Spatial Predictive Queries

Abstract

In this seminar, we address spatial predictive queries both in Euclidian spaces and over road networks. We provide a definition for various types of spatial predictive queries, describe current research trends, and envision future directions. We present practical application scenarios and emphasize the roadblocks that are holding industry back from the commercialization of spatial predictive queries. This seminar targets audience in mobile data management, spatiotemporal query processing, mobile crowd sourcing, and tracking of moving objects. I. TUTORIALOUTLINE The next generation of location based services would offer and recommend services for users according to their current locations as well as their future destinations [20], [21], [23], [25], [26], [29], [34], [37], [38], thanks to the widespread of mobile devices [17]. Nowadays, search engines, e.g., Microsoft Bing and Google, offer satisfactory location-aware search results that are based on the user's current location. However, these search engines remain in an infancy phase developing search techniques that takes into consideration the user's future (or intended) destination. Experience tells that the advertisement that targets the user's current location (e.g., a coupon in a nearby shopping mall) is not very effective. More specifically, it is too late to attract the user's attention to a nearby service because most users, by then, are already heading to a preplanned destination. Industry believes that targeting the user with services that are around his future destination is more valuable to the user (from a relevance perspective) and more profitable to industry (from a market share perspective). However, experience tells that users are either reluctant to share their future destinations with search engines or are unaware of the value they may get by doing so. Consequently, the prediction of the user's future location combined with spatial query processing has been gaining tremendous interest in both the research and industrial communities. Entering this era of "future-location-aware" search engines requires the ability to process spatial predictive queries.In this seminar, we survey the existing research and envision the future of spatial predictive query processing and optimization. The seminar is organized in the following key sections: A. Part 1: Spatial Predictive Queries, What and Why?In this part, we provide basic definitions for different types of predictive queries. Then, we show the importance of this topic through real-world example applications and systems. In general, predictive queries [19], [21], [25], [26] aim at answering inquiries about the anticipated future locations of a set of moving objects, either in an Euclidean space or over a road network. The fundamental types of spatial predictive queries include: (a) predictive point query [19], [18], that finds out the objects that are most likely to show up around a specific location point in the space within a future time window, (b) predictive range query [26], [49], [60], where a user...