PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit Twente, op gezag van de rector magnificus, prof. dr. Contents 1 Introduction 1 1.

Efficient algorithms have been developed for mining frequent patterns in traditional data where the content of each transaction is definitely known. There are many applications that deal with real data sets where the contents of the transactions are uncertain. Limited research work has been dedicated for mining frequent patterns from uncertain data. This is done by extending the state of art ho...

6 Chance constraints and the choice of uncertainty sets 15 6.1 Value at risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 6.2 Safe convex approximations for chance constraints . . . . . . . . . . . . . . . 17 6.3 Tightest convex bounds and conditional value at risk . . . . . . . . . . . . . 18 6.4 Analytic approximation using moment generating functions . . . . . ....

An uncertain data model is a system for representing incomplete or uncertain data. An uncertain database asserts that a database is in one of multiple alternative states (possible worlds), each being a standard database. A probability distribution can be assigned to the set of possible worlds. A detailed account of relational uncertain data models and their evolution, as well as related computa...

Lineage, also called Boolean provenance, event expression, or why-provenance, is a form of provenance or origin of the answer(s) to a query executed on a database. Lineage is expressed as a Boolean formula with variables assigned to the tuples in the database, where joint usage of the tuples (by the database join operation) is captured by Boolean conjunction (AND, ^) and alternative usage (proj...

The application of statistical methods to natural language processing has been remarkably successful over the past two decades. But, to deal with recent problems arising in this field, machine learning techniques must be generalized to deal with uncertain data, or datasets whose elements are distributions over sequences, such as weighted automata. This paper reviews a number of recent results r...

Uncertain data are abundant in numerous spatiotemporal applications providing location-based services. This chapter will first review these applications and explain why uncertainty is inherent, sometimes even intended, in their underlying data. Then, we will address two classical spatial queries: range search and nearest neighbor retrieval. We will see that the presence of uncertainty requires ...