The Web is a central part of modern everyday life. Many people access it on a daily basis for a variety of reasons such as to retrieve news, watch videos, engage in social networks, buy goods in online shops or simply to procrastinate. Yet, we are still uncertain about how humans navigate the Web and the potential of factors influencing this process. To shed light on this topic, this thesis deals with modeling aspects of human navigation on the Web and the effects arising due to manipulations of this process. Mainly, this work provides a solid theoretical framework which allows to examine the potential effects of two different strategies aiming to guide visitors of a website. The framework builds upon the random surfer model, which is shown to be a sufficiently accurate model of human navigation on the Web in the first part of this work. In a next step, this thesis examines to which extent various click biases influence the typical whereabouts of the random surfer. Based on this analysis, this work demonstrates that exploiting common human cognitive biases exhibits a high potential of manipulating the frequencies with which the random surfer visits certain webpages. However, besides taking advantage of these biases, there exist further possibilities to steer users who navigate a website. Specifically, simply inserting new links to a webpage opens up new routes for visitors to explore a website. To investigate which of the two guiding strategies bears the higher potential, this work applies both of them to webgraphs of several websites and provides a detailed comparison of the emerging effects. The results presented in this thesis lead to actionable insights for website administrators and further broaden our understanding of how humans navigate the Web. Additionally, the presented model builds the foundation for further research in this field.

Social tagging systems enable users to collaboratively assign freely chosen keywords (i.e., tags) to resources (e.g., Web links). In order to support users in finding descrip- tive tags, tag recommendation algorithms have been proposed. One issue of current state-of-the-art tag recommendation algorithms is that they are often designed in a purely data-driven way and thus, lack a thorough understanding of the cognitive processes that play a role when people assign tags to resources. A prominent exam- ple is the activation equation of the cognitive architecture ACT-R, which formalizes activation processes in human memory to determine if a specific memory unit (e.g., a word or tag) will be needed in a specific context. It is the aim of this thesis to investigate if a cognitive-inspired approach, which models activation processes in human memory, can improve tag recommendations. For this, the relation between activation processes in human memory and usage practices of tags is studied, which reveals that (i) past usage frequency, (ii) recency, and (iii) semantic context cues are important factors when people reuse tags. Based on this, a cognitive-inspired tag recommendation approach termed BLL AC +MP r is developed based on the activation equation of ACT-R. An extensive evaluation using six real-world folksonomy datasets shows that BLL AC +MP r outperforms current state-of-the-art tag recommendation algorithms with respect to various evaluation metrics. Finally, BLL AC +MP r is utilized for hashtag recommendations in Twitter to demonstrate its generalizability in related areas of tag-based recommender systems. The findings of this thesis demonstrate that activation processes in human memory can be utilized to improve not only social tag recommendations but also hashtag recommendations. This opens up a number of possible research strands for future work, such as the design of cognitive-inspired resource recommender systems

In recent years, various recommendation algorithms have been proposed to support learners in technology-enhanced learning environments. Such algorithms have proven to be quite effective in big-data learning settings (massive open online courses), yet successful applications in other informal and formal learning settings are rare. Common challenges include data sparsity, the lack of sufficiently flexible learner and domain models, and the difficulty of including pedagogical goals into recommendation strategies. Computational models of human cognition and learning are, in principle, well positioned to help meet these challenges, yet the effectiveness of cognitive models in educational recommender systems remains poorly understood to this date. This thesis contributes to this strand of research by investigating i) two cognitive learner models (CbKST and SUSTAIN) for resource recommendations that qualify for sparse user data by following theory-driven top down approaches, and ii) two tag recommendation strategies based on models of human cognition (BLL and MINERVA2) that support the creation of learning content meta-data. The results of four online and offline experiments in different learning contexts indicate that a recommendation approach based on the CbKST, a well-founded structural model of knowledge representation, can improve the users' perceived learning experience in formal learning settings. In informal settings, SUSTAIN, a human category learning model, is shown to succeed in representing dynamic, interest based learning interactions and to improve Collaborative Filtering for resource recommendations. The investigation of the two proposed tag recommender strategies underlined their ability to generate accurate suggestions (BLL) and in collaborative settings, their potential to promote the development of shared vocabulary (MINERVA2). This thesis shows that the application of computational models of human cognition holds promise for the design of recommender mechanisms and, at the same time, for gaining a deeper understanding of interaction dynamics in virtual learning systems.