Abstract: Provided is a method, system, and computer program product for dynamically correcting text of a communication sent by a user. A processor may receive a first communication sent by a first user. The processor may forward the first communication to one or more other users. The processor may receive a second communication sent by the first user. The second communication includes a command to correct an error in the first communication. The processor may determine a location of the error in the first communication. The processor may correct the error in the first communication. The processor may forward a corrected version of the first communication to the one or more other users.

Abstract: A computer-implemented method for generating regulatory content requirement descriptions is disclosed and involves receiving requirement data including a plurality of requirements including hierarchical information extracted from regulatory content. The method involves identifying parent requirements based on the existence of child requirements on a lower hierarchical level and generating requirement pairs including the parent requirement and at least one child requirement. The method also involves feeding each of the pairs through a conjunction classifier which has been trained to generate a classification output indicative of the pair being not a conjunction (NC), a single requirement conjunction (CSR), or a multiple requirement conjunction (CMR). The method involves generating a set of requirement descriptions based on the classification output generated for each parent requirement.

Abstract: Embodiments of the present disclosure disclose a text translation method, a text translation apparatus, a device and a storage medium. The method includes: obtaining a source language text; and translating the source language text with a modified translation model to obtain a target language text corresponding to the source language text, the modified translation model being obtained by modifying an original translation model based on a text evaluation result of one or more translated texts for training, the translated text for training being an output result after translating through the original translation model, and the text evaluation result for evaluating a contextual semantic relation in the translated text for training.

Abstract: An example system includes a processor to receive data for a multi-objective task. The processor is to also perform the multi-objective task on the received data via a trained primal network. The primal network and a dual network are trained for a multi-objective task using a Lagrangian loss function representing a number of objectives. The primal network is trained to minimize the Lagrangian loss function and the dual network is trained to maximize the Lagrangian loss function.

Abstract: The present disclosure discloses a method and apparatus for recognizing a text type, and a device. A specific embodiment of the method comprises: preprocessing a pre-acquired text to obtain a keyword set of the text; calculating an occurrence probability value of each keyword in the keyword set in the text; importing, for the each keyword in the keyword set, the keyword and the occurrence probability value corresponding to the keyword into a pre-established document topic generation model, and determining an occurrence probability value of each topic in the text preset in the document topic generation model; and recognizing a type to which the text belongs based on the occurrence probability value of the each topic in the text. The embodiment improves an accuracy of recognizing the type of the text.

Abstract: A computer implemented method of document classification includes receiving a text document. A first classification is generated for the document, and a text corpus is searched for one or more terms from the document. Searched terms having an incidence in the text corpus lower than a threshold incidence are flagged, and at least one classification is generated after removing at least one flagged term from the document. An output is generated if the further classification is different from the first classification.

Abstract: A computer generates a plurality of encoded sentences in the first language by encoding a plurality of sentences in the first language in prescribed units. Next, the computer generates a plurality of encoded sentences in the second language by encoding a plurality of sentences in the second language, each of which is associated with each of the plurality of sentences in the first language, in the prescribed units. The computer generates alignment information based on a code included in each of the plurality of encoded sentences in the first language and a code included in an encoded sentence in the second language, which is associated with each of the plurality of encoded sentences in the first language. The alignment information indicates an alignment between a plurality of codes in the first language and a plurality of codes in the second language.

Abstract: A recurrent neural network (RNN) method implemented on a computer system is used to produce summaries of unstructured text generated by multiple networks of individuals interacting over time by encoding the unstructured text into intermediate representations and decoding the intermediate representations into summaries of each network. Parameter data for the RNN is obtained by using multiple different versions of the same source texts to train the computer system. The method and computer system can be used to identify which of the networks match a query by determining which network generates the query with low or lowest cost.

Abstract: Advanced reporting systems based on ontology development for quantitative imaging biomarkers, information modeling to integrate heterogeneous data types relevant to quantitative medical analysis, and a knowledge representation framework for their statistical validation.

Abstract: Systems and methods for translating a source segment are disclosed. In embodiments, a computer-implemented method for translating a source segment comprises receiving, by a computing device, the source segment in a first language to be translated into a second language; identifying, by the computing device, linguistic markers within the source segment and associated noise values to produce a tagged source segment, wherein the linguistic markers are associated with one or more linguistic patterns likely to introduce noise into a translation channel; transforming, by the computing device, the tagged source segment into an amplified source segment; and sending, by the computing device, the amplified source segment to a machine translation module, wherein the machine translation module is configured to process the amplified source segment to produce a return amplified match in the second language.

Abstract: Methods, systems, and computer-readable storage media for contextual interpretation of a Japanese word are provided. A first set of characters representing Japanese words is received. The first set of characters are received is input to a neural network. The neural network is trained to processes characters based on bi-directional context interpretation. The first set of characters is processed by the neural network through a plurality of learning layers that process the first set of characters in an order of the first set of characters and in a reverse order to determine semantical meanings of the characters in the first set of characters. An alphabet representation of at least one character of the first set of characters representing a Japanese word is output. The alphabet representation corresponds to a semantical meaning of the at least one character within the first set of characters.

Abstract: An analytics platform for the extraction of structured observations from largely narrative sources using a hybrid approach of user configuration and machine learning is provided. The analytics platform collects and normalizes data from public and private sources and applies extractions to the data to create a world view of objects, traits, and relationships of interest and maintains that world view as data and/or extractions are updated. The platform is further configured to apply queries to the extracted world view for a variety of purposes including scoring objects for prioritized attention, generating notifications when specific conditions are met, providing data sets for exploratory analysis, and triggering the automatic collection of enhancing data from external sources.

Abstract: Present embodiment include a prosody subsystem of a natural language understanding (NLU) framework that is designed to analyze collections of written messages for various prosodic cues to break down the collection into a suitable level of granularity (e.g., into episodes, sessions, segments, utterances, and/or intent segments) for consumption by other components of the NLU framework, enabling operation of the NLU framework. These prosodic cues may include, for example, source prosodic cues that are based on the author and the conversation channel associated with each message, temporal prosodic cues that are based on a respective time associated with each message, and/or written prosodic cues that are based on the content of each message. For example, to improve the domain specificity of the agent automation system, intent segments extracted by the prosody subsystem may be consumed by a training process for a ML-based structure subsystem of the NLU framework.

Abstract: A method and apparatus include receiving a first sentence including a first set of words, and a second sentence including a second set of words. A first set of vectors corresponding to the first set of words of the first sentence, and a second set of vectors corresponding to the second set of words of the second sentence are generated using a word embedding model. A similarity matrix based on the first set of vectors and the second set of vectors is generated. An alignment score associated with the first set of vectors and the second set of vectors is determined using the similarity matrix. The alignment score is transmitted to permit information retrieval based on a similarity between the first sentence and the second sentence.

Abstract: Provided is a method for using an intelligent industrial assistant. The method may include receiving package data comprising at least one dialogue template. A natural language input may be received from a user. At least a portion of the natural language input may be mapped to a first dialogue template of the at least one dialogue template. At least one command may be determined based on the first dialogue template. Command data associated with the at least one command may be communicated. A system and computer program product are also disclosed.

Abstract: A natural language processing system that includes a sentence selector and a question answering module. The sentence selector receives a question and sentences that are associated with a context. For a question and each sentence, the sentence selector determines a score. A score represents whether the question is answerable with the sentence. Sentence selector then generates a minimum set of sentences from the scores associated with the question and sentences. The question answering module generates an answer for the question from the minimum set of sentences.

Abstract: A method, system and a computer program product are provided for generating a natural language model that is substantially independent of languages and domains by transforming monolingual embeddings into a multilingual embeddings in a first shared embedding space using a cross-lingual learning process, and then transforming the multilingual embeddings into cross-domain, multilingual embeddings in a second shared embedding space using a cross-domain learning process, where the multilingual embeddings and/or cross-domain, multilingual embeddings are evaluated to measure a degree to which the embeddings associate a set of target concepts with a set of attribute words.

Abstract: A technique for translating text strings includes receiving a source language text string from an application, determining that a translated text string that includes a translation in a target language of the source language text string is not available for use by the application, transmitting the source language text string to a translation service for translation, receiving the translated text string from the translation service, and causing the translated text string to be available for use by the application.

Abstract: Techniques are described herein for learning property graph representations edge-by-edge. In an embodiment, an input graph is received. The input graph comprises a plurality of vertices and a plurality of edges. Each vertex of the plurality of vertices is associated with vertex properties of the respective vertex. A vertex-to-property mapping is generated for each vertex of the plurality of vertices. The mapping maps each vertex to a vertex-property signature of a plurality of vertex-property signatures. A plurality of edge words is generated. Each edge word corresponds to one or more edges that each begin at a first vertex having a particular vertex-property signature of the plurality of vertex property signatures and end at a second vertex having a particular vertex-property signature of the plurality of vertex property signatures. A plurality of sentences is generated. Each sentence comprises edge words directly connected along a path of a plurality of paths in the input graph.

Abstract: Systems, methods, and computer-readable storage media for providing for intelligent switching of languages and/or pronunciations in a text-to-speech system. As the system receives text, the text is analyzed to identify portions which should have speech constructed using a pronunciation distinct from the remaining portions of the text. The text-to-speech system uses multiple pronunciation dictionaries to generate and produce speech corresponding to the text, where the identified portions of the text are in a different language or have a different accent from the remainder of the text. Having generated speech corresponding to the text in multiple languages, accents, or dialects, the system combines the portions, then communicates the speech to the text recipient.