User-Centric Similarity and Proximity Measures for Spatial Personalization

User-Centric Similarity and Proximity Measures for Spatial Personalization

Yanwu Yang (Chinese Academy of Sciences, China), Christophe Claramunt (Naval Academy Research Institute, France), Marie-Aude Aufaure (Ecole Centrale Paris, France) and Wensheng Zhang (Chinese Academy of Sciences, China)
DOI: 10.4018/978-1-61350-474-1.ch008
OnDemand PDF Download:
List Price: $37.50


Spatial personalization can be defined as a novel way to fulfill user information needs when accessing spatial information services either on the web or in mobile environments. The research presented in this paper introduces a conceptual approach that models the spatial information offered to a given user into a user-centered conceptual map, and spatial proximity and similarity measures that considers her/his location, interests and preferences. This approach is based on the concepts of similarity in the semantic domain, and proximity in the spatial domain, but taking into account user’s personal information. Accordingly, these spatial proximity and similarity measures could directly support derivation of personalization services and refinement of the way spatial information is accessible to the user in spatially related applications. These modeling approaches are illustrated by some experimental case studies.
Chapter Preview


Personalization offers many opportunities to improve the way information is delivered to a given user. Web personalization should help to remove irrelevant information, to adapt web content and structure, and thus to improve the quality of web services and user’s satisfaction. E-commerce and distance learning are some of domains where web personalization has been successfully applied, while personalization has been also applied to other domains, particularly where the information is complex and diverse per nature.

Spatial web personalization can be considered as an emerging research area of interest as the amount of geo-referenced data distributed either on the web or mobile environments grows exponentially. Let us consider that spatial data represents any item of information that is in one way or the other related to a given geographical location. Namely, geographical entities that appear on a map, or items that represent an information or an object located on Earth (e.g., a photograph, a given landmark).

The development of spatial Web services has been widely active over the past few years, together with significant progresses on spatial information retrieval and location-based services that open novel perspectives for real-time diffusion of geographical data (Harle & Hopper, 2008; Kwon & Shin, 2008). With respect to whether a given user is located in the virtual (such as the web) or in the physical environment, spatially related information services are closely related to the fields of spatial web information retrieval and location-based services.

In the past decade, the progressive integration of spatial information within web pages has been widely investigated to facilitate spatially related information retrieval on the web (Ding et al., 2000; Gravano et al., 2003). Two complementary strategies have been applied: a technique based on the spatial distribution of web links to a given page, and another based on the distribution of spatial references in a given web page. With consideration of user’s location, (Gravano et al., 2003) identified whether a given query submitted to a search engine should be oriented towards either “local” or “global” pages. For example, a query oriented to some general biological information on “wildflowers” should consider global information, while a search for some “houses for sale” after local information, as this later case is locally oriented per definition.

More specifically, location-based spatial queries usually consider spatial queries that retrieve information based on the users’ current locations which is likely to be mobile. When considering location-based spatial queries, a given user (and thus the entities of interest) is likely to be in displacement in the physical environment. Its movement can be represented as a sequence of time-stamped locations. Over the past few years, there has been several works oriented towards mobile spatial queries and sequential patterns (Jayaputera & Taniar, 2005), including nearest neighbor (NN), k nearest neighbor, continuous k nearest neighbor (Zhao et al., 2008; 2009), reverse nearest neighbor (Lee et al., 2008) and range search (Xuan et al., 2008). When considering road networks, (Zhang et al., 2004) studied spatial queries in the presence of obstacles, where the distance between two points is defined as the length of the shortest path that connects them without crossing any obstacles. (Zhang et al., 2003) considered a moving query as a query which is reevaluated only when the query exits the validity scope. Taking location dependency (Goh & Taniar, 2004a) into account, Goh and Taniar (2004b; 2005; 2007) built user profiles based on past mobile visiting data, filters and to mine mobile pattern and association rules from mobile users. (Wang et al., 2006; 2008) presented an approach that derives group patterns of mobile users based on their movement data, where a group pattern is defined as a group of users that are within a distance threshold from one another for at least a minimum duration.

Complete Chapter List

Search this Book: