Abstract | ||
---|---|---|
Behavior analysis using trajectory data presents a practical and interesting challenge for KDD. Conventional analyses address discriminative tasks of behaviors, e.g., classification and clustering typically using the subsequences extracted from the trajectory of an object as a numerical feature representation. In this paper, we explore further to identify the difference in the high-level semantics of behaviors such as roles and address the task in a cross-domain learning approach. The trajectory, from which the features are sampled, is intuitively viewed as a domain, and we assume that its intrinsic structure is characterized by the underlying role associated with the tracked object. We propose a novel hybrid method of spectral clustering and density approximation for comparing clustering structures of two independently sampled trajectory data and identifying patterns of behaviors unique to a role. We present empirical evaluations of the proposed method in two practical settings using real-world robotic trajectories. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/ICDM.2011.125 | Data Mining |
Keywords | Field | DocType |
learning (artificial intelligence),pattern clustering,robots,KDD,behavior discriminative tasks,behavior high-level semantics,behavior pattern identification,cross-domain learning approach,density approximation,robotic trajectories,role-behavior analysis,spectral clustering,trajectory data,density-based outlier detection,time-series subsequence clustering,trajectory data mining,transfer learning | Data mining,Spectral clustering,Pattern clustering,Computer science,Transfer of learning,Artificial intelligence,Cluster analysis,Discriminative model,Trajectory,Pattern recognition,Robot,Machine learning,Semantics | Conference |
ISSN | ISBN | Citations |
1550-4786 | 978-1-4577-2075-8 | 2 |
PageRank | References | Authors |
0.36 | 24 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Shin Ando | 1 | 140 | 13.89 |
Einoshin Suzuki | 2 | 853 | 93.41 |