Omniscience Corporation

Aol Advertising Inc.

Comcast Cable Communications, LLC

Manu Shukla

Pero Subasic

Alexander Wesley Contryman

Jiliang Chang

Guoying Zhang

Jacob Ryan Van Gogh

Kenin Coloma

Oliver Jojic

Hans Marius Holtan

Yan Qu

Neil Wayne Cohen

Jaimyoung Kwon

William Bertrand

Brian Whitney

Robert S Cabacungan

Rohan A Mehta

Jay L Chang

Kenin A H Coloma

A method and apparatus for performing automated underwriting based on document images are described. The method may include receiving a document image corresponding to a form and data contained within the form for which an underwriting decision is to be rendered. The method may also include performing a first machine learning based analysis of the document image by extracting one or more keys and one or more values corresponding to the keys from the document image using a first machine learning model trained to digitize the form from the document image. The method may also include performing a second machine learning based analysis using the one or more keys and one or more values corresponding to keys as an input into a second machine learning model trained to make the decision based on key, value pairs. Furthermore, the method may include transmitting, to an organization system, a user interface generated using the decision and a confidence of the decision determined from the second machine learning based analysis.

Systems and methods for descriptive machine learning based underwriting

Aspects of the subject technology include an event processing and prospect identifying platform. It accepts as input a set of storylines (a sequence of entities and their relationships) and processes them as follows: (1) uses different algorithms (LDA, SVM, information gain, rule sets) to identify themes from storylines; (2) identifies top locations and times in storylines and combines with themes to generate events that are meaningful in a specific scenario for categorizing storylines; and (3) extracts top prospects as people and organizations from data elements contained in storylines. The output comprises sets of events in different categories and storylines under them along with top prospects identified. Aspects use in-memory distributed processing that scales to high data volumes and categorizes generated storylines in near real-time.

Event categorization and key prospect identification from storylines

A method and apparatus for recognizing and extracting data from a form depicted within an image of a document are described. The method may include receiving the image of the document, the image depicting the form and data contained one the form. The method may also include transforming the image of the document to a set of one or more key, value pairs by processing the image of the document with a sequence of two or more trained machine learning based image analysis processes, wherein keys are relevant to forms of the type depicted in the form, and wherein each value is associated with a key. The method may also include generating a data output that comprises the set of key, value pairs for textual data recognized and extracted from the form depicted in the image.

Systems and methods for machine learning based content extraction from document images

Aspects of the present disclosure relate to a distributed storytelling framework. A server receives an adjacency list comprising a set of nodes linked together by edges. The server converts the adjacency list to a set of generated storylines, each storyline being represented as a key-value pair. A key represents a first node and a value represents a second node linked to the first node by an edge. The server combines first and second storylines, of the set of generated storylines, to generate an additional storyline in response to a value from a first storyline matching a key from a second storyline. The additional storyline includes a single key and multiple values, and is added to the set of generated storylines. The server repeats combining storylines, of the set of generated storylines, to generate additional storylines. The server provides an output corresponding to at least one of the generated storylines.

Distributed storytelling framework for intelligence analysis

Drones have become ubiquitous in performing risky and labor intensive areal tasks cheaply and safely. To allow them to be autonomous, their flight plan needs to be pre-built for them. Existing works do not precalculate flight paths but instead focus on navigation through camera based image processing techniques, genetic or geometric algorithms to guide the drone during flight. That makes flight navigation complex and risky. We present automated flight plan builder DIFPL which pre-builds flight plans for drones to survey a large area. The flight plans are built for subregions and fed into drones which allow them to navigate autonomously. DIFPL employs distributed paradigm on Hadoop MapReduce framework. Distribution is achieved by processing sections or subregions in parallel. Experiments performed with network and elevation datasets validate the efficiency of DIFPL in building optimal flight plans.

Distributed drone flight path builder system

Systems and methods are disclosed for contextual analysis and segmentation of information objects. In accordance with one implementation, information objects, such as web pages and user profiles, may be processed to obtain a list of key terms. An index may be created containing each of the information objects and associated key terms. Information objects may then be matched to contextual segments (i.e., categories of information objects) by submitting terms associated with the contextual segments against the index. Further, thresholding may be applied, so that only the most relevant information objects for a contextual segment are assigned to the contextual segment.

Systems and methods for contextual analysis and segmentation of information objects

In one aspect, a recommendation system offers item recommendations to users based on one or more items known to be liked by the users. An item may be recommended to a user if a similarity indicator for the item, established by determining how much more likely than expected the user will like the item based on the user liking another item, exceeds a predetermined threshold. Multiple items may be recommended to a user based on relative similarity indicators.

Recommendation system

Systems and methods are disclosed for contextual analysis and segmentation of information objects. According to one implementation, information objects, such as web pages and user profiles, may be analyzed to identify key terms. These key terms may be included in a contextual representation of an information object. By comparing the contextual representations of a plurality of information objects, one or more contextual segments (i.e., categories of information objects) may be created. Each contextual segment may also be associated with its own contextual representation. Once a contextual segment has been created, information objects may be assigned to the contextual segment. These contextual segments may be used to deliver targeted advertising, for example.

Systems and methods for contextual analysis and segmentation using dynamically-derived topics

A computer-implemented method is provided for managing online user profiles. The method includes receiving information about online activity of a user, the information including at least a unique user ID and a URL requested by the user; fetching textual content associated with the URL requested by the user; generating one or more keywords based on the textual content; and generating or updating a user profile of the user to include the one or more keywords. A system for managing online user profiles is also provided.

Growing community of inventors

Sterling, VA, United States of America

Manu Shukla