The present invention comprises a novel Virtual Facilities Manager architecture that optimizes home health by combining multiple prediction engines, fed by continuously monitored and processed sensor-based and environmental data (historical as well as current), with an integrated Homecare Network of homeowners and service and other providers. This enables a continuous feedback process in which abnormal conditions (symptoms of an underlying problem, including suboptimal performance) are detected and addressed by issuing contextual alerts with associated actions (related alert-action pairs) in an iterative manner over time. By maintaining a Home Health Record including Scores (e.g., reliability and energy efficiency) representing the systemic state of one or more homes, the present invention detects not only maintenance issues, but also suboptimal performance “problems,” and addresses them with the same iterative troubleshooting approach until such scores return to an acceptable level.

Some embodiments are directed to a system that includes multiple fire sensors and equipment, and mobile devices of the users of the system. These fire sensors and equipment are configured in a network so as to allow communication with a processor. A processor in the system interrogates the status of the fire sensors and fire safety elements on a continuous basis. Upon a change in system status, the processor interprets the change and classifies according to operation state or probability of a fire or other imminent threat. The data, processing and storage provide the means by which occupants and potential occupants can better navigate a building during the occurrence of a fire. The system enables communication between users of the system through their mobile devices used and communication between devices that may be present in the

Some embodiments are directed to a system that includes multiple fire sensors and equipment, and mobile devices of the users of the system. These fire sensors and equipment are configured in a network so as to allow communication with a processor. A processor in the system interrogates the status of the fire sensors and fire safety elements on a continuous basis. Upon a change in system status, the processor interprets the change and classifies according to operation state or probability of a fire or other imminent threat. The data, processing and storage provide the means by which occupants and potential occupants can better navigate a building during the occurrence of a fire. The system enables communication between users of the system through their mobile devices used and communication between devices that may be present in the

A device and method for receiving real-time warning messages about a hazardous situation, for which an alert or a mitigating control action may be warranted, with the means to measure the effects forecasted by the parameters contained in the warning messages, adjust the parameters to more accurately reflect local conditions, and provide feedback about the performance and accuracy of the system sending the warning messages.

A device and method for receiving real-time warning messages about a hazardous situation, for which an alert or a mitigating control action may be warranted, with the means to measure the effects forecasted by the parameters contained in the warning messages, adjust the parameters to more accurately reflect local conditions, and provide feedback about the performance and accuracy of the system sending the warning messages.

A method for predicting air quality with the aid of machine learning models includes: (A) providing air pollution data to perform an eXtreme Gradient Boosting (XGBoost) regression algorithm for obtaining a XGBoost prediction value; (B) providing the air pollution data to perform a Long Short-Term Memory (LSTM) algorithm for obtaining an LSTM prediction value; (C) combining the air pollution data, the XGBoost prediction value and the LSTM prediction value to generate air pollution combination data; (D) performing an XGBoost classification algorithm to obtain a suggestion for whether to issue an air pollution alert; and (E) performing the XGBoost regression algorithm on the air pollution combination data to obtain an air pollution prediction value. Two layers of machine learning models are built, and a situation where prediction results are too conservative when a single model does not have enough data can be improved.

A glove comprises a textile component and an electronic module component wherein the textile component comprises in order: a flame-resistant fabric having inner and outer surfaces, a first temperature detecting sensor located next to the inner surface of the flame-resistant fabric, at least one thermally insulating fabric having inner and outer surfaces, a second temperature detecting sensor located on the inner surface of the thermally insulating fabric that is closest to the skin of the wearer and an electronic module component capable of predicting body burns comprising a removable encapsulated electronic module located in a pocket of the glove.

A glove comprises a textile component and an electronic module component wherein the textile component comprises in order: a flame-resistant fabric having inner and outer surfaces, a first temperature detecting sensor located next to the inner surface of the flame-resistant fabric, at least one thermally insulating fabric having inner and outer surfaces, a second temperature detecting sensor located on the inner surface of the thermally insulating fabric that is closest to the skin of the wearer and an electronic module component capable of predicting body burns comprising a removable encapsulated electronic module located in a pocket of the glove.

The present disclosure generally relates to a system for crime risk forecasting using cyber security and deep learning comprises a data input unit for receiving a pre-stored crime event dataset and real time crime event data input along with geographical details of an area; a classification processing unit for categorizing pre-stored crime event dataset and real time crime event data input according to crime type; a graphical user interface for entering a target geographic area for forecasting upcoming crime risk; a central processing unit for generating a crime risk forecast based on the historical crime incident stored in the pre-stored crime event dataset using a deep leaning technique; and a control unit coupled to a display for displaying a crime risk ranking generated based on the crime risk forecast and one or more crime risk event for the target geographic area.

Apparatuses, systems, devices, and computer program products for deterrence techniques are described. A method may include processing data from an outdoor camera. A method may include detecting a person based at least in part on processed data from an outdoor camera. A method may include emitting a sound from an outdoor camera to deter a person in response to detecting the person.