|M.Sc Student||Noa Binski|
|Subject||Wikipedia as a source of survay landmarks for enriching|
|Department||Department of Civil and Environmental Engineering||Supervisor||Dr. Dalyot Sagi|
|Full Thesis text|
Car navigation systems provide with an easy and simple solution to the basic concept of reaching a destination. Although they usually achieve this goal, they are mostly using a limited and poor sequence of instructions that do not consider the basic human nature of using landmarks in route instructions.
This research addresses the concept of enriching navigation route instructions by adding supplementary route information in the form of landmarks. Research show that route instructions using landmarks are much easy to understand and follow, when compared to a limited set of commands that do not contribute to gaining spatial understanding of the environment. Furthermore, landmarks help in creating a mental map, which has an important role in navigation and orientation of one in the surrounding environment.
Landmarks are available in authoritative geographic resources, as well as external repositories. An alternative possibility to retrieving landmark information is by using a voluntarily source of information, user-generated one. Such source must be easy to access, available, and have a very large and updated volume of relevant landmark information. Wikipedia, which is the world’s largest free encyclopedia, which also includes information on many geospatial entities, including landmarks, was investigated and used for the task in this research.
This research developed a methodology and algorithms for extracting valuable landmark information form Wikipedia, which not only relies on the geo-tagged data of entries, but also includes a filtering and classification procedure of available landmarks. This is coupled with a ranking algorithm based on the entries’ categories and attributes. These are aimed at retrieving the most relevant and valuable survey landmark information required for the enrichment of navigation routes.
Methodology and algorithms were tested and analyzed on different routes in different cities: New York, London, Berlin and Tel Aviv. An investigation of the outcome of these experiments proved the developed algorithms were robust for retrieving the landmarks that should contribute the most for the goal, enriching all routes with valuable environmental information that should help drivers to better understand and follow the car navigation system instructions, thus creating a mental map.