On the practicalities of place-based virtual communities: Ontology-based querying, application architecture, and performance
Introduction
Mobile devices such as PDAs, mobile phones and pagers have woven themselves into people’s lives. There is a trend towards a virtual community co-existing with a physical place. In other words, places contain not only physical objects, real people, but also digital services and digital devices, forming, increasingly, over a place, a kind of Place-Based Virtual Community (PBVC). This paper identifies and addresses the need for capturing, storing, and understanding the context dependencies between a physical place and the social context of users within a community in order to build context-aware place-based services. In previous work (Nguyen, Loke, Torabi, & Lu, 2011), we proposed a systematic approach for developing place-based context-aware applications via the introduction of the PlaceComm framework, namely, we build an ontology, and then encapsulate the queries over the ontology as agents to provide services. A key distinguishing idea of this work is that the ontology also serves as a “standardized” schema for knowledge about a place. The purpose of such a “standardized” schema is that different context-aware applications for a place can be programmed against this schema, and different places can be expected to have the kind of knowledge mentioned in the schema.
This paper extends previous work as follows. We provide an updated description on the PBVC ontology and an updated description on the PlaceComm architecture, compared to what was presented in Nguyen et al. (2011). This paper presents more details on the format of items in the knowledge base, and presents in detail a comprehensive series of queries to illustrate the ideas of the work, not previously described elsewhere. We address questions and research gaps relating to the practical aspects of the PlaceComm approach to mobile services, namely, (i) what queries can be used for PBVCs and whether such queries can be efficiently evaluated against the ontology we developed, and (ii) what architectures for applications that use PlaceComm and how they relate to the generic PlaceComm architecture (e.g., whether the architectures of specific applications be viewed as a specialisation of the generic PlaceComm architecture). Also, here, we discuss trust aiding mechanisms and place discovery, not in previous work. This paper provides a comprehensive performance evaluation not previously published elsewhere. This paper relates the applications (in Section 5) to the generic PlaceComm framework in a new way, showing the specific architectures and how they are derived from the generic PlaceComm architecture. We demonstrate the system’s versatility by developing prototype applications based on our implemented framework.
As advocated by Gordon and Silva, 2011, Quelch and Jocz, 2012, this paper highlights the “return of focus” to locality and place that the mobile Internet is enabling.
The paper is organized as follows. Section 2 reviews the background for the paper. Section 3 details the notion of Place-based Virtual Community, from definition to theory and representation in the form of a PBVC ontology built using the Web Ontology Language (OWL).1 Section 4 details the PlaceComm framework from its conceptual model to its agent architecture. Section 5 introduces a number of application prototypes built on the PBVC framework. Section 6 presents further evaluation of the PlaceComm framework, including performance measurement and scalability of querying. We conclude the paper in Section 7 with future directions.
Section snippets
Context-aware frameworks and services
There are detailed surveys on context-aware frameworks and services, e.g., in Baldauf et al., 2007, Endres et al., 2005, Sheng et al., 2010, Raz et al., 2006. Our work differs from earlier context-aware middleware and toolkits (Dey, 2001) in that these toolkits and middleware generally do not support reuse of a common knowledge base across different applications or allow applications to contribute knowledge. Site-specific services accessed via smartphones were considered in Toye, Sharp,
The PBVC ontology
The PBVC ontology was first mentioned in Nguyen, Loke, Torabi, and Lu (2009). We published initial ideas on the concepts, architecture, and ontology of the PBVC in Nguyen et al., 2007, Nguyen et al., 2008, Nguyen et al., 2010, Nguyen et al., 2010, but this paper provides a more comprehensive and update-to-date view of the ontology. Although, in building the PBVC ontology we incorporated concepts from other ontologies, the core part of the ontology is novel. Also, we identified particular
Implementation of the PlaceComm framework
This section details the PlaceComm (multiagent) framework, which is built based on the PBVC concepts. The ontology and framework were first discussed in our previous paper (Nguyen et al., 2011), but here, we provide a more update-to-date description. PlaceComm has been prototyped using a multiagent architecture with the JADE platform (Bellifemine, Poggi, & Rimassa, 2001). In previous work, we implemented the protocols for members joining and leaving a PBVC (Nguyen et al., 2007).
The layered
Applications using PlaceComm
In this section, we first illustrate the support for place discovery and community-based trust queries in our framework, and then describe applications that have been built based on the PlaceComm framework. All of the applications described in this section have been published via conferences, workshops and journal articles (Nguyen et al., 2007, Nguyen et al., 2008, Nguyen et al., 2009, Nguyen et al., 2009, Nguyen et al., 2010, Nguyen et al., 2010). However, here, we briefly summarize three of
Performance evaluation of the PlaceComm framework
In this section, we detail an evaluation and discuss the challenge of building context-aware applications with regard to performance.
Conclusion and future work
The notion of place should be central when building mobile services, and we believe this has not received adequate emphasis so far. On this note, we presented the idea of basing place-specific services on a place ontology. A key distinguishing feature of the Place-Comm framework is a common context knowledge base, shared among applications and contributed by agents. This makes the context more reusable, and context contributed in one service may be used in other services. In addition, the
References (42)
- et al.
A survey of context modelling and reasoning techniques
Pervasive and Mobile Computing
(2010) - Bainbridge, D., Jones, M., & Arter, D. (2007). A map-based place-browser for a PDA, Technical report, University of...
- et al.
A survey on context-aware systems
International Journal of Ad Hoc and Ubiquitous Computing
(2007) - et al.
Jade: A fipa 2000 compliant agent development environment
- et al.
Managing an integrated ubicomp environment using ontologies and reasoning
- et al.
Handbook of research on urban informatics: The practice and promise of the real-time city
(2009) - et al.
Real time Rome
Networks and Communication Studies – Official Journal of the IGU’s Geography of Information Society Commission
(2007) - et al.
Browsing the world: Bridging pervasive computing and the web
- Chen, H., Perich, F., Finin, T., & Joshi, A. (2004). Soupa: Standard ontology for ubiquitous and pervasive...
Understanding spaces as places: extending interaction design paradigms
Cognition, Technology & Work
(2004)