I completed this thesis in April 2017 as the final work of my M.Sc. degree at the University of Freiburg. Please do not hesitate to contact me if you have any questions or ideas.

When tourists visit new cities or countries, they might be interested in visiting local attractions. To achieve this, they can approach a local tourist information center from which they typically obtain a list of those sites. The user can then determine, based on the places' descriptions, which attractions are relevant and can search for a way to reach those places. This task is typically very time-consuming, especially the latter, since the user is in an unknown city and needs to understand how local public transportation functions.

Recommender systems (RS) are tools which aim to assist the user in a decision process and seek to alleviate information overload. Whereas there are several RS focused on touristic recommendations, not many of them integrate the information with the public transportation network. When it comes to developing countries, the situation is typically worse. In these countries there exist several challenges that prevent tools to produce reliable trip plans, such as traffic jams, accidents, construction work and unorganized transportation networks.

The goal of this Thesis is to combine two different paradigms, i.e. that of Recommender Systems and Route Planning, to create a tool that makes it possible to produce trip plans even when delays or other events are considered. To demonstrate the efficacy of our tool, we make use of real data from the city of Izmir. In addition to this, we carry extensive evaluation which shows that in terms of performance our tool is able to satisfy the initial requirements.

• to my supervisors Victor Anthony Arrascue Ayala and Dr. Marco Muniz for always supporting me and encouraging me.
• to Prof. Dr. Georg Lausen and Prof. Dr. Peter Fischer for the kind opportunity.