This paper contributes to the theory of the modal $\mu$-calculus by proving some model-theoretic results. More in particular, we discuss a number of semantic properties pertaining to formulas of the modal $\mu$-calculus. For each of these properties we provide a corresponding syntactic fragment, in the sense that a $\mu$-formula $\xi$ has the given property iff it is equivalent to a formula $\xi'$ in the corresponding fragment. Since this formula $\xi'$ will always be effectively obtainable from $\xi$, as a corollary, for each of the properties under discussion, we prove that it is decidable in elementary time whether a given $\mu$-calculus formula has the property or not. The properties that we study all concern the way in which the meaning of a formula $\xi$ in a model depends on the meaning of a single, fixed proposition letter $p$. For example, consider a formula $\xi$ which is monotone in $p$; such a formula a formula $\xi$ is called continuous (respectively, fully additive), if in addition it satisfies the property that, if $\xi$ is true at a state $s$ then there is a finite set (respectively, a singleton set) $U$ such that $\xi$ remains true at $s$ if we restrict the interpretation of $p$ to the set $U$. Each of the properties that we consider is, in a similar way, associated with one of the following special kinds of subset of a tree model: singletons, finite sets, finitely branching subtrees, noetherian subtrees (i.e., without infinite paths), and branches. Our proofs for these characterization results will be automata-theoretic in nature; we will see that the effectively defined maps on formulas are in fact induced by rather simple transformations on modal automata. Thus our results can also be seen as a contribution to the model theory of modal automata.

Areas in which graph databases are applied - such as the semantic web, social networks and scientific databases - are prone to inconsistency, mainly due to interoperability issues. This raises the need for understanding query answering over inconsistent graph databases in a framework that is simple yet general enough to accommodate many of its applications. We follow the well-known approach of consistent query answering (CQA), and study the data complexity of CQA over graph databases for regular path queries (RPQs) and regular path constraints (RPCs), which are frequently used. We concentrate on subset, superset and symmetric difference repairs. Without further restrictions, CQA is undecidable for the semantics based on superset and symmetric difference repairs, and Pi_2^P-complete for subset repairs. However, we provide several tractable restrictions on both RPCs and the structure of graph databases that lead to decidability, and even tractability of CQA. We also compare our results with those obtained for CQA in the context of relational databases.

A database may for various reasons become inconsistent with respect to a given set of integrity constraints. In the late 1990s, the formal approach of consistent query answering was proposed in order to query such databases. Since then, a lot of efforts have been spent to classify the complexity of consistent query answering under various classes of constraints. It is known that for the most common constraints and queries, the problem is in coNP and might be coNP-hard, yet several relevant tractable classes have been identified. Additionally, the results that emerged suggested that given a set of key constraints and a conjunctive query, the problem of consistent query answering is either in PTime or is coNP-complete. However, despite all the work, as of today this dichotomy remains a conjecture. The main contribution of this article is to explain why it appears so difficult to obtain a dichotomy result in the setting of consistent query answering. Namely, we prove that such a dichotomy with respect to common classes of constraints and queries is harder to achieve than a dichotomy for the constraint satisfaction problem, which is a famous open problem since the 1990s.

A database may for various reasons become inconsistent with respect to a given set of integrity constraints. To overcome the problem, a formal approach to querying such inconsistent databases has been proposed and since then, a lot of efforts have been spent to classify the complexity of consistent query answering under various classes of constraints. It is known that for the most common constraints and queries, the problem is in CONP and might be CONP-hard, yet several relevant tractable classes have been identified. Additionally, the results that emerged suggested that given a set of key constraints and a conjunctive query, the problem of consistent query answering is either in PTIME or is CONP-complete. However, despite all the work, as of today this dichotomy remains a conjecture. The main contribution of this paper is to explain why it appears so difficult to obtain a dichotomy result in the setting of consistent query answering. Namely, we prove that such a dichotomy w.r.t. common classes of constraints and queries, is harder to achieve than a dichotomy for the constraint satisfaction problem, which is a famous open problem since the 1990s.

The framework of database repairs is a principled approach to managing inconsistency in databases. In particular, the consistent answers of a query on an inconsistent database provide sound semantics and the guarantee that the values obtained are those returned by the query on every repair of the given inconsistent database. In this paper, we carry out a systematic investigation of the data complexity of the consistent answers of conjunctive queries for set-based repairs and with respect to classes of constraints that, in recent years, have been extensively studied in the context of data exchange and data integration. Our results, which range from polynomial-time computability to undecidability, complement or improve on earlier work, and provide a fairly comprehensive picture of the data complexity of consistent query answering. We also address the problem of finding a "representative" or "useful" repair of an inconsistent database. To this effect, we introduce the notion of a universal repair, as well as relaxations of it, and then apply it to the investigation of the data complexity of consistent query answering.

Building on the fact that descriptive frames are coalgebras for the Vietoris functor on the category of Stone spaces, we introduce and study the concept of a Vietoris bisimulation between two descriptive modal models, together with the associated notion of bisimilarity. We prove that our notion of bisimilarity, which is defined in terms of relation lifting, coincides with Kripke bisimilarity (with respect to the underlying Kripke models), with behavioural equivalence, and with modal equivalence, but not with Aczel-Mendler bisimilarity. As a corollary, we obtain that the Vietoris functor does not preserve weak pullbacks. Comparing Vietoris bisimulations between descriptive models to Kripke bisimulations on the underlying Kripke models, we prove that the closure of such a Kripke bisimulation is a Vietoris bisimulation. As a corollary, we show that the collection of Vietoris bisimulations between two descriptive models forms a complete lattice. Finally, we provide a game-theoretic characterization of Vietoris bisimilarity.