Origin-destination matrices help understand the movement of people within cities. This work is built upon the premise that stakeholders, e.g. decision makers, need to analyze mobility flows from spatio-temporal perspectives that are appropriate to their context of analysis. The data retrieved from sensors and Intelligent Transportation Systems are useful for this purpose due to their lower acquisition costs and fine granularity, although it is complex to use such data in an integrated way, as they might have heterogeneous representations of spatio-temporal attributes and granularities. Most of the related works on the analysis of OD flows consider matrices with a fixed spatio-temporal aggregation level, and do not explore the intrinsic issue of data heterogeneity. Herein we report our findings on building the semantic foundation of knowledge-assisted visualization tools for analyzing OD matrices from multiple stakeholder levels. We propose a set of ontology design patterns for modeling the semantics of OD data, and the relations between the spatio-temporal constructs that stakeholders ought to choose when visualizing urban mobility flows. Our approach aims to be reusable by researchers and practitioners. We describe a practical implementation using estimated flows from smart card data from Porto, Portugal.

Markov models have been widely used to represent and analyze user Web navigation data. In previous work, we have proposed a method to dynamically extend the order of a Markov chain model and a complimentary method for assessing the predictive power of such a variable-length Markov chain. Herein, we review these two methods and propose a novel method for measuring the ability of a variable-length Markov model to summarize user Web navigation sessions up to a given length. Although the summarization ability of a model is important to enable the identification of user navigation patterns, the ability to make predictions is important in order to foresee the next link choice of a user after following a given trail so as, for example, to personalize a Web site. We present an extensive experimental evaluation providing strong evidence that prediction accuracy increases linearly with summarization ability