International Business Machines Corporation

Adobe, Inc.

Nandakishore Kambhatla

Wlodek Wlodzimierz Zadrozny

Karthik Visweswariah

Debapriyo Majumdar

Rose Catherine Kanjirathinkal

Shajith I Mohamed

Sugata Ghosal

Abraham P Ittycheriah

Rashmi Gangadharaiah

Salim Estephan Roukos

Anup Kumar Ghosh

Catherine Gody Wolf

Xiaoqiang Luo

Asidhara Lahiri

Hongyan Jing

Dimitri Kanevsky

Soumitra Sarkar

Rafah A Hosn

Niyati Himanshu Chhaya

Munish Goyal

Gyana Ranjan Parija

Maroun Touma

Anshul Sheopuri

Priyanka Agrawal

Yiming Ye

Tetsunosuke Fujisaki

Vijil Enara Chenthamarakshan

Dragomir Radkov Radev

Aparna Garimella

Imed Zitouni

Vinay Aggarwal

Amit Kumar R Singh

Soujanya Soni

Saikat Mukhopadhyay

Priyanka Dasgupta

Aditya Telang

Sambuddha Roy

Rohit Manohar Lotlikar

Milind V Vaidya

Nickle Lamoreaux

Joyce Yue Chai

Sunil Subramanyam Govindappa

Binny Rieder

Joanna Daly

Amit K R Singh

Ananya Ganesh

Pooja Arya

Pavithra Krishnan

Shivaram Kulkarni

Simon Thomas

Systems and methods for text simplification are described. Embodiments of the present disclosure identify a simplified text that includes original information from a complex text and additional information that is not in the complex text. Embodiments then compute an entailment score for each sentence of the simplified text using a neural network, wherein the entailment score indicates whether the sentence of the simplified text includes information from a sentence of the complex text corresponding to the sentence of the simplified text. Then, embodiments generate a modified text based on the entailment score, the simplified text, and the complex text, wherein the modified text includes the original information and excludes the additional information. Embodiments may then present the modified text to a user via a user interface.

Text simplification with minimal hallucination

Various embodiments are provided for implementing intelligent decision support system in a computing environment by a processor. Data of historical decisions may be collected and examples of decisions by domain experts may be generated. One or more machine learning models may be generated using different splits of the historical data and the annotated data. The one or more machine learning models may be combined and used to generate ensemble machine learning models that generate recommendations for the decisions. Users interact with a user interface displaying the data, recommendations, reasons for recommendations and a conversational dialog system for querying about the data, recommendations and guidance for decision making.

Intelligent decision support system

A method for detecting solutions to a problem using content in online discussion sources. The method includes receiving a request, such request identifying a problem, and searching multiple online discussion sources for content related to the problem. Responsive to finding content related to the problem, the method searches the multiple online discussion sources for a plurality of solutions to the problem. Responsive to finding a plurality of solutions to the problem, the method forms groups containing the solutions from each of the multiple online discussion sources. The method then determines a likeliness to solve the problem for each of the groups and ranks the groups based on the determined likeliness to solve the problem. The method then determines that the rank of at least one group meets a threshold value, wherein the threshold value is based on a confidence in the likeliness to solve the problem.

Using content found in online discussion sources to detect problems and corresponding solutions

Embodiments of the present invention provide a system, method, and program product to measure problems from a social media discussion. In exemplary embodiments, a computer extracts one or more problems from the social media discussion. The computer extracts one or more severity indicators and one or more complexity indicators from the social media discussion. The computer clusters the one or more problems into one or more sets of unique problems in a manner that related problems are clustered together into the one or more unique problems. The computer determines an overall severity and an overall complexity of the sets of unique problems.

Measuring problems from social media discussions

Embodiments of the present invention relate to an approach for reusing information/knowledge. Specifically, embodiments of the present invention provide an approach for retrieving previously stored data to satisfy queries (e.g., jobs/tickets) for solutions to problems while maintaining privacy/security of the data as well as ensuring the quality of the results. In a typical embodiment, a query for a solution to a problem is received and details are extracted therefrom. Using the details, a search is performed on a set of data stored in at least one computer storage device. Based on the search, a set of results will be generated and classified into a set of categories. In any event, the quality of each of the set of results will be assessed based on the usefulness of the set of results.

Computer-implemented information reuse

A decision support tool which takes a high-level view of supply and demand in the realm of labor resources and produces globally-optimal recommendations for assignment of practitioners to open seats. These assignments preferably optimize a utility function composed of different business metrics and can be customizable for different organizations and priorities.

Decision support system and method for distributed decision making for optimal human resource deployment

A Bell Tree data structure is provided to model the process of chaining the mentions, from one or more documents, into entities, tracking the entire process; where the data structure is used in an entity tracking process that produces multiple results ranked by a product of probability scores.

Mention-synchronous entity tracking: system and method for chaining mentions

Systems and methods are provided for automated semantic role labeling for languages having complex morphology. In one aspect, a method for processing natural language text includes receiving as input a natural language text sentence comprising a sequence of white-space delimited words including inflicted words that are formed of morphemes including a stem and one or more affixes, identifying a target verb as a stem of an inflicted word in the text sentence, grouping morphemes from one or more inflicted words with the same syntactic role into constituents, and predicting a semantic role of a constituent for the target verb.

Systems and methods for automatic semantic role labeling of high morphological text for natural language processing applications

A method, a system and a computer program product for analyzing a document are disclosed. In response to receiving the document, the document is partitioned into a plurality of segments using a set of pre-defined attributes. The plurality of segments of the document is mapped with corresponding segments of at least one template selected from a set of stored templates. A first template from the set of stored templates is selected and a group of segments in the first template is identified by computing at least one of a structural similarity and a textual similarity associated with the group of segments compared with the plurality of segments of the document. A subset of segments from the group of segments is aligned with corresponding segments from the plurality of segments of the document. A set of scores is computed using a set of pre-defined criteria, in response to the mapping. The document is analyzed based on the computed set of scores.

Analyzing documents using stored templates

Mention-synchronous entity tracking system and method for chaining mentions

This patent describes a novel system, method, and program product that are used in interactive natural language dialog. One or more presentation managers operating on a computer system present information from the computer system to one or more users over network interface(s) and accept queries from the users using one or more known input/output modalities (e.g. Speech, typed in text, pointing devices, etc.). A natural language parser parses one or more natural language phrases received over one or more of the network interfaces by one or more of the presentation managers into one or more logical forms (parsed user input), each logical form having a grammatical and structural organization. A dialog manager module maintains and directs interactive sessions between each of the users and the computer system. The dialog manager receives logical forms from one or more of the presentation managers and sends these to a taxonomical mapping process which matches the items of interest to the user against the content organization in the content database to match business categories and sends modified logical forms back to the dialog manager.

System, method and program product for interactive natural dialog

The invention is directed towards an automated system for extracting voice messages from a voice mail system and for providing unified access to voice mail and electronic mail or voice mail and the internet. For a given user, a voice mail remote access server connects to the user's voice mail system through a telephone or data network, and uses speech recognition and understanding to navigate through the prompts of the voice mail system and extract the user's voice mail. Depending upon the access mechanism preferred by the user, the voice messages are sent as e-mail messages with attachments (audio files) to the user or made accessible to the user's world wide web server or displayed to the user using a stand-alone voice mail player application.

System and method for accessing voice mail from a remote server

Disclosed is a system and a method for combining the computational resources of numerous embedded devices to enable any of them to perform complex tasks like speech recognition or natural language understanding. A distinguished master device communicates with a network of embedded devices, and organizes them as the nodes of a neural network. To each node (embedded device) in the neural network, the master device sends the activation function for that node and the connectivity pattern for that node. The master device sends the inputs for the network to the distinguished input nodes of the network. During computation, each node computes the activation function of all of its inputs and sends its activation to all the nodes to which it needs to send output to. The outputs of the neural network are sent to the master device. Thus, the network of embedded devices can perform any computation (like speech recognition, natural language understanding, etc.) which can be mapped onto a neural network model.

Method and apparatus for executing neural network applications on a network of embedded devices

An automated system generates and revises grammars for speech recognizers in a speech recognition system. Given an initial grammar, expressed in terms of non-terminals in Backus-Naur Form (BNF) notation, a sentence generator generates a list of all sentences accepted by the grammar. From this list, a corpus of inappropriate or irrelevant sentences which are accepted by the grammar (counter-examples) is identified. A grammar revisor program uses the original grammar and the list of counter examples, to generate a pruned list from which a revised grammar is generated. The revision process is iterated several times either concatenating or merging pairs of non-terminals until the revised grammar is deemed satisfactory in that it accepts as legal only relevant sentences. The revised grammar is used by the speech recognizer, thus reducing errors in the overall system.

Method and apparatus for creating speech recognition grammars

Growing community of inventors

Bengaluru, India

Nandakishore Kambhatla