Browse Prior Art Database

Event Correlation and Visualization for Investigation Research to Predicted Events

IP.com Disclosure Number: IPCOM000246431D
Publication Date: 2016-Jun-06
Document File: 3 page(s) / 62K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for correlating data from various streaming data sources and providing the ability to visualize the relationship of events / incidents by representing the relationship bond strength by thickness of lines and color coding.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Event Correlation and Visualization for Investigation Research to Predicted Events

Disclosed is a method and system for collecting data from multiple streaming data sources and performing correlation analysis on defined or undefined events and incidents to facilitate investigative research using a modified cascade correlation (CC) algorithm. The method and system identifies relationship bond strength of correlated events / incidents from contextual analysis using a modified CC algorithm from multiple data sources, and provides the ability to visualize the relationship of events / incidents by representing the relationship bond strength by thickness of lines and color coding. The method also enables historical analysis and investigation analysis that is used to create relationship among the events to attempt to find direct or indirect related information among the events to predict future events.

As shown in Figure 1, the system includes a centralized cognitive module that gathers information from various streaming data sources such as, social network content, online newspapers, blogs, video surveillance feeds, and forensic reports.

Figure 1

The centralized cognitive module performs contextual analysis to find one or more events and/or incidents. The event or incident could be an accident, illegal activity, birthday party, robbery, get together, etc. There are two approaches that can be employed. The contextual analysis could be performed on wide range of event or incident types. Alternatively, the contextual analysis could be focus on a specific type of event / incident for the needs of the interested party.

The centralized cognitive module identifies various events from different source analysis, and then extracts various dimensions of each events, dimension can be who have attended, location, and type of event. The extracted dimensions are part of the event metadata. Based on the metadata information detail of any even...