Disjunctive Logic Programming (DLP) is an advanced formalism for knowledge representation and reasoning. DLP is very expressive, and allows to represent even problems of high complexity (every problem in the complexity class sigma^P_2 = NP^NP) in a simple and declarative way. The recent implementation of a couple of efficient DLP systems (namely, DLV and GnT/SModels) increased the interest in D LP, and allowed the implementation of several knowledge-based applications also in the are of planning, diagnosis, product configuration, etc. The highly expressive knowledge representation language supported by DLP systems, makes it evident that these systems have an high potential for exploitation in the area of Knowledge Management (KM), where one has to represent and make reasoning on complex knowledge. In particular, a main issue in Knowledge Discovery (a fundamental process in KM) is the specification of ontologies for modeling the application domain and their exploitation to derive new knowledge. To this end, the advanced knowledge modeling features (like, e.g., inference rules, integrity constraints, weak constraints, and non-monotonic inference mechanisms) provided by DLP systems can significantly enhance the knowledge modeling ability and the inference power of existing ontologies specification languages. Nevertheless, a full exploitation of DLP systems for KM applications requires them to be suitably enhanced for this purpose (e.g., DLP systems do not permit to naturally deal with statistical data, and do not provide support for objects and inheritance). This is precisely the goal of this project proposal. We will first single out the main limitations of DLP systems for ontologies specification. We will then propose DLP extensions that overcome these limitations, and design suitable algorithms and data-structures to efficiently implement the proposed extensions in the DLP system DLV making it suitable for the development of KM applications.