UrbanHubble: Location Prediction and Geo-Social Analytics in LBSN
Massive amounts of geo-social data is generated daily. In this paper, we propose UrbanHubble, a location-based predictive analytics tool that entails a broad range of state-of-the-art location prediction and recommendation algorithms. Besides, UrbanHubble
- PDF / 2,099,350 Bytes
- 4 Pages / 439.37 x 666.142 pts Page_size
- 100 Downloads / 148 Views
Abstract. Massive amounts of geo-social data is generated daily. In this paper, we propose UrbanHubble, a location-based predictive analytics tool that entails a broad range of state-of-the-art location prediction and recommendation algorithms. Besides, UrbanHubble consists of a visualization component that depicts the real-time complex interactions of users on a map, the evolution of friendships over time, and how friendship triggers mobility.
1
Introduction
The volume of data generated from human social interactions in Location-Based Social Networks (LBSN) is breathtaking. Such data encapsulates all visited locations and mimics the identity, behaviors, and affiliations of an individual or group. This has fueled enormous research interests to study location-based social interactions or group dynamics. One profound user behavior that has emerged during mobile social networking is the generation of check-in. Check-in is a phenomena whereby a person deliberately broadcasts her current location to a group of friends in an LBSN. Numerous location prediction techniques have been proposed. To the best of our knowledge, there is no platform that consists of a broad array of innovative state-of-the-art location prediction techniques such as [1,3,5,6,8,9]. The availability of such a framework would assist researchers to quickly compare and evaluate state-of-the-art prediction techniques. Thus, saving their time and allowing them to focus more on the new techniques they aspire to develop. Towards this end, we were motivated to create UrbanHubble1 , an innovative LBSN predictive analytics tool, which entails a broad spectrum of state-of-the-art LBSN prediction algorithms. Specifically, the algorithms include [1,3,5,6,8,9]. While Spot [4] also provides a platform to analyze LBSN, it consists of three algorithms. In contrast, we provide more algorithms than [4] and most importantly, our framework contains the most recent or relevant location prediction techniques. In addition to the aforementioned algorithms, UrbanHubble consists of a visualization component that shows the real-time complex social interactions of users on a map, the evolution of friendships over time, and how friendship triggers mobility or vice-versa. 1
http://dme.rwth-aachen.de/en/urbanhubble
c Springer International Publishing Switzerland 2015 A. Bifet et al. (Eds.): ECML PKDD 2015, Part III, LNAI 9286, pp. 329–332, 2015. DOI: 10.1007/978-3-319-23461-8 41
330
R. Assam et al.
While UrbanHubble is primarily intended for researchers, its visualization interface can also be used across an array of industries such as in location-based advertisement, where customers behaviors’ strongly depend on the location context, and advertisers are interested to efficiently identify patterns to hyper target such customers, or in urban planning or for traffic monitoring. Given the importance and enormous potentials of LBSN research, we believe these use-cases connote the demand for UrbanHubble.
2
UrbanHubble Tool
In this section, we provide a detailed description of the
Data Loading...