Present-day routing services are able to provide travel directions for users of all modes of transport. Most of them are focusing on functional journeys (i.e. journeys linking given origin and destination with minimum cost) and pay less attention to recreational trips, in particular leisure walks in an urban context. These walks have predefined time or distance and as their purpose is the process of walking itself, the attractiveness of chosen paths starts playing an important role in route selection. Conventional map data that are informing routing algorithms cannot be used for extracting street attractiveness as they do not contain a subjective component, or in other words, do not tell whether or not people enjoy their presence at a particular place. Recent research demonstrates that the crowd-sourced data available from the photo- sharing websites have a potential for being a good source of this measure, thus becoming a base for a routing system that suggests attractive leisure walks. This PhD research looks at existing projects, which aim to utilize user-generated photographic data for journey planning, and suggests new techniques that make the estimation of street attractiveness based on this source more reliable. First, we determine the artifacts in photo- graphic datasets that may negatively impact the resulting attractiveness scores. Based on the findings, we suggest filtering methods that improve the compliance of the spatial distributions of photographs with the chosen purpose. Second, we discuss several approaches of assigning attractiveness scores to street segments and make conclusions about their differences. Finally, we experiment with the routing itself and develop a prototype system that suggests leisure walks through attractive streets in an urban area. The experiments we perform cover Central London and involve four photographic sources: Flickr, Geograph, Panoramio and Picasa. A Visual analytic (VA) approach is used throughout the work to glean new insights. Being able to combine computation and the analytical capabilities of the human brain, this research method has proven to work well with complex data structures in a variety of tasks. The thesis contributes to VA as an example of what can be achieved by means of the visual exploration of data.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:675987 |
| Date | January 2014 |
| Creators | Kachkaev, A. |
| Publisher | City University London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://openaccess.city.ac.uk/12460/ |
Page generated in 0.0017 seconds