University of Michigan

Michigan State University

Mingyan Liu

Manish Karir

Jing Zhang

Michael R Bailey

Yang Liu

Tongtong Li

Michael Donald Bailey

Armin Sarabi

Kun Jin

Tongxin Yin

Bruno Giordani

Jinxian Deng

Boxin Sun

Zakir Durumeric

Voyko Kavcic

A method and system for assessing brain impairment includes generating electroencephalogram (EEG) signals from a plurality of electrodes, removing artifacts from the EEG signals to generate artifact reduced signals, generating current source density signals from the artifact reduced signals, performing multiscale analysis of dynamic functional connectivity of a brain based on the current source density signals using a plurality of different sized time windows, generating hard classifiers for each of the plurality of different size time windows, selecting classifiers from the hard classifiers to form selected classifiers, performing majority voting on a discrimination of normal cognition and mild cognitive impairment using the selected classifiers, generating an EEG-based health score of overall brain activity based on majority voting of the selected classifiers and generating a display corresponding to the health score of brain activity.

Mild cognitive impairment detection and prediction system

A computer-implemented method is presented for scanning a computer network. The method includes: a) sending a particular network probe to a network address in a computer network; b) receiving a response to the network probe from the network address; c) appending the response to a set of features forming a feature vector; d) determining a next network probe to conduct at the network address; and e) predicting, by the computer processor, the response from the next network probe using the feature vector and a model, where the model is trained using a machine learning method and outputs a probability that a given network address will respond to a network probe.

Adaptive network probing using machine learning

Embodiments are disclosed for profiling network-level malicious activity. Profiling embodiments include observing malicious activity, representing such activity in accordance with a set of representative features, capturing temporal evolution of this malicious behavior and its dynamics, and using this temporal evolution to reveal key risk related properties of these networks. Embodiments are further disclosed addressing the connectedness of various networks and similarity in network-level maliciousness. Embodiments directed to similarity analyses include focusing on the notion of similarity&#x2014;a quantitative measure of the extent to which the dynamic evolutions of malicious activities from two networks are alike, and mapping this behavioral similarity to their similarity in certain spatial features, which includes their relative proximity to each other and may be used to help predict the future maliciousness of a particular network. The embodiments described may be applicable to various network aggregation levels.

Rating network security posture and comparing network maliciousness

Network security and robustness is analyzed by developing correlations among network maliciousness observations to determine attack susceptibility. Network traffic is analyzed at the autonomous system (AS) level, among connected Internet Protocol (IP) routing prefixes, to identify these observations. The traffic is monitored for any of a number of specified mismanagement metrics. Correlations among these metrics are determined and a unified network mismanagement metric is developed, indicating network susceptibility to potentially malicious attack.

Network maliciousness susceptibility analysis and rating

Growing community of inventors

Ann Arbor, MI, United States of America

Mingyan Liu