Kostas Stefanidis

A survey on representation, composition and application of preferences in database systems

Preferences have been traditionally studied in philosophy, psychology, and economics and applied to decision making problems. Recently, they have attracted the attention of researchers in other fields, such as databases where they capture soft criteria for queries. Databases bring a whole fresh perspective to the study of preferences, both computational and representational. From a representational perspective, the central question is how we can effectively represent preferences and incorporate them in database querying. From a computational perspective, we can look at how we can efficiently process preferences in the context of database queries. Several approaches have been proposed but a systematic study of these works is missing. The purpose of this survey is to provide a framework for placing existing works in perspective and highlight critical open challenges to serve as a springboard for researchers in database systems. We organize our study around three axes: preference representation, preference composition, and preference query processing.

Adding Context to Preferences

To handle the overwhelming amount of information currently available, personalization systems allow users to specify the information that interests them through preferences. Most often, users have different preferences depending on context. In this paper, we introduce a model for expressing such contextual preferences. Context is modeled as a set of multidimensional attributes. We formulate the context resolution problem as the problem of (a) identifying those preferences that qualify to encompass the context state of a query and (b) selecting the most appropriate among them. We also propose an algorithm for context resolution that uses a data structure, called the profile tree, that indexes preferences based on their associated context. Finally, we evaluate our approach from two perspectives: usability and performance.

Fast group recommendations by applying user clustering

Recommendation systems have received significant attention, with most of the proposed methods focusing on personal recommendations. However, there are contexts in which the items to be suggested are not intended for a single user but for a group of people. For example, assume a group of friends or a family that is planning to watch a movie or visit a restaurant. In this paper, we propose an extensive model for group recommendations that exploits recommendations for items that similar users to the group members liked in the past. We do not exhaustively search for similar users in the whole user base, but we pre-partition users into clusters of similar ones and use the cluster members for recommendations. We efficiently aggregate the single user recommendations into group recommendations by leveraging the power of a top-k algorithm. We evaluate our approach in a real dataset of movie ratings.

Entity resolution in the web of data

This tutorial provides an overview of the key research results in the area of entity resolution that are relevant to addressing the new challenges in entity resolution posed by the Web of data, in which real world entities are described by interlinked data rather than documents. Since such descriptions are usually partial, overlapping and sometimes evolving, entity resolution emerges as a central problem both to increase dataset linking but also to search the Web of data for entities and their relations.

Fast contextual preference scoring of database tuples

To provide users with only relevant data from the huge amount of available information, personalization systems utilize preferences to allow users to express their interest on specific pieces of data. Most often, user preferences vary depending on the circumstances. For instance, when with friends, users may like to watch thrillers, whereas, when with their kids, they may prefer to watch cartoons. Contextual preference systems address this challenge by supporting preferences that depend on the values of contextual attributes such as the surrounding environment, time or location. In this paper, we address the problem of finding interesting data items based on contextual preferences that assign interest scores to pieces of data based on context. To this end, we propose a number of pre-processing steps. Instead of pre-computing scores for all data items under all potential context states, we exploit the hierarchical nature of context attributes to identify representative context states. Furthermore, we introduce a method for grouping preferences based on the similarity of the scores that they produce. This method uses a bitmap representation of preferences and scores with various levels of precision that lead to approximate rankings with different degrees of accuracy. We evaluate our approach using both real and synthetic data sets and present experimental results showing the quality of the scores attained using our methods.

PerK: personalized keyword search in relational databases through preferences

Keyword-based search in relational databases allows users to discover relevant information without knowing the database schema or using complicated queries. However, such searches may return an overwhelming number of results, often loosely related to the user intent. In this paper, we propose personalizing keyword database search by utilizing user preferences. Query results are ranked based on both their relevance to the query and their preference degree for the user. To further increase the quality of results, we consider two new metrics that evaluate the goodness of the result as a set, namely coverage of many user interests and content diversity. We present an algorithm for processing preference queries that uses the preferential order between keywords to direct the joining of relevant tuples from multiple relations. We then show how to reduce the complexity of this algorithm by sharing computational steps. Finally, we report evaluation results of the efficiency and effectiveness of our approach.

Preference-aware publish/subscribe delivery with diversity

In publish/subscribe systems, users describe their interests via subscriptions and are notified whenever new interesting events become available. Typically, in such systems, all subscriptions are considered equally important. However, due to the abundance of information, users may receive overwhelming amounts of events. In this paper, we propose using a ranking mechanism based on user preferences, so that only top-ranked events are delivered to each user. Since many times top-ranked events are similar to each other, we also propose increasing the diversity of delivered events. Furthermore, we examine a number of different delivering policies for forwarding ranked events to users, namely a periodic, a sliding-window and a history-based one. We have fully implemented our approach in SIENA, a popular publish/subscribe middleware system, and report experimental results of its deployment.

Nearest keyword search in XML documents

This paper studies the nearest keyword (NK) problem on XML documents. In general, the dataset is a tree where each node is associated with one or more keywords. Given a node q and a keyword w, an NK query returns the node that is nearest to q among all the nodes associated with w. NK search is not only useful as a stand-alone operator but also as a building brick for important tasks such as XPath query evaluation and keyword search. We present an indexing scheme that answers NK queries efficiently, in terms of both practical and worst-case performance. The query cost is provably logarithmic to the number of nodes carrying the query keyword. The proposed scheme occupies space linear to the dataset size, and can be constructed by a fast algorithm. Extensive experimentation confirms our theoretical findings, and demonstrates the effectiveness of NK retrieval as a primitive operator in XML databases.

Managing Contextual Preferences

To handle the overwhelming amount of information currently available, personalization systems allow users to specify through preferences which pieces of data interest them. Most often, users have different preferences depending on context. In this paper, we introduce a model for expressing such contextual preferences. Context is modeled using a set of hierarchical attributes, thus allowing context specification at various levels of detail. We formulate the context resolution problem as the problem of selecting appropriate preferences based on context for personalizing a query. We also propose algorithms for context resolution based on data structures that index preferences by exploiting the hierarchical nature of the context attributes. Finally, we evaluate our approach from two perspectives: usability and performance. Usability evaluates the overheads imposed on users for specifying context-dependent preferences, as well as their satisfaction from the quality of the results. Our performance results focus on the context resolution using the proposed indexes.

Modeling and storing context-aware preferences

Today, the overwhelming volume of information that is available to an increasingly wider spectrum of users creates the need for personalization. In this paper, we consider a database system that supports context-aware preference queries, that is, preference queries whose result depends on the context at the time of their submission. We use data cubes to store the associations between context-dependent preferences and database relations and OLAP techniques for processing context-aware queries, thus allowing the manipulation of the captured context data at different levels of abstractions. To improve query performance, we use an auxiliary data structure, called context tree, which indexes the results of previously computed preference-aware queries based on their associated context. We show how these cached results can be used to process both exact and approximate context-aware preference queries.