Ranking queries are widely used in data exploration, data analysis and decision making scenarios. While most of the currently proposed ranking techniques focus on deterministic data, several emerging applications involve data that are imprecise or uncertain. Ranking uncertain data raises new challenges in query semantics and processing, making conventional methods inapplicable. Furthermore, the interplay between ranking and uncertainty models introduces new dimensions for ordering query results that do not exist in the traditional settings. This lecture describes new formulations and processing techniques for ranking queries on uncertain data. The formulations are based on marriage of traditional ranking semantics with possible worlds semantics under widely-adopted uncertainty models. In particular, we focus on discussing the impact of tuple-level and attribute-level uncertainty on the semantics and processing techniques of ranking queries. Under the tuple-level uncertainty model, we describe new processing techniques leveraging the capabilities of relational database systems to recognize and handle data uncertainty in score-based ranking. Under the attribute-level uncertainty model, we describe new probabilistic ranking models and a set of query evaluation algorithms, including sampling-based techniques. We also discuss supporting rank join queries on uncertain data, and we show how to extend current rank join methods to handle uncertainty in scoring attributes. Table of Contents: Introduction / Uncertainty Models / Query Semantics / Methodologies / Uncertain Rank Join / Conclusion

Ranking queries are widely used in data exploration, data analysis and decision making scenarios. While most of the currently proposed ranking techniques focus on deterministic data, several emerging applications involve data that are imprecise or uncertain. Ranking uncertain data raises new challenges in query semantics and processing, making conventional methods inapplicable. Furthermore, the interplay between ranking and uncertainty models introduces new dimensions for ordering query results that do not exist in the traditional settings. This lecture describes new formulations and processing techniques for ranking queries on uncertain data. The formulations are based on marriage of traditional ranking semantics with possible worlds semantics under widely-adopted uncertainty models. In particular, we focus on discussing the impact of tuple-level and attribute-level uncertainty on the semantics and processing techniques of ranking queries. Under the tuple-level uncertainty model, we describe new processing techniques leveraging the capabilities of relational database systems to recognize and handle data uncertainty in score-based ranking. Under the attribute-level uncertainty model, we describe new probabilistic ranking models and a set of query evaluation algorithms, including sampling-based techniques. We also discuss supporting rank join queries on uncertain data, and we show how to extend current rank join methods to handle uncertainty in scoring attributes. Table of Contents: Introduction / Uncertainty Models / Query Semantics / Methodologies / Uncertain Rank Join / Conclusion

This book constitutes the thoroughly refereed proceedings of the 16th East-European Conference on Advances in Databases and Information Systems (ADBIS 2012), held in Poznan, Poland, in September 2012. The 32 revised full papers presented were carefully selected and reviewed from 122 submissions. The papers cover a wide spectrum of issues concerning the area of database and information systems, including database theory, database architectures, query languages, query processing and optimization, design methods, data integration, view selection, nearest-neighbor searching, analytical query processing, indexing and caching, concurrency control, distributed systems, data mining, data streams, ontology engineering, social networks, multi-agent systems, business process modeling, knowledge management, and application-oriented topics like RFID, XML, and data on the Web.

Probabilistic databases are databases where the value of some attributes or the presence of some records are uncertain and known only with some probability. Applications in many areas such as information extraction, RFID and scientific data management, data cleaning, data integration, and financial risk assessment produce large volumes of uncertain data, which are best modeled and processed by a probabilistic database. This book presents the state of the art in representation formalisms and query processing techniques for probabilistic data. It starts by discussing the basic principles for representing large probabilistic databases, by decomposing them into tuple-independent tables, block-independent-disjoint tables, or U-databases. Then it discusses two classes of techniques for query evaluation on probabilistic databases. In extensional query evaluation, the entire probabilistic inference can be pushed into the database engine and, therefore, processed as effectively as the evaluation of standard SQL queries. The relational queries that can be evaluated this way are called safe queries. In intensional query evaluation, the probabilistic inference is performed over a propositional formula called lineage expression: every relational query can be evaluated this way, but the data complexity dramatically depends on the query being evaluated, and can be #P-hard. The book also discusses some advanced topics in probabilistic data management such as top-k query processing, sequential probabilistic databases, indexing and materialized views, and Monte Carlo databases. Table of Contents: Overview / Data and Query Model / The Query Evaluation Problem / Extensional Query Evaluation / Intensional Query Evaluation / Advanced Techniques

State-of-the-art database systems manage and process a variety of complex objects, including strings and trees. For such objects equality comparisons are often not meaningful and must be replaced by similarity comparisons. This book describes the concepts and techniques to incorporate similarity into database systems. We start out by discussing the properties of strings and trees, and identify the edit distance as the de facto standard for comparing complex objects. Since the edit distance is computationally expensive, token-based distances have been introduced to speed up edit distance computations. The basic idea is to decompose complex objects into sets of tokens that can be compared efficiently. Token-based distances are used to compute an approximation of the edit distance and prune expensive edit distance calculations. A key observation when computing similarity joins is that many of the object pairs, for which the similarity is computed, are very different from each other. Filters exploit this property to improve the performance of similarity joins. A filter preprocesses the input data sets and produces a set of candidate pairs. The distance function is evaluated on the candidate pairs only. We describe the essential query processing techniques for filters based on lower and upper bounds. For token equality joins we describe prefix, size, positional and partitioning filters, which can be used to avoid the computation of small intersections that are not needed since the similarity would be too low.

The chase has long been used as a central tool to analyze dependencies and their effect on queries. It has been applied to different relevant problems in database theory such as query optimization, query containment and equivalence, dependency implication, and database schema design. Recent years have seen a renewed interest in the chase as an important tool in several database applications, such as data exchange and integration, query answering in incomplete data, and many others. It is well known that the chase algorithm might be non-terminating and thus, in order for it to find practical applicability, it is crucial to identify cases where its termination is guaranteed. Another important aspect to consider when dealing with the chase is that it can introduce null values into the database, thereby leading to incomplete data. Thus, in several scenarios where the chase is used the problem of dealing with data dependencies and incomplete data arises. This book discusses fundamental issues concerning data dependencies and incomplete data with a particular focus on the chase and its applications in different database areas. We report recent results about the crucial issue of identifying conditions that guarantee the chase termination. Different database applications where the chase is a central tool are discussed with particular attention devoted to query answering in the presence of data dependencies and database schema design. Table of Contents: Introduction / Relational Databases / Incomplete Databases / The Chase Algorithm / Chase Termination / Data Dependencies and Normal Forms / Universal Repairs / Chase and Database Applications

Incomplete data is part of life and almost all areas of scientific studies. Users tend to skip certain fields when they fill out online forms; participants choose to ignore sensitive questions on surveys; sensors fail, resulting in the loss of certain readings; publicly viewable satellite map services have missing data in many mobile applications; and in privacy-preserving applications, the data is incomplete deliberately in order to preserve the sensitivity of some attribute values. Query processing is a fundamental problem in computer science, and is useful in a variety of applications. In this book, we mostly focus on the query processing over incomplete databases, which involves finding a set of qualified objects from a specified incomplete dataset in order to support a wide spectrum of real-life applications. We first elaborate the three general kinds of methods of handling incomplete data, including (i) discarding the data with missing values, (ii) imputation for the missing values, and (iii) just depending on the observed data values. For the third method type, we introduce the semantics of k-nearest neighbor (kNN) search, skyline query, and top-k dominating query on incomplete data, respectively. In terms of the three representative queries over incomplete data, we investigate some advanced techniques to process incomplete data queries, including indexing, pruning as well as crowdsourcing techniques.

As an alternative to traditional client-server systems, Peer-to-Peer (P2P) systems provide major advantages in terms of scalability, autonomy and dynamic behavior of peers, and decentralization of control. Thus, they are well suited for large-scale data sharing in distributed environments. Most of the existing P2P approaches for data sharing rely on either structured networks (e.g., DHTs) for efficient indexing, or unstructured networks for ease of deployment, or some combination. However, these approaches have some limitations, such as lack of freedom for data placement in DHTs, and high latency and high network traffic in unstructured networks. To address these limitations, gossip protocols which are easy to deploy and scale well, can be exploited. In this book, we will give an overview of these different P2P techniques and architectures, discuss their trade-offs, and illustrate their use for decentralizing several large-scale data sharing applications. Table of Contents: P2P Overlays, Query Routing, and Gossiping / Content Distribution in P2P Systems / Recommendation Systems / Top-k Query Processing in P2P Systems

Data usually comes in a plethora of formats and dimensions, rendering the exploration and information extraction processes challenging. Thus, being able to perform exploratory analyses in the data with the intent of having an immediate glimpse on some of the data properties is becoming crucial. Exploratory analyses should be simple enough to avoid complicate declarative languages (such as SQL) and mechanisms, and at the same time retain the flexibility and expressiveness of such languages. Recently, we have witnessed a rediscovery of the so-called example-based methods, in which the user, or the analyst, circumvents query languages by using examples as input. An example is a representative of the intended results, or in other words, an item from the result set. Example-based methods exploit inherent characteristics of the data to infer the results that the user has in mind, but may not able to (easily) express. They can be useful in cases where a user is looking for information in an unfamiliar dataset, when the task is particularly challenging like finding duplicate items, or simply when they are exploring the data. In this book, we present an excursus over the main methods for exploratory analysis, with a particular focus on example-based methods. We show how that different data types require different techniques, and present algorithms that are specifically designed for relational, textual, and graph data. The book presents also the challenges and the new frontiers of machine learning in online settings which recently attracted the attention of the database community. The lecture concludes with a vision for further research and applications in this area.

Due to the Internet Revolution, human conversational data -- in written forms -- are accumulating at a phenomenal rate. At the same time, improvements in speech technology enable many spoken conversations to be transcribed. Individuals and organizations engage in email exchanges, face-to-face meetings, blogging, texting and other social media activities. The advances in natural language processing provide ample opportunities for these "informal documents" to be analyzed and mined, thus creating numerous new and valuable applications. This book presents a set of computational methods to extract information from conversational data, and to provide natural language summaries of the data. The book begins with an overview of basic concepts, such as the differences between extractive and abstractive summaries, and metrics for evaluating the effectiveness of summarization and various extraction tasks. It also describes some of the benchmark corpora used in the literature. The book introduces extraction and mining methods for performing subjectivity and sentiment detection, topic segmentation and modeling, and the extraction of conversational structure. It also describes frameworks for conducting dialogue act recognition, decision and action item detection, and extraction of thread structure. There is a specific focus on performing all these tasks on conversational data, such as meeting transcripts (which exemplify synchronous conversations) and emails (which exemplify asynchronous conversations). Very recent approaches to deal with blogs, discussion forums and microblogs (e.g., Twitter) are also discussed. The second half of this book focuses on natural language summarization of conversational data. It gives an overview of several extractive and abstractive summarizers developed for emails, meetings, blogs and forums. It also describes attempts for building multi-modal summarizers. Last but not least, the book concludes with thoughts on topics for further development. Table of Contents: Introduction / Background: Corpora and Evaluation Methods / Mining Text Conversations / Summarizing Text Conversations / Conclusions / Final Thoughts