A Dynamic Warehousing and Mining System for Large-Scale Human Social Cultural Behavioral Data

We propose to develop a novel Dynamic Warehousing and Mining (DWM) framework for in-situ collection and analysis of Human Social Cultural Behavior (HSCB) data. DWM is a large-scale, dynamic approach for maximizing the ability of intelligence analyst of various kinds to collect, organize and analyze text-based communication to assess HSCB dimensions for a given group and predict current belief states and likely intended actions for that group. For Phase I of this project, the HSCB dimensions of interest are a group"s (i.e., terrorist cell, corporate organization, village, nation) power structure, sense of fatalism, and affective change over time. These are presented as just first steps in a more complete HSCB analysis solution. DWM has several key features: 1) it collects HSCB data from large quantity of unstructured information in an"undetected"fashion, 2) it incorporates linguistic feature analysis and data mining techniques for multiple HSCB dimensions, 3) it utilizes agent technologies to pull in unfiltered data in an untended manner for prolonged durations, and 4) it provides high-dimensional data cubing technology for instant analysis.