Capital One Services, LLC

Paul Cho

Vannia Gonzalez Macias

Scott Garcia

Adithya Ramanathan

Rahul Gupta

Gary B Williams

Piper Alexandra Coble

Lin Ni Lisa Cheng

Ryan M Parker

Chris Demchalk

Ankur Ankur

Matthew Kevin Sullivan

Prarthana Bhattarai

Ralph Lozano

Natalie Kuhn

Brian Nguyen

Sergey Petrunin

Alexander Bussan

Eric Campbell

Mukund Manikantan

Talha Koc

Talyne Derderian Walsh

Exemplary embodiments provide methods, mediums, and systems for performing a reusable, intelligent semantic search across a potentially large number of records. Embodiments may be particularly useful for responding to requests for information from regulatory agencies. In one embodiment, an embedding model is trained to embed queries in an embedding space. When a new query is received, the new query is embedded with the embedding model. A set of documents (e.g., previous responses to regulatory inquiries) may be searched using the embedded query and an indexing model that allows for efficient searches of embedding spaces. A number of results may be returned from the document store, and the results may be ranked by a ranking model. User feedback about the quality of the results may be received, and the ranking model may be retrained based on the feedback.

Methods, mediums, and systems for reusable intelligent search workflows

Methods and systems are disclosed herein for using anomaly detection in timeseries data of user sentiment to detect incidents in computing systems and identify events within an enterprise. An anomaly detection system may receive social media messages that include a timestamp indicating when each message was published. The system may generate sentiment identifiers for the social media messages. The sentiment identifiers and timestamps associated with the social media messages may be used to generate a timeseries dataset for each type of sentiment identifier. The timeseries datasets may be input into an anomaly detection model to determine whether an anomaly has occurred. The system may retrieve textual data from the social media messages associated with the detected anomaly and may use the text to determine a computing system or event associated with the detected anomaly.

Detecting system events based on user sentiment in social media messages

Logic may provide a hierarchical label structure for a document with a predicted set of hierarchical labels. Logic may provide a historical label performance database comprising performance data associated with each assignee in a complete set of assignees for each label in a complete set of the hierarchical labels. Logic may generate a first vector for the hierarchical label structure for the document. Logic may generate a second vector for the assignees in an identified set of assignees, comprising each hierarchical label in the predicted set of hierarchical labels for the document, the identified set comprising one or more of the assignees in the complete set of assignees. Logic may perform a similarity search to identify a predicted assignee from the identified set of assignees and logic may predict a selected assignee of the identified set of assignees to associate with the document via the similarity search.

Methods and arrangements for similarity search based on multi-label text classification

A system, processes, and a computer-readable storage medium are provided method includes receiving a search query including one or more of a textual phrase, a document, or data including metadata from a user. For example, a processor may execute a theme-generating machine learning algorithm to generate a theme of the search query. Profiles of other users are identified as similar to a profile of the user may be located. A search history of each of the located other users may be located in the collected search history. Content in the searchable content that corresponds to the generated theme of the search query and according to a result of the evaluation of the located search history of each of the other users may be obtained. The obtained content may be scored based on a scoring algorithm and output a set of scored search results to the input/output device.

Techniques and systems for smart natural language processing of search results

Growing community of inventors

Boston, MA, United States of America

Paul Cho