Abstract: An intelligent text cleaning method includes: acquiring a text set, and preprocessing the text set to obtain a word vector text set; subjecting the word vector text set to a full-text matrix numeralization to generate a principal word vector matrix and a text word vector matrix; inputting the principal word vector matrix to a BiLSTM model to generate an intermediate text vector; inputting the text word vector matrix to a convolution neural network model to generate a target text vector; and concatenating the intermediate text vector and the target text vector to obtain combined text vectors, inputting the combined text vectors to a pre-constructed semantic recognition classifier model, outputting an aggregated text vector, subjecting the aggregated text vector to reverse recovery using a word2vec reverse algorithm, and outputting a standard text. The present application realizes accurate text cleaning.

Abstract: The present invention may be a method, a system, and/or a computer program product. An embodiment of the present invention provides a method for paraphrasing, on a client computer, text in a webpage, the method comprising the following: transferring a request for a webpage including a plurality of passages of text to a server; receiving the webpage from the server in response to the request; judging whether or not the received webpage has text which is a subject of paraphrase; in a case where the judgment is positive, paraphrasing the text; and displaying, on a display, the webpage including the paraphrased text. Another embodiment of the present invention provides a method for updating on a server, text in a webpage, the method comprising the following: receiving, from each of the devices, a set of URLs of a webpage, a location path of text which is a subject of paraphrase in the webpage, and paraphrased text; and replacing text in the webpage with text among the received text.

Abstract: A method and system for detecting anomalies in mission-critical environments using word representation learning are provided. The method includes parsing at least one received data set into a text structure; isolating a protocol language of the at least one received data set, wherein the protocol language is a standardized pattern for communication over at least one communication protocol; generating at least one document from the contents of the received at least one data set, wherein the at least one document includes at least one parsed text structure referencing a unique identifier; detecting insights in the at least one generated document, wherein insights are detected in at least one representation having at least one dimension, wherein the representation is mapped to at least one learned hyperspace; extracting rules from the detected insights; and detecting anomalies by applying the extracted rules on patterns for communication over at least one communication protocol.

Abstract: A computer-implemented method, computer system, and computer program product for for generation of a chatbot. The method may include receiving data in a first format. The method may include generating one or more clusters from the received data. The method may include labeling the generated one or more clusters. The method may include exporting the one or more labeled clusters into a cluster database. The method may include generating the chatbot using the one or more labeled clusters exported from the cluster database. The method may include executing a validation script into the chatbot to generate a report. The method may include receiving a determination on an accuracy of the chatbot based on the generated report. In response to determining that the chatbot is not accurate, the method may include determining whether a manual adjustment directly on the chatbot is needed.

Abstract: In an illustrative embodiment, methods and systems for automatically labeling unstructured data include accessing unstructured data representing data entry and analyzing the unstructured data by applying natural language processing to a text component of the unstructured data to obtain a set of term counts of words and/or phrases identified in the text component. Analyzing may include applying at least one clustering algorithm to the set of term counts to determine a term cluster, identifying a preexisting term cluster most closely matching the term cluster, and applying, to the unstructured data, a predefined label corresponding to the preexisting term cluster. The unstructured data may be analyzed to obtain formatting counts of formatting elements, and a formatting cluster may be determined and applied to match to a preexisting formatting cluster, thus deriving a predefined label corresponding to the preexisting formatting cluster.

Abstract: Developer and runtime environments supporting multi-modal input for computing systems are disclosed. The developer environment includes a gesture library of human body gestures (e.g., hand gestures) that a previously-trained, system-level gesture recognition machine is configured to recognize. The developer environment further includes a user interface for linking a gesture of the gesture library with a semantic descriptor that is assigned to a function of the application program. The application program is executable to implement the function responsive to receiving an indication of the gesture recognized by the gesture recognition machine within image data captured by a camera. The semantic descriptor may be additionally linked to a different input modality than the gesture, such as a natural language input.

Abstract: Machine learning, artificial intelligence, and other computer-implemented methods are used to identify various semantically important chunks in documents, automatically label them with appropriate datatypes and semantic roles, and use this enhanced information to assist authors and to support downstream processes. Chunk locations, datatypes, and semantic roles can often be automatically determined from what is here called “context”, to wit, the combination of their formatting, structure, and content; those of adjacent or nearby content; overall patterns of occurrence in a document, and similarities of all these things across documents (mainly but not exclusively among documents in the same document set).

Abstract: A method, an apparatus, a computer device and a readable medium for knowledge hierarchical extraction of a text are disclosed. The method comprises: performing word segmentation on a designated text to obtain a word list, the word list including at least one word arranged in a sequence in the designated text; analyzing part-of-speech of each word in the word list in the designated text, to obtain a part-of-speech list corresponding to the word list; predicting a SPO triple included in the designated text according to the word list, the part-of-speech list and a pre-trained knowledge hierarchical extraction model. By the technical solutions, the SPO triple included in any designated text however loose its organization and structure is may be accurately extracted based on the pre-trained knowledge hierarchical extraction model. Compared to the prior art, the efficiency and accuracy of knowledge hierarchical extraction may be effectively improved.

Abstract: A system maintains a knowledge layout to support the analysis of active events and determination of epicenter and aftershock nodes via an event reach stack. At an input layer of the event reach stack, the system may receive active event data. At a semantic layer, the system may parse the active event data to determine event phrases. Based on the event phrases, the system may identify epicenter nodes directly affected by the active event. At an analytic model layer, the system may successively determine aftershock nodes by traversing the knowledge layout outward from the epicenter nodes. The system then directs the response to the active event to the aftershock and epicenter nodes, via action at a focus response layer of the event reach stack.

Abstract: Techniques are provided for using machine learning techniques to analyze textual content. In one technique, a potential item is identified within a document. An analysis of the potential item is performed at multiple levels of granularity that includes two or more of a sentence level, a segment level, or a document level. The analysis produces multiple outputs, one for each level of granularity in the multiple levels of granularity. The outputs are input into a machine-learned model to generate a score for the potential item. Based on the score, the potential item is presented on a computing device. In response to user selection of the potential item, an association between the potential item and the document is created. The association may be used later to identify a set of users to which the document (or data thereof) is to be presented.

Abstract: In some embodiments, a style transfer computing system generates a set of discriminator models corresponding to a set of styles based on a set of training datasets labeled for respective styles. The style transfer computing system further generates a style transfer language model for a target style combination that includes multiple target styles from the set of styles. The style transfer language model includes a cascaded language model and multiple discriminator models selected from the set of discriminator models. The style transfer computing system trains the style transfer language model to minimize a loss function containing a loss term for the cascaded language model and multiple loss terms for the multiple discriminator models. For a source sentence and a given target style combination, the style transfer computing system applies the style transfer language model on the source sentence to generate a target sentence in the given target style combination.

Abstract: Reading material on video gives the reader a seamless reading experience by displaying on a device of their choice a series of segments containing letters, words, phrases, sentences and/or paragraphs on a background of the drafter's choice. One segment flows into the other until the reading material is completed. These sequential segments are set to be viewed seamlessly with audio accompaniment. Words, sentences or paragraphs are set to music, where recognizable features of the music are played at the appearance of a certain word or the beginning of a sentence or paragraph. The appearance of a word, sentence or paragraph may be accompanied by the appearance of an image representing the word, sentence or paragraph, along with a recognizable designated musical element.

Abstract: Embodiments of the present disclosure provide a method and an apparatus for automatically discovering medical knowledge. In this method, one or more linking concepts having a semantic relation with a starting concept are obtained from a medical literature library. The starting concept represents a disease. Next, one or more target concepts having a semantic relation with the one or more linking concepts are obtained from the medical literature library, and an association degree of each of the one or more target concepts with respect to the starting concept is calculated. The association degree indicates a probability that the target concept can cope with the starting concept. Further, the one or more target concepts are sorted according to the calculated association degrees. In this method, explainable target concepts can be obtained by using semantic analysis, and these target concepts are sorted to increase a possibility of discovering useful medical knowledge.

Abstract: A bipartite application implements a table auto-completion (TAC) algorithm on the client side and the server side. A client module runs a local model of the TAC algorithm on a user device and a server module runs a global model of the TAC algorithm on a server machine. The local model is continuously adapted through on-the-fly training, with as few as a negative example, to perform TAC on the client side, one document at a time. Knowledge thus learned by the local model is used to improve the global model on the server side. The global model can be utilized to automatically and intelligently extract table information from a large number of documents with significantly improved accuracy, requiring minimal human intervention even on complex tables.

Abstract: A technique for more efficiently integrating external content for web page display is disclosed. The technique includes identifying a plurality of page features for displaying a web page, selecting a corresponding content resolver from a plurality of content resolvers stored in the memory for each of the identified page features. The selecting may be in accordance with the end user device. The technique may also include rendering the web page for display on the end user device in accordance with one or more parameters specified in the corresponding content resolver.

Abstract: Methods and apparatus for converting an OOXML file into a format which can be rendered, displayed and edited by a browser on a device such as a cell phone without the device having to directly work with the document in the OOXML format are described. In various embodiments the contents of an OOXML file are converted into a JSON object, communicated to a client device which can display and edit the document contents. A revised version of the JSON object including edits is returned and converted back into an OOXML file. The method allows for preserving document format and other information while avoiding the need for a client device to support display and direct editing of a document in OOXML format.

Abstract: Disclosed are a method, a device, a system and/or a manufacture of iterative development and/or scalable deployment of a spreadsheet-based formula algorithm. In one embodiment, a system for scalable application of a data model defined in a spreadsheet to a dataset includes a translation server and an execution server. The translation server receives a spreadsheet file including a formula algorithm. The formula algorithm includes one or more spreadsheet formulas stored in cells. The translation server generates an extrapolated algorithm expressed in a programming language based on the formula algorithm. The execution server receives the extrapolated algorithm and the dataset and verifies calculation independence when applied to a data entry. The extrapolated algorithm is applied against the dataset. An iteration engine may continuously reapply the extrapolated algorithm to update an output data as the dataset evolves and/or receive an update to the formula algorithm and reapply the extrapolated algorithm.

Abstract: In medical writing, manual process of creating, updating and maintaining documents is expensive, time consuming. This disclosure provides a method of an automatic medical document writing by receiving, a plurality of input documents as an input; processing, the inputted plurality of documents by extracting into an at least one section to generate a list of sections; classifying, at least one category corresponding to the at least one section to generate a summary set; generating, at least one of a local context and a global context based on the summary set; parsing, at least one sentence based on the generated local context and global context to generate at least one sequence of the plurality of sentences; processing, the at least one sequence of the plurality of sentence to generate a queue of the at least one sequence; validating, the queue of the at least one sequence to obtain a combined summary set.

Abstract: A method is provided for validating a hypothesis sentence. The method extracts, from a document database D using a hypothesis sentence that includes a causal part and a consequence part, a set D1 of documents related to the causal part. The method extracts, from set D1, a set S of sentences that include expressions of opinion. The method obtains a word list W of words that have a high co-occurrence in the set D1. The method selects, from the set S, a set S1 of sentences that are positionally close to any of the words in the word list W. The method selects, from the set S, a set S2 of sentences that are related to the words in the consequence part of the hypothesis. The method extracts and displays sentences included in both the set S1 and the set S2 as opinion sentences in relation to the hypothesis sentence.

Abstract: A method, system, program and apparatus for sharing content portions is disclosed. An ordinary user, using a standard browser is empowered to be able to select portions of content within a currently displayed internet document. The user is presented with the ability to share the content portions selected by him specifically, and share them with his social graph and/or publish them to other locations. All this is accomplished without requiring additional software components be installed by the user.