1096results about "Natural language analysis" patented technology

Various embodiments provide a method that comprises receiving a set of segments from a text field, analyzing the set of segments to determine at least one of a target subtext or a target meaning associated with the set of segments, and identifying a set of candidate emoticons where each candidate emoticon in the set of candidate emoticons has an association between the candidate emoticon and at least one of the target subtext or the target meaning. The method may further comprise presenting the set of candidate emoticons for entry selection at a current position of an input cursor, receiving an entry selection for a set of selected emoticons from the set of candidate emoticons, and inserting the set of selected emoticons into the text field at the current position of the input cursor.

A computer assisted design system (100) that includes a computer system (102) and text input device (103) that may be provided with text elements from a keyboard (104). A user may also provide oral input (107) to the text input device (103) or to a voice recognition software with in-built artificial intelligence algorithms (110) which can convert spoken language into text elements. The computer system (102) includes an interaction design heuristic engine (116) that acts to understand and translate text and language into a visual form for display to the end user.

The present disclosure relates to a digital device and a method for controlling the same, which displays an application related to a text message, extracts additional information from the application and inserts the extracted additional information to the text message, in order to provide additional information associated with a text message.

A method and system identifying the language of a textual passage is disclosed. The method and system includes parsing the textual passage into n-grams and assigning an initial weight to each n-gram, and adjusting the weight initially assigned to a word or n-gram parsed from the textual passage. The initially assigned weight is adjusted in a manner proportionate to the inverse of the number of languages within which such words or n-grams appear. Reducing the weight assigned to such words or n-grams diminishes—without completely eliminating—their importance in comparison to other words or n-grams parsed from the same textual passage when determining the language of a passage. The method and system of the present invention appropriately weighs the short words or n-grams common to multiple languages without affecting the short words or n-grams that are uncommon to several languages.

A method and system identifying the language of a textual passage is disclosed. The method and system includes parsing the textual passage into n-grams and assigning an initial weight to each n-gram, and adjusting the weight initially assigned to a word or n-gram parsed from the textual passage. The initially assigned weight is adjusted in a manner proportionate to the inverse of the number of languages within which such words or n-grams appear. Reducing the weight assigned to such words or n-grams diminishes—without completely eliminating—their importance in comparison to other words or n-grams parsed from the same textual passage when determining the language of a passage. The method and system of the present invention appropriately weighs the short words or n-grams common to multiple languages without affecting the short words or n-grams that are uncommon to several languages.

Systems, apparatuses, and methods for automating avatar selection for groups are disclosed. A representative method includes providing a software application accessible to multiple users, and a database of available avatar characteristics that are selectively combinable to create user-specific avatars with which the users can represent themselves to each other in the application. User(s) associated with a group provide an instruction that associates a first set of the available avatar characteristics with the group. For users associated with the group, access is then restricted to the first set of avatar characteristics, and for users not associated with the group, access is restricted to a different second set of available avatar characteristics. Avatar characteristics in the first set may be selected by an automated procedure that analyzes user-specific avatars of some users, and selects which of the available avatar characteristics to include in the first set based on the analysis.

A context-free grammar can be represented by a weighted finite-state transducer. This representation can be used to efficiently compile that grammar into a weighted finite-state automaton that accepts the strings allowed by the grammar with the corresponding weights. The rules of a context-free grammar are input. A finite-state automaton is generated from the input rules. Strongly connected components of the finite-state automaton are identified. An automaton is generated for each strongly connected component. A topology that defines a number of states, and that uses active ones of the non-terminal symbols of the context-free grammar as the labels between those states, is defined. The topology is expanded by replacing a transition, and its beginning and end states, with the automaton that includes, as a state, the symbol used as the label on that transition. The topology can be fully expanded or dynamically expanded as required to recognize a particular input string.

A parse / build engine that can handle multi-format financial messages. The engine converts the different format messages into a common format, and the common format message is then processed by the business service application. A parser examines the message and determines an appropriate schema for the particular format of message received. The schema is a data structure in a schema registry that includes a grammar structure for the received format as well as pointers to handlers for converting the different fields of the message into the internal message format using the grammar structure (the “grammar” can include field sequence, field type, length, character encoding, optional and required fields, etc.). The handlers are individually compiled. As formats change, new formats or changes to old formats can be dynamically added to the parse / build engine by loading new schema and handlers.

A method for a computer system includes receiving an input source comprising a plurality of sentences, determining a set of semiotic relationships from the plurality of sentences, performing a contextual correlation in response to the set of semiotic relationships and in response to an ideas index to determine a reduced set of semiotic relationships from the set of semiotic relationships, omparing the reduced set of semiotic relationships to each of a plurality of semiotic relationships associated with a set of documents to determine comparison values, determining a reduced set of documents from the set of documents in response to the comparison values, and outputting portions of the reduced set of documents to a user.

A method is provided for automatically classifying text into categories. In operation, a plurality of tokens or features are manually or automatically associated with each category. A weight is then coupled to each feature, wherein the weight indicates a degree of association between the feature and the category. Next, a document is parsed into a plurality of unique tokens with associated counts, wherein the counts are indicative of the number of times the feature appears in the document. A category score representative of a sum of products of each feature count in the document times the corresponding feature weight in the category for each document is then computed. Next, the category scores are sorted by perspective, and a document is classified into a particular category, provided the category score exceeds a predetermined threshold.

A method and program product to generate summaries for text documents. A user can also specify a query, topic, and terms that he / she is interested in. This method determines the importance of each sentence by using the linguistic salience of the word to the user profile, the similarity among the word, the query and topic provided by a user and the sum of scores of the sentence comprising the word. After computing the score for each word, this method computes the score for each sentence in the set of sentences according to the score of words composing it and the position of the sentence in a section and a paragraph.

Apparatuses, systems, methods, and computer program products are disclosed for an automated messaging response. A message parsing module may parse a textual message to determine whether the message includes a question. A question determination module may determine a question type for the question. A response presentation module may present a response interface that includes a plurality of selectable responses to the question.

A method is provided for automatically classifying text into categories. In operation, a plurality of tokens or features are manually or automatically associated with each category. A weight is then coupled to each feature, wherein the weight indicates a degree of association between the feature and the category. Next, a document is parsed into a plurality of unique tokens with associated counts, wherein the counts are indicative of the number of times the feature appears in the document. A category score representative of a sum of products of each feature count in the document times the corresponding feature weight in the category for each document is then computed. Next, the category scores are sorted by perspective, and a document is classified into a particular category, provided the category score exceeds a predetermined threshold.

A method for machine learning-based account manager virtual assistant message processing includes receiving a message of a user, preprocessing the message to generate a normalized data set, analyzing the normalized data using a classification machine learning model, analyzing the normalized data using a deep learning machine learning model, generating a response based on the output of the first and second machine learning models, and transmitting the response to the user. An account manager virtual assistant computing system includes a processor and a memory storing instructions that cause the account manager virtual assistant system to receive a message from a user, process the message using a first machine learning model, process the message using a second machine learning model, when the message is a request for quote, generate a response and transmit the response to the user.

A question answering system estimates an answer type of an answer from a input question statement, extracts keywords from the question statement and retrieves a document database, extracts language expressions (answer candidates) from the extracted document data, assigns evaluation points thereto. When there are a plurality of answer candidates having the same language expression, the system sorts those evaluation points in descending ranking order of evaluation, calculates values of evaluation points using such weighting that the value to be processed for each evaluation point diminishes, regards the sum total of those values as the evaluation point of the answer candidate, and outputs the answer candidate whose counted evaluation point is equal to or greater than a predetermined evaluation as an answer.

Computer-implemented systems and methods are provided for suggesting emoticons for insertion into text based on an analysis of sentiment in the text. An example method includes: determining a first sentiment of text in a text field; selecting first text from the text field in proximity to a current position of an input cursor in the text field; identifying one or more candidate emoticons wherein each candidate emoticon is associated with a respective score indicating relevance to the first text and the first sentiment based on, at least, historical user selections of emoticons for insertion in proximity to respective second text having a respective second sentiment; providing one or more candidate emoticons having respective highest scores for user selection; and receiving user selection of one or more of the provided emoticons and inserting the selected emoticons into the text field at the current position of the input cursor.

The invention relates to processing of data during semantic analysis of text data and building of document semantic models. The method consists of two main steps. In step one, an ontology is extracted from external information sources that contain descriptions of particular domain objects. In step two, the text information in the document is linked to ontology concepts and a document semantic model is built. Electronic resources that may or may not be connected through a hyperlink structure are used as information sources. The technical result is achieved, in particular, by identifying all terms in the document and linking them to ontology concepts, so that each term correlates to one concept (its meaning), and then ranking term meanings by importance to the document.

A chatbot system and method with interactive chat log. A chatbot includes a processor, an interactive dialog interface and said knowledge database. The chatbot generates a chat log page including messages based on a chat session. Without leaving the chat log page, the chatbot uses the chat log page for updating chatbot messages and corresponding entries in a knowledge database.

Processing natural language user inputs into a more formal, machine-readable, structured query representation used for making an item recommendation. Analyses of user inputs are coordinated via a knowledge graph constructed from categories, attributes, and attribute values describing relatively frequently occurring prior interactions of various users with an electronic marketplace. The knowledge graph has directed edges each with a score value based on: the conditional probabilities of category / attribute / attribute value interactions calculated from user behavioral patterns, associations between user queries and structured data based on historical buyer behavioral patterns in the marketplace, metadata from items made available for purchase by sellers used to better define buyers' requirements, and / or world knowledge of weather, locations / places, occasions, and item recipients that map to inventory-related data, for generating relevant prompts for further user input. The knowledge graph may be dynamically updated during a multi-turn interactive dialog.

Systems, apparatuses, and methods for automating avatar selection for groups are disclosed. A representative method includes providing a software application accessible to multiple users, and a database of available avatar characteristics that are selectively combinable to create user-specific avatars with which the users can represent themselves to each other in the application. User(s) associated with a group provide an instruction that associates a first set of the available avatar characteristics with the group. For users associated with the group, access is then restricted to the first set of avatar characteristics, and for users not associated with the group, access is restricted to a different second set of available avatar characteristics. Avatar characteristics in the first set may be selected by an automated procedure that analyzes user-specific avatars of some users, and selects which of the available avatar characteristics to include in the first set based on the analysis.

A method and a system for extracting information from a natural language text corpus based on a natural language query are disclosed. In the method the natural language text corpus is analyzed with respect to surface structure of word tokens and surface syntactic roles of constituents, and the analyzed natural language text corpus is then indexed and stored. Furthermore a natural language query is analyzed with respect to surface structure of word tokens and surface syntactic roles of constituents. From the analyzed natural language query one or more surface variants are then created, where these surface variants are equivalent to the natural language query with respect to lexical meaning of word tokens and surface syntactic roles of constituents. The surface variants are then compared with the indexed and stored analyzed natural language text corpus, and each portion of text comprising a string of word tokens that matches the any one of the surface variants or the natural language query is extracted from the indexed and stored analyzed natural language text corpus.

A method and apparatus are provided for adapting a language model. The method and apparatus provide supervised class-based adaptation of the language model utilizing in-domain semantic information.

Method and apparatus for reconstructing new documents from a group of old ones by removing the existing redundant information. Redundant information (images, text paragraphs) from retrieved multimedia documents is removed. Each document consists of two main parts stored in different databases. The first part of a document represents text paragraphs, the second part consists of the images and drawings related with the text paragraphs. An information reduction methodology examines first the text paragraphs of each document related with a specific topic, and removes the redundant information, such as same or similar paragraphs, by keeping pointers useful for a future reconstruction of the original documents. The remaining text paragraphs and the set of points are used to compose the first version of a new document. The invention also examines all the images related with the set of original documents and removes the same or similar images while keeping pointers that could assist a future reconstruction of the original documents. The invention merges text-paragraphs and images and creates the first stage new document.

Markov Chain Monte Carlo (MCMC) sampling of elements of a domain to be sampled is performed to generate a set of samples. The MCMC sampling is performed over a search tree of decision sequences representing the domain to be sampled and having terminal nodes corresponding to elements of the domain. In some embodiments the MCMC sampling is performed by Metropolis-Hastings (MH) sampling. The MCMC sampling is constrained using a bound on nodes of the search tree. The constraint may entail detecting a node whose bound value ensures that an acceptable element cannot be identified by continuing traversal of the tree past that node, and terminating the traversal in response. The constraint may entail selecting a node to serve as a starting node for a sampling attempt in accordance with a statistical promise distribution indicating likelihood that following a decision sequence rooted at the node will identify an acceptable element.

An artificial intelligence data analytics system, including: a data analytics algorithm accessible by an analyst operable for receiving data from one or more sources, applying one or more analytical tools to the data, and displaying summary information to the analyst; an artificial intelligence algorithm accessible by the analyst operable for applying a preferred artificial intelligence methodology to the data and automatically triggering and executing an action on behalf of the analyst; and an artificial intelligence learning algorithm accessible by the analyst operable for statistically scoring a plurality of potential artificial intelligence methodologies to be used by the artificial intelligence algorithm, receiving a corresponding analyst preference as to the preferred artificial intelligence methodology, and causing the artificial intelligence algorithm to utilize the preferred artificial intelligence methodology.

Systems and processes for operating a digital assistant are provided. In accordance with one or more examples, a method includes, receiving training data for a data-driven learning network. The training data include a plurality of word sequences. The method further includes obtaining representations of an initial set of semantic categories associated with the words included in the training data; and training the data-driven learning network based on the plurality of word sequences included in the training data and based on the representations of the initial set of semantic categories. The training is performed using the word sequences in their entirety. The method further includes obtaining, based on the trained data-driven learning network, representations of a set of semantic embeddings of the words included in the training data; and providing the representations of the set of semantic embeddings to at least one of a plurality of different natural language processing tasks.

A method and apparatus are provided for adapting a language model. The method and apparatus provide supervised class-based adaptation of the language model utilizing in-domain semantic information.

A method for authoring a grammar for use in a language processing application is provided. The method includes receiving at least one grammar configuration parameter relating to how to configure a grammar and creating the grammar based on the at least one grammar configuration parameter.

A machine translation quality determination mechanism uses comparisons of subsequent and potentially numerous reverse translations of a translated human language back to the source language. The process of translating from source language to target language to source language may iterate many times to ultimately yield information as to an assertion of low quality translation. Thus, the present invention continuously iterates this “back-and-forth” translation until the resulting source human language text is not reasonably equivalent to the original source human language or until the process iterates a predetermined number of times. If the back-and-forth translation results in a source human language text that is not reasonably equivalent to the original source text, then the translation is identified as low quality.