document set such that each node in the redundancy graph represents a unique cluster of information. The nodes can be linked in an order with respect to the information provided in the document set and bundles of redundant information from the document set can be mapped to individual nodes. A data structure (e.g., a hash table) of a paragraph identifier associated with a probability value can be constructed for eliminating inconsistencies with respect to node redundancy. Additionally, a sequence of unique nodes can also be integrated into the graph construction process. The nodes can be connected to the paragraphs associated with the document set via a hyperlink and/or via a label with respect to each node.

Abstract: An apparatus for editing a document containing a plurality of pages each of which includes one or more first objects is provided. The apparatus includes: an adding portion that adds a second object to each of the pages; an extraction portion that extracts, from the pages, a possible improper page that is a page to which the second object is probably not added properly, the possible improper page being extracted for each cause due to which the second object is not added properly; a quantity display portion that displays, for each cause, a quantity of the possible improper page thus extracted; an excluding portion that excludes, from the possible improper page, a selected page that is a page corresponding to the cause selected by a user; and an informing portion that informs the user of a to-be-confirmed page corresponding to the possible improper page remaining after excluding the selected page.

Abstract: A method and apparatus are provided for summarizing a text message, such as an email message or a transcribed audio message. A portion of each text message, such as a sentence, is extracted as an indicative summary of the text message based on a degree of overlap of words in the sentence with a set of words, such as words in the message subject or words in a related root message. The extracted portion is based on a score for each portion of the text message, such as a sentence. An interface is also provided for presenting the indicative summaries of a set of related text messages to a user.

Abstract: A text content summary is created from speech content. A focus more signal is issued by a user while receiving the speech content. The focus more signal is associated with a time window, and the time window is associated with a part of the speech content. It is determined whether to use the part of the speech content associated with the time window to generate a text content summary based on a number of the focus more signals that are associated with the time window. The user may express relative significance to different portions of speech content, so as to generate a personal text content summary.

Abstract: Information extraction and annotation systems and methods for use in annotating and determining annotation instances are provided herein. Exemplary methods include receiving annotated documents, the annotated documents comprising annotated fields, analyzing the annotated documents to determine contextual information for each of the annotated fields, determining discriminative sequences using the contextual information, generating a proposed rule or a feature set using the discriminative sequences and annotated fields, and providing the proposed rule or the feature set to a document annotator.