A system includes an ingestion component configured to receive a request from an entity for content related to a content item and a user identity. The request has a content identifier representative of the content item and a token. A request processing component of the system is configured to access a database associated with the system and identify the content item and the user identity using the content identifier and the token, wherein the database has information associating the token with the user identity and associating the content identifier with the content item. In response to identification of the content item and the user identity, the request processing component directs a recommendation engine associated with the system to identify the content related to the content item and the user identity. Information identifying the content related to the content item and the user identity is then transmitted back to the entity.

There is described a neural network system implemented by one or more computers for determining graph similarity. The neural network system comprises one or more neural networks configured to process an input graph to generate a node state representation vector for each node of the input graph and an edge representation vector for each edge of the input graph; and process the node state representation vectors and the edge representation vectors to generate a vector representation of the input graph. The neural network system further comprises one or more processors configured to: receive a first graph; receive a second graph; generate a vector representation of the first graph; generate a vector representation of the second graph; determine a similarity score for the first graph and the second graph based upon the vector representations of the first graph and the second graph.

Described are a system, method, and computer program product for user network activity anomaly detection. The method includes receiving network resource data associated with network resource activity of a plurality of users and generating a plurality of layers of a multilayer graph from the network resource data. Each layer of the plurality of layers may include a plurality of nodes, which are associated with users, connected by a plurality of edges, which are representative of node interdependency. The method also includes generating a plurality of adjacency matrices from the plurality of layers and generating a merged single layer graph based on a weighted sum of the plurality of adjacency matrices. The method further includes generating anomaly scores for each node in the merged single layer graph and determining a set of anomalous users based on the anomaly scores.

A sleep training device is disclosed comprising certain electronic components, audio components, light-emitting diodes (LEDs), physical switches, a capacitive touch component, or a combination thereof. The sleep training device can have a housing configured to house the electronic components, the audio components, the LEDs, the capacitive touch component, and at least part of the physical switches. The sleep training device can receive an instruction to schedule a sleep program based on a set parameter from another device in wireless communication with the sleep training device, initiate the sleep program once the set parameter is met, and disable at least one of the physical switches and the capacitive touch component when the sleep program is initiated.

Aspects of the present disclosure are presented for an AI system featuring specially designed AI hardware that incorporates security features to provide iron clad trust and security to run AI applications/solution models. Presented herein are various security features for AI processing, including: a trust and integrity verifier of data during operation of an AI solution model; identity and trust establishment between an entity and the AI solution model; secure isolation for a virtual AI multilane system; a real-time attack detection and prevention mechanism; and built in detection mechanisms related to rogue security attack elements insertion during manufacturing. Aspects also include security to implement an AI network interconnecting multiple user devices in an AI environment.

Systems and methods for responding to incidents in a building are provided. One method includes performing operations including retrieving data relating to one or more on-premises building devices of the building, determining the incident relating to the one or more on-premises building devices, and determining a plurality of potential responses to the incident. The operations further include analyzing the plurality of potential responses and determining relative risks of the potential responses to the incident and transmitting, to a first on-premises building device of the on-premises building devices, data indicating the incident and at least one of the plurality of potential responses based on the relative risks of the potential responses.

Systems and methods for responding to incidents in a building are provided. One method includes performing operations including retrieving data relating to one or more on-premises building devices of the building, determining the incident relating to the one or more on-premises building devices, and determining a plurality of potential responses to the incident. The operations further include analyzing the plurality of potential responses and determining relative risks of the potential responses to the incident and transmitting, to a first on-premises building device of the on-premises building devices, data indicating the incident and at least one of the plurality of potential responses based on the relative risks of the potential responses.

Methods and systems for contingency communication are disclosed. In one embodiment, a method for providing emergency services may be performed by a base station operating in a communication system in an embodiment, the method for providing emergency services includes transmitting a beacon signal to indicate an emergency status to enable portable devices to operate in a stress mode. A distress signal may be transmitted by a mobile device in response to the beacon signal to the base station, wherein the distress signal carries information at least comprising user identity associated with the mobile device, geolocation of the mobile device, or biometrics of a user of the mobile device.

An animal tracking system comprising a first LoRa receiver configured as a gateway to receive transmissions on a first set of channels or a second set of channels and forward said transmissions to a first remote server, and a transmitting tag configured to be affixed to an animal. The transmitting tag includes a memory configured to store digital information, a passive RFID receiver configured to detect a close proximity to an RFID transmitter and store detection of the close proximity in comprising in the memory, a processor configured to determine an alarm status according to a frequency of detection of the close proximity, an internal energy source, and a LoRa transmitter configured to transmit signals for a distance greater than 100 meters via energy from the internal energy source.

A system may include: a server comprising a rule cache; a user device communicably coupled to the server; a computer-readable medium comprising instructions that cause the server to: monitor a plurality of third-party data sources; obtain, via a queueing service, a plurality of pieces of content from the plurality of third-party data sources; for each piece of content, fetch a rule ID from a list of rule IDs on the user device, wherein the rule ID is fetched based on the content and a pre-selected setting on the user device; use the rule ID to fetch a rule from the rule cache, the rule comprising a script, the script comprising executable code; execute the script on the piece of content to determine if the piece of content matches the rule; and in response to determining that the piece of content matches the rule, send an alert to the user device.