A Visual Analytics Model with Computational Intelligence Techniques for Adaptive Situation Awareness
Abstract
Visual Analytics (VA) is an emerging technique for logical reasoning using interactive computational methods and visual interfaces. Analysts who are using visualization methods for big data concept exploration in Persistent Surveillance Systems (PSS) increasingly expect to comprehend more distinct relationships and prominent concepts in support of their hypotheses or decisions. In this dissertation a novel Visual Analytics System for PSS (PSS-VA) is presented. PSS-VA facilitates seamless data analysis, data mining, and data visualization for improved situational awareness and actionable decision making. The proposed framework is distinct for its functionalities and suitable for variety of surveillance domains. The proposed PSS-VA takes advantage of advanced Computational Intelligence techniques such as: Principal Component Analysis (PCA), Latent Dirichlet Allocation (LDA), and Vector Space Modeling (VSM) for exploitation and knowledge discovery from metadata of multimodality sensors presented to PSS-VA in a modified Transducer Mark-up Language (TML) data structure. The former two techniques are employed for correlation-based probabilistic exploration of dimensionally reduced data while the latter technique is used to arrive at abstract perceptions enabling rapid knowledge-based relationship discovery than otherwise difficult to comprehend. The experimental results of the developed PSS-VA system demonstrate the effectiveness and efficiency of this proposed approach for ontology-based group activity patterns recognition and exploitation. In overall, the proposed PSS-VA system can significantly enhance situation awareness and expedite intelligent actionable decision making.
Subject Area
Information Technology|Information science|Artificial intelligence
Recommended Citation
Mohammad Serkhail Habibi,
"A Visual Analytics Model with Computational Intelligence Techniques for Adaptive Situation Awareness"
(2014).
ETD Collection for Tennessee State University.
Paper AAI3623017.
https://digitalscholarship.tnstate.edu/dissertations/AAI3623017