Pierre Senellart

Querying and updating probabilistic information in XML

We present in this paper a new model for representing probabilistic information in a semi-structured (XML) database, based on the use of probabilistic event variables. This work is motivated by the need of keeping track of both confidence and lineage of the information stored in a semi-structured warehouse. For instance, the modules of a (Hidden Web) content warehouse may derive information concerning the semantics of discovered Web services that is by nature not certain. Our model, namely the fuzzy tree model, supports both querying (tree pattern queries with join) and updating (transactions containing an arbitrary set of insertions and deletions) over probabilistic tree data. We highlight its expressive power and discuss implementation issues.

On the complexity of managing probabilistic XML data

In [3], we introduced a framework for querying and updating probabilistic information over unordered labeled trees, the probabilistic tree model. The data model is based on trees where nodes are annotated with conjunctions of probabilistic event variables. We briefly described an implementation and scenarios of usage. We develop here a mathematical foundation for this model. In particular, we present complexity results. We identify a very large class of queries for which simple variations of querying and updating algorithms from [3] compute the correct answer. A main contribution is a full complexity analysis of queries and updates. We also exhibit a decision procedure for the equivalence of probabilistic trees and prove it is in co-RP. Furthermore, we study the issue of removing less probable possible worlds, and that of validating a probabilistic tree against a DTD. We show that these two problems are intractable in the most general case.

On the complexity of deriving schema mappings from database instances

We introduce a theoretical framework for discovering relationships between two database instances over distinct and unknown schemata. This framework is grounded in the context of data exchange. We formalize the problem of understanding the relationship between two instances as that of obtaining a schema mapping so that a minimum repair of this mapping provides a perfect description of the target instance given the source instance. We show that this definition yields intuitive results when applied on database instances derived from each other by basic operations. We study the complexity of decision problems related to this optimality notion in the context of different logical languages and show that, even in very restricted cases, the problem is of high complexity.

Web page rank prediction with markov models

In this paper we propose a method for predicting the ranking position of a Web page. Assuming a set of successive past top-k rankings, we study the evolution of Web pages in terms of ranking trend sequences used for Markov Models training, which are in turn used to predict future rankings. The predictions are highly accurate for all experimental setups and similarity measures.

Web Data Management: Ontologies, RDF, and OWL

INTRODUCTION The vision of the Semantic Web is that of a world-wide distributed architecture where data and services easily interoperate. This vision is not yet a reality in the Web of today, in which given a particular need, it is difficult to find a resource that is appropriate to it. Also, given a relevant resource, it is not easy to understand what it provides and how to use it. To solve such limitations, facilitate interoperability, and thereby enable the Semantic Web vision, the key idea is to also publish semantics descriptions of Web resources. These descriptions rely on semantic annotations , typically on logical assertions that relate resources to some terms in predefined ontologies . This is the topic of the chapter. An ontology is a formal description providing human users a shared understanding of a given domain. The ontologies we consider here can also be interpreted and processed by machines thanks to a logical semantics that enables reasoning. Ontologies provide the basis for sharing knowledge, and, as such, they are very useful for a number of reasons: Organizing data. It is very easy to get lost in large collections of documents. An ontology is a natural means of “organizing” (structuring) it and thereby facilitates browsing through it to find interesting information. It provides an organization that is flexible, and that naturally structures the information in multidimensional ways. For instance, an ontology may allow browsing through the courses offered by a university by topic or department, by quarter or time, by level, and so forth.

Putting into Practice: Large-Scale Data Management with Hadoop

The chapter proposes an introduction to HADOOP and suggests some exercises to initiate a practical experience of the system. The following assumes that you dispose of a Unixlike system (Mac OS X works just fine; Windows requires Cygwin). HADOOP can run in a pseudo-distributed mode which does not require a cluster infrastructure for testing the software, and the main part of our instructions considers this mode. Switching to a real cluster requires some additional configurations that are introduced at the end of the chapter. Since HADOOP is a relatively young system that steadily evolves, looking at the on-line, up-to-date documentation is of course recommended if you are to use it on a real basis. We illustrate HADOOP, MAPREDUCE and PIG manipulations on the DBLP data set, which can be retrieved from the following URL:

Web Data Management: An Introduction to Distributed Systems

For personal use only, not for distribution.

Putting into Practice: Wrappers and Data Extraction with XSLT

For personal use only, not for distribution.

Distributed Computing with MapReduce and Pig

For personal use only, not for distribution.

Web Data Management: Distributed Access Structures

For personal use only, not for distribution.