Urban Airship, Inc.

Other

Airship Group, Inc.

Gary Todd Johnson

Erik A Onnen

Jeffrey A Towle

Dave K Revell

Nathan A Putnam

Kerry D Roosmalen

Jeffrey F Wagner

Elizabeth Marjory Orr

Michael S Despain

Darren K Riley

Neel Banerjee

Disclosed is a churn prediction system that predicts with a high level of accuracy which users will and which users will not stop opening the app over a 30-day time period. To this end a model is created using historical event data where the churn-related behavior of each user is known. New event data is then applied to the model to determine the likelihood of each user churning in the future. With these prediction scores a user is then qualified as falling into one of three classifications: low-risk, medium-risk, or high-risk of churn.

Churn prediction with machine learning

Push reporting is disclosed. A first payload comprising feature reporting information and meta information is received from a first device at a first time. A second payload comprising feature reporting information and meta information is received from a second device at a second time. At respective third and fourth times, respective third and fourth payloads are received from the first and second devices, respectively. The third and fourth payloads include additional feature reporting information and meta information. The processing of the information received from the first and second devices is coordinated. A request for information is received. Metrics are provided as output. The metrics are based at least in part on the processing of the information received from the first and second devices.

Push reporting

Push Reporting is disclosed. A first payload comprising feature reporting information and meta information is received from a first device at a first time. A second payload comprising feature reporting information and meta information is received from a second device at a second time. At respective third and fourth times, respective third and fourth payloads are received from the first and second devices, respectively. The third and fourth payloads include additional feature reporting information and meta information. The processing of the information received from the first and second devices is coordinated. A request for information is received. Metrics are provided as output. The metrics are based at least in part on the processing of the information received from the first and second devices.

A configurable base is formed from a plurality of interlocking tiles that may fit within a frame. Anchoring tiles have receptacles for accepting anchors. Smooth panels support heavy items. Accessories are formed to either attach to the anchor or integrally with the anchor. Accessories include tie downs, sidewalls, cross frame and stanchions, and wheel-chocks. The anchor fits into the receptacles for placement anywhere desired on the base. An embodiment of the anchor has a lock for added security. A coupler provides at least one aperture for securing items, preferably positioned on the frame. An alternative configurable bed is formed by orthogonal cross members arranged within a frame. Open spaces between the orthogonal cross members accept anchors. Another alternative configurable bed is formed from a single tile having a plurality of receptacles. A method of configuring a cargo management system allows a dealer employee to compare customer needs to established footprints, alter the footprint design if needed, and generate a cost estimate for the configuration.

Cargo management system for trailers and truckbeds and method of use

A backpack blower includes a control handle carried by a flexible arm which permits the control handle to be positioned at any location in a three dimensional position range. User comfort and controllability are enhanced.

Growing community of inventors

Portland, OR, United States of America

Gary Todd Johnson