International Business Machines Corporation

Servicenow, Inc.

Joshua S Allen

Andrew Ronald Freed

Kyle G Christianson

Hassan Nadim

Andrew Jason Lavery

Shikha Srivastava

Richard Adam King

Vijay Pandiarajan

Corville O Allen

Eric Woods

Samar Choudhary

Mario Alessandro Maldari

Robert L Nielsen

Bradford Austin Fisher

Aaron T Smith

Jason M Brown

Thai T La

Ryan Brink

Bryan M Ellington

Jacob E Yackenovich

Paul L H Carroll

Fernando Jose Suarez Saiz

Sorabh Murgai

Kenneth James Parzygnat

Raghuveer Raavi

Kimberly S Holmes

A method retrains a cognitive interaction information system. One or more processors monitors an interactive communication between a user and the cognitive interactive information system. The processor(s) identifies an incorrect answer returned by the cognitive interactive information system to a question asked by the user; reevaluates the question utilizing alternative intent classifications to form alternative answers; and presents the alternative answers to the user. The processor(s) receive a selection from the alternative answers from the user; updates processing rules to generate updated processing rules that are based on the selection; and retrains the cognitive interaction information system based on the updated processing rules.

Retraining a conversation system based on negative feedback

Systems and methods for generating and annotating cell documents include extracting tables from a document using a table extraction engine. Headers are extracted for each of the tables using a header detection engine. Cells are extracted from each of the tables using a cell extraction engine. A cell document is generated for each of the cells which are each correlated to corresponding portions of the headers, each cell document recording the correlation between the cells and the headers. Each cell document is annotated to generate annotated cell documents with a cell recognition model trained to perform natural language processing on the cell documents by classifying each term in each of the cell documents and extracting relationships between the terms of each of the cell documents.

Information extraction from open-ended schema-less tables

A computer-implemented method for paging row-based data stored as an object file. The method includes receiving, from a client, a request for a set of data, wherein the set of data is included in a file, the file is stored as an object file, and the object file is stored in an object based storage system and the file includes row based data for a plurality of rows. The method also includes, generating, by a micro-service, an index of the file, wherein the index file includes a byte to row translation of a portion of the file. The method includes, storing the index in the object based storage system. The method further includes, retrieving, from the index, the set of data. The method includes, returning, to the client, the set of data.

Paging row-based data stored as objects

A method, system and computer-usable medium for detecting headers in various documents, such as PDF and HTML files. The files are converted to a two dimensional array or table, having orthogonal rows and columns. Either rows or columns are determined to include headers. For determining if rows include headers. For each row in the array or table, pair wise comparison is performed for each cell of each column that is orthogonal to that row. The pair wise comparison scores or values are summed up for each orthogonal column to that row and the sum across for all the orthogonal columns to row provide a score or value for that row. Row scores are evaluated relative to one another to determine likelihood of headers in the row. For determining if columns have headers, similar calculation is performed between columns and their orthogonal rows.

Table header detection using global machine learning features from orthogonal rows and columns

A method, system and computer-usable medium are disclosed for finding vertically and horizontally aligned cells in a complex table structure. A file or document, such as an HTML file that defines a complex table includes spanning rows and columns is expanded into a two dimensional (2D) array or table with orthogonal rows and columns, where the spanning rows and columns included cells with copied values or object references. The expanded 2D array or table can be deduplicated row or column wise to determine header alignment of the table.

Intelligent normalization and de-normalization of tables for multiple processing scenarios

A method, computer system, and a computer program product for detecting one or more semantic headers in one or more tabular structures by utilizing a custom pre-trained embeddings model is provided. The present invention may include receiving the custom pre-trained embeddings model. The present invention may also include computing one or more dot product values associated with the one or more tabular structures from the one or more documents based on the context of each cell associated with the one or more tabular structures in the one or more documents. The present invention may then include generating one or more similarity feature values based on the computed one or more dot product values. The present invention may further include detecting the one or more semantic headers associated with the one or more tabular structures based on the one or more similarity feature values.

Semantic header detection using pre-trained embeddings

Embodiments provide a system and method for medical record error detection. Using a cognitive system having natural language processing capabilities, the error detection system can analyze a corpus of medical records, which can correspond to a particular patient, as well as a new medical record, and parse each into one or more record elements. By comparing the record elements, the error detection system can detect one or more potential errors resulting from a record element in the new medical record being inconsistent or contradictory with the record elements in the prior medical records, or ambiguous. The error detection system can output a list of candidate answers for correcting the one or more potential errors based upon a cognitive analysis of the corpus of medical records. The error detection system can further output a list of suggested therapies based on the selection of one or more of the candidate answers.

Medical record error detection system and method

A method, system and computer-usable medium are disclosed for associating data cells with headers and header labels. In certain embodiments, a table having rows and columns is received, wherein the table includes a plurality of cells, wherein each cell is populated with at least one of a header name, data value, or no information. A determination is made as to whether a cell is a header cell or data cell. If the cell is a header cell, current list of column and current list of row headers are dynamically updated. The current list of column and row headers are assigned to the cell regardless of whether the cell is a header cell or data cell. Headers associated with header cells are used to identify label candidates for the header name of the header cell. The labels may be used to provide additional context for headers within a data cell.

System for extracting header labels for header cells in tables having complex header structures

A method improves a functionality of a conversational agent that is generated by an artificial intelligence (AI) system. A conversational agent receives a first utterance from a first entity. The AI system identifies an unverified response to the first utterance; sends the unverified response to the first entity; and receives a positive feedback indication about the unverified response from the first entity. The AI system searches a data store in order to identify an entry for a second utterance by a second entity, where the second entity has sent a positive feedback for the unverified response. The AI system sends the second utterance and the unverified response to the first entity, and receives a positive feedback for the unverified response to the second utterance from the first entity in order to mark the unverified response as a verified response, which responds to future receipts of the first utterance.

Training artificial intelligence to respond to user utterances

A method, system and computer-usable medium are disclosed for associating data cells with headers and tables having one or more embedded header structures. In certain embodiments, a table having rows and columns is received, wherein the table includes a plurality of cells, wherein each cell is populated with at least one of a header name, data value, or no information, the table having at least one embedded header. A determination is made as to whether a cell is a header cell or data cell. If the cell is a header cell, a count of consecutive column headers is maintained. A current list of column headers is dynamically updated based on the count of the consecutive column headers. Upon encountering a data cell, the current list of column headers is assigned to the data cell.

System for associating data cells with headers in tables having complex header structures

Systems and methods for generating and annotating cell documents include extracting tables from a document using a table extraction engine. Headers are extracted for each of the tables using a header detection engine. Cells are extracted from each of the tables using a cell extraction engine. A cell document is generated for each of the cells which are each correlated to corresponding portions of the headers, each cell document recording the correlation between the cells and the the headers. Each cell document is annotated to generate annotated cell documents with a cell recognition model trained to perform natural language processing on the cell documents by classifying each term in each of the cell documents and extracting relationships between the terms of each of the cell documents.

A device comprises an input component configured to receive an event identification request from a provider. The event identification request comprises one or more requirements usable to identify an actionable event for resolution. The requirements specify a geographical boundary and one or more goods. The device also comprises a processor communicably coupled to the input component. The processor is adapted to transmit the event identification request to a repository configured to match the event identification request with one or more actionable events based on the one or more requirements.

Matching actionable events with goods and services providers

A method comprises receiving event data describing an event from an internet-coupled mobile communication device and receiving a geographical location of the event. The geographical location is identified by the internet-coupled mobile communication device. The method also comprises receiving an event identification request from a provider device, and determining that the event data and the geographical information match the event identification request. The method further comprises transmitting the event data and the geographical information to the provider device.

An approach is provided for evaluating a potential post based on historical data. In the approach, historically highly rated attributes that to previously received highly rated posts are identified. The process receives a potential post from an online Subject Matter Expert (SME). The process analyzes the potential post, using a Natural Language Processing (NLP) routine performed by computer processors. The analysis identifies a lack of one or more of the historically highly rated attributes in the potential post. The process then notifies the SME, based on the analysis, regarding the lack of one or more of the historically highly rated attributes in the potential post.

Growing community of inventors

Durham, NC, United States of America

Joshua S Allen