University of Washington

Microsoft Technology Licensing, LLC

James Anthony Fogarty

Shwetak N Patel

Eric C Larson

Jon E Froehlich

Desney S Tan

Ashish Kapoor

Simon A J Winder

Steven Mark Drucker

Tadayoshi Kohno

Morgan Dixon

Kayur Dushyant Patel

Daniel Leventhal

Michael Toomim

James Landay

Nathan Morris

Xianhang Zhang

Simon AJ Winder

By monitoring pressure transients in a liquid within a liquid distribution system using only a single sensor, events such as the opening and closing of valves at specific fixtures are readily detected. The sensor, which can readily be coupled to a faucet bib, transmits an output signal to a computing device. Each such event can be identified by the device based by comparing characteristic features of the pressure transient waveform with previously observed characteristic features for events in the system. These characteristic features, which can include the varying pressure, derivative, and real Cepstrum of the pressure transient waveform, can be used to select a specific fixture where a valve open or close event has occurred. Flow to each fixture and leaks in the system can also be determined from the pressure transient signal. A second sensor disposed at a point disparate from the first sensor provides further event information.

Sensing events affecting liquid flow in a liquid distribution system

Systems and methods for implementing real-time pixel-based reverse engineering of interface structures are disclosed herein. The system uses the pixels representing widgets and content of a user interface to analyze, interpret, and/or enhance the interface without requiring cooperation from the interface. The system captures pixel data and reverse engineers the composition and structure of the interface from the captured pixel data. Using the reverse engineered interface, generated based on the captured pixel data, the facility can identify and record user interactions with the interface, modify the display of the interface to meet user requirements, preferences, or customization options, or enhance the interface to provide functionality not otherwise available, etc. The system can perform these processes across a variety of interfaces regardless of their underlying implementation and without any modifications to the interface.

Systems and methods for implementing pixel-based reverse engineering of interface structure

An interactive concept learning image search technique that allows end-users to quickly create their own rules for re-ranking images based on the image characteristics of the images. The image characteristics can include visual characteristics as well as semantic features or characteristics, or may include a combination of both. End-users can then rank or re-rank any current or future image search results according to their rule or rules. End-users provide examples of images each rule should match and examples of images the rule should reject. The technique learns the common image characteristics of the examples, and any current or future image search results can then be ranked or re-ranked according to the learned rules.

Interactive concept learning in image search

A multiple model data exploration system and method for running multiple machine-learning models simultaneously to understand and explore data. Embodiments of the system and method allow a user to gain a greater understanding of the data and to gain new insights into their data. Embodiments of the system and method also allow a user to interactively explore the problem and to navigate different views of data. Many different classifier training and evaluation experiments are run simultaneously and results are obtained. The results are aggregated and visualized across each of the experiments to determine and understand how each example is classified for each different classifier. These results then are summarized in a variety of ways to allow users to obtain a greater understanding of the data both in terms of the individual examples themselves and features associated with the data.

Exploring data using multiple machine-learning models

Access to resource(s) intended to be shared with specific groups of individuals is controlled using concise tests of shared knowledge instead of (or in addition) to accounts and access control lists. Users can readily learn the concept and choose questions that will control the access by the desired group with little effort. Such questions can be relatively secure to guesses by those not intended to have access, particularly if the number of allowed guesses is relatively limited. Users can generally predict the security of their questions, but sometimes underestimate the ability of attackers to use Web searching or enumeration to discover answers. In such cases, the system can automatically discover weak questions and then suggest alternatives. By lowering the threshold to access control, shared knowledge tests can enable more types of information to acquire collaborative value on the Internet and on other types of networks.

Access control by testing for shared knowledge

Growing community of inventors

Seattle, WA, United States of America

James Anthony Fogarty