Cross-modal event summarization system based on network of networks representations

    公开(公告)号:US10757061B1

    公开(公告)日:2020-08-25

    申请号:US15680069

    申请日:2017-08-17

    IPC分类号: H04L12/58 G06Q30/02

    摘要: Described is a system for automated event summarization. A multi-layer network representing a multimodal data set is generated, where nodes within a given layer represent information tokens in a given modality. A topically diverse set of nodes is ranked and selected from each layer to represent temporal event highlights. Temporal event highlights are linked into storylines. Using the storylines, the system monitors a progression of an event or opinions regarding a topic. A temporal summary of the progression of the event or the opinions regarding the topic is generated.

    Video scene analysis system for situational awareness

    公开(公告)号:US10528818B1

    公开(公告)日:2020-01-07

    申请号:US15143471

    申请日:2016-04-29

    摘要: Described is a video scene analysis system. The system includes a salience module that receives a video stream having one more pairs of frames (each frame having a background and a foreground) and detects salient regions in the video stream to generate salient motion estimates. The salient regions are regions that move differently than dominant motion in the pairs of video frames. A scene modeling module generates a sparse foreground model based on salient motion estimates from a plurality of consecutive frames. A foreground refinement module then generates a Task-Aware Foreground by refining the sparse foreground model based on task knowledge. The Task-Aware Foreground can then be used for further processing such as object detection, tracking or recognition.

    Real-time signal detection over very wide band in low SNR environments

    公开(公告)号:US10306486B1

    公开(公告)日:2019-05-28

    申请号:US15188879

    申请日:2016-06-21

    摘要: A method of and apparatus removing of a plurality of relatively narrow banded signals in a relatively wide banded input signal. The method involves and the apparatus provides for compressively sensing one relatively narrow banded signal in the relatively wide banded input signal and removing one relatively narrow banded signal from the relatively wide banded input signal before detecting and removing another relatively narrow banded signal in the relatively wide banded input signal, the step of and apparatus for compressing sensing occurring with respect to both (i) the input signal with the previously detected narrow banded signals removed therefrom and (ii) a frequency shifted version of (i).

    Network of networks diffusion control

    公开(公告)号:US09680702B1

    公开(公告)日:2017-06-13

    申请号:US14294037

    申请日:2014-06-02

    IPC分类号: H04L12/24

    摘要: Diffusion control in a network of networks (NoN) uses a connection degree of node pairs to control diffusion within the NoN. A method of increasing diffusion includes selecting an unconnected node pair having a lowest connection degree and connected nodes of the selected unconnected node pair. A method of decreasing diffusion includes selecting a connected node pair having a highest connection degree and disconnecting nodes of the selected connected node pair. A network diffusion control system includes the NoN and a diffusion controller to one or both of increase a diffusion rate and decrease the diffusion rate within the NoN. A non-transitory computer readable medium includes instructions that implement the diffusion control within the NoN.

    System and method for selecting measurement nodes to estimate and track state in dynamic networks

    公开(公告)号:US11200354B1

    公开(公告)日:2021-12-14

    申请号:US16746436

    申请日:2020-01-17

    摘要: Described is a system for selecting measurement nodes in a distributed physical system of agents. In operation, the distributed physical system is represented as a multi-layer network having a communication layer and an agent layer. The communication layer represents the amount of collective communication activities between any pair of areas and the agent layer represents movement of agents within the distributed physical system such that the communication layer and agent layer collectively generate network dynamics. The network dynamics are modeled as hybrid partial differential equations (PDEs) with measurable interconnected states in the communication layer. Notably, placement of a minimum set of measurement nodes is determined within the distributed physical system to provide full-state observability of the distributed physical system. The system can then track the full system state and apply compensation to one or more agents in the distributed physical system based on tracking the full system state.

    Real-time signal detection over very wide band in low SNR environments

    公开(公告)号:US11096071B2

    公开(公告)日:2021-08-17

    申请号:US16383484

    申请日:2019-04-12

    摘要: A method of and apparatus removing of a plurality of relatively narrow banded signals in a relatively wide banded input signal. The method involves and the apparatus provides for compressively sensing one relatively narrow banded signal in the relatively wide banded input signal and removing one relatively narrow banded signal from the relatively wide banded input signal before detecting and removing another relatively narrow banded signal in the relatively wide banded input signal, the step of and apparatus for compressing sensing occurring with respect to both (i) the input signal with the previously detected narrow banded signals removed therefrom and (ii) a frequency shifted version of (i).

    Model for the competitive seeding of opinions in networks with hybrid gossip-threshold dynamics

    公开(公告)号:US11023904B1

    公开(公告)日:2021-06-01

    申请号:US15944267

    申请日:2018-04-03

    IPC分类号: G06Q30/02 G06F17/11 G06Q50/00

    摘要: Described is a system for determining how opinions spread through a network. Opinion dynamics are applied to a network, each node having a corresponding opinion. Each node is described by an active state or an inactive state such that inactive nodes can update their opinions, and active nodes are fixed in their opinion at the time of activation. Inactive nodes can be influenced by both active nodes and inactive nodes. The opinion dynamics proceed in discrete time steps with an influence step for updating each inactive node's opinion, and a stochastic action step for determining whether an inactive node becomes activated. The system identifies how opinions spread through the network using the applied opinion dynamics, resulting in a set of opinion dynamics data. The opinion dynamics data is used to control information that a device or account is allowed to post to social media platform.

    Automatic anomaly detector
    9.
    发明授权

    公开(公告)号:US10749883B1

    公开(公告)日:2020-08-18

    申请号:US16113837

    申请日:2018-08-27

    IPC分类号: G06F21/55 H04L29/06

    摘要: Described is an automatic anomaly detector that receives a time-series of normal and abnormal activities that include features related to entities within a computing system. A feature coherence graph for the features is constructed, with the graph then clustered such that feature spaces of entities are expanded to include features that live within a same cluster but belong to separate entities. The feature spaces are unified by mapping representations of the features spaces into a Euclidean space of feature vectors. The feature vectors related to each feature are then aligned. Sets of clusters of related abnormal activities are then generated by regressing each feature vector over only those features that it possesses. The sets of clusters are used to detect anomalous behavior. The system then identifies a node within the computer system generating the anomalous behavior and initiates an action to minimize a threat posed by the node.

    Method for determining contagion dynamics on a multilayer network

    公开(公告)号:US10178120B1

    公开(公告)日:2019-01-08

    申请号:US15217393

    申请日:2016-07-22

    IPC分类号: H04L29/06

    摘要: Described is a system for predicting temporal evolution of contagions on multilayer networks. The system determines a functional epidemic threshold for disappearance of a contagion on a multilayer network model according to a score value s=λβ/δ, where λ corresponds to an adjacency matrix of the first layer of the multilayer network model, β represents a spread rate of the contagion, and δ represents a recovery rate. A prediction of future behavior of the contagion on the multilayer network model using the functional epidemic threshold is output and utilized to inform decisions regarding connectivity within a multilayer network in order to prevent spread of the contagion on a multilayer network.