In this paper, we explore the issue of inconsistency handling in DatalogMTL, an extension of Datalog with metric temporal operators. Since facts are associated with time intervals, there are different manners to restore consistency when they contradict the rules, such as removing facts or modifying their time intervals. Our first contribution is the definition of relevant notions of conflicts (minimal explanations for inconsistency) and repairs (possible ways of restoring consistency) for this setting and the study of the properties of these notions and the associated inconsistency-tolerant semantics. Our second contribution is a data complexity analysis of the tasks of generating a single conflict / repair and query entailment under repair-based semantics.

We study the complexity of ontology-mediated querying when ontologies are formulated in the guarded fragment of first-order logic (GF). Our general aim is to classify the data complexity on the level of ontologies where query evaluation w.r.t. an ontology O is considered to be in PTime if all (unions of conjunctive) queries can be evaluated in PTime w.r.t. O and coNP-hard if at least one query is coNP-hard w.r.t. O. We identify several large and relevant fragments of GF that enjoy a dichotomy between PTime and coNP, some of them additionally admitting a form of counting. In fact, almost all ontologies in the BioPortal repository fall into these fragments or can easily be rewritten to do so. We then establish a variation of Ladner's Theorem on the existence of NP-intermediate problems and use this result to show that for other fragments, there is provably no such dichotomy. Again for other fragments (such as full GF), establishing a dichotomy implies the Feder-Vardi conjecture on the complexity of constraint satisfaction problems. We also link these results to Datalog-rewritability and study the decidability of whether a given ontology enjoys PTime query evaluation, presenting both positive and negative results.

We study temporal description logics (TDLs) based on the branching-time temporal logic CTL and the lightweight DL EL in the presence of rigid roles and restricted TBoxes. While TDLs designed in this way are known to be inherently nonelementary or even undecidable over general TBoxes, there is hope for a better computational behaviour over acyclic or empty TBoxes. We begin by showing that the basic DL ALC combined with CTL in the described way is indeed decidable, but still inherently nonelementary. As our main contribution, we identify several TDLs of elementary complexity, obtained by combining EL with CTL fragments that allow only restricted sets of temporal operators. We obtain upper complexity bounds ranging from PTime to coNExpTime and mostly tight lower bounds. This contrasts the fact that the respective ALC variants are already inherently nonelementary.

In ontology-based data access (OBDA), ontologies are used as an interface for querying instance data. Since in typical applications the size of the data is much larger than the size of the ontology and query, data complexity is the most important complexity measure. In this paper, we propose a new method for investigating data complexity in OBDA: instead of classifying whole logics according to their complexity, we aim at classifying each individual ontology within a given master language. Our results include a P/coNP-dichotomy theorem for ontologies of depth one in the description logic ALCFI, the equivalence of a P/coNP-dichotomy theorem for ALC/ALCI-ontologies of unrestricted depth to the famous dichotomy conjecture for CSPs by Feder and Vardi, and a non-P/coNP-dichotomy theorem for ALCF-ontologies.