Pattern-Aware Prediction for Moving Objects

タイトル: 「Pattern-Aware Prediction for Moving Objects」
講演者: Xiaofan Zhou客員教授（オーストラリア・クイーンズランド大学教授）
主催: 情報連携基盤センター
共催: 名古屋大学大学院情報科学研究科

概要: Predication of future locations of moving objects can enable a wide range of applications, such as in ITS and location-based services, especially when accurate predications can be made beyond the immediate future . Existing location prediction techniques are limited in their ability to support such applications since they are generally capable only of very-short-term predictions (e.g., up to the next road junction or in next a few minutes). In this talk, we will explore the potential and techniques for pattern-aware predictions that can address this limit. A hybrid prediction model is proposed, to consider not only the current motion of an object but also trajectory patterns discovered using novel data mining algorithms form history trajectory data. We will also discuss identification and prediction for a group of moving objects that form a convoy (i.e., moving together over a period of time). This talk is based on our recent work published in ICDE and VLDB conferences in 2008.

Xiaofan Zhou客員教授講演会