This article tackles the problem of efficient construction of iCPI trees, frequently used in co-location pattern discovery in spatial databases. It discusses the methods for parallelization of... Show More

This article tackles the problem of efficient construction of iCPI trees, frequently used in co-location pattern discovery in spatial databases. It discusses the methods for parallelization of iCPI-tree construction and plane-sweep algorithms used in state-of-the-art algorithms for co-location pattern mining. The main contribution of this paper is threefold: (1) a general algorithm for parallel iCPI-tree construction is presented, (2) two variants of parallel plane-sweep algorithm (which can be used in conjunction with the aforementioned iCPI-tree construction algorithm) are introduced and (3) all three algorithms are implemented on CUDA GPU platform and their performance is tested against an efficient multithreaded parallel implementation of iCPI-tree construction on CPU. Experiments prove that our solutions allow for large speedups over CPU version of the algorithm. This paper is an extension of the conference paper (Andrzejewski & Boinski, 2014).

Empty answers are a major problem by processing pattern matching queries in graph databases. Especially, there can be multiple reasons why a query failed. To support users in such situations... Show More

Empty answers are a major problem by processing pattern matching queries in graph databases. Especially, there can be multiple reasons why a query failed. To support users in such situations, differential queries can be used that deliver missing parts of a graph query. Multiple heuristics are proposed for differential queries, which reduce the search space. Although they are successful in increasing the performance, they can discard query subgraphs relevant to a user. To address this issue, the authors extend the concept of differential queries and introduce top-k differential queries that calculate the ranking based on users' preferences and significantly support the users' understanding of query database management systems. A user assigns relevance weights to elements of a graph query that steer the search and are used for the ranking. In this paper the authors propose different strategies for selection of relevance weights and their propagation. As a result, the search is modelled along the most relevant paths. The authors evaluate their solution and both strategies on the DBpedia data graph.

This article introduces how temporal data can be maintained and processed by utilizing Column-oriented NoSQL databases (CoNoSQLDBs). Although each column in a CoNoSQLDB can store multiple data... Show More

This article introduces how temporal data can be maintained and processed by utilizing Column-oriented NoSQL databases (CoNoSQLDBs). Although each column in a CoNoSQLDB can store multiple data versions with their corresponded timestamps, its implicit temporal interval representation can cause wrong or misleading results during temporal query processing. In consequence, the original table representation supported by CoNoSQLDBs is not suitable for storing temporal data. To maintain the temporal data in the CoNoSQLDB tables, two alternative table representations can be adopted, namely, explicit history representation (EHR) and tuple time-stamping representation (TTR) in which each tuple (data version) has an explicit temporal interval. For processing TTR, the temporal relational algebra is extended to TTRO operator model with minor modifications. For processing EHR, a novel temporal operator model called CTO is proposed. Both TTRO and CTO contain eight temporal data operators, namely, Union, Difference, Intersection, Project, Filter, Cartesian product, Theta-Join and Group by with a set of aggregation functions, such as SUM, AVG, MAX and etc. Moreover, the authors implement each temporal operator by utilizing MapReduce framework to indicate which temporal operator model is more suitable for temporal data processing in the context of CoNoSQLDBs.

Constraints ubiquitously exist in many real-life applications for entity resolution. However, it is always challenging to effectively specify and use such constraints for performing ER tasks. In... Show More

Constraints ubiquitously exist in many real-life applications for entity resolution. However, it is always challenging to effectively specify and use such constraints for performing ER tasks. In particular, not every constraint is equally robust. Adding weights to express the “confidence” on constraints thus becomes a natural choice. In this paper, the authors study entity resolution (ER), the problem of determining which records in one or more databases refer to the same entities, in the presence of weighted constraints. They propose a unified framework that allows us to associate a weight for each constraint, capturing the confidence for its robustness in an ER model. The authors develop an approach to learn weighted constraints based on domain knowledge, and investigate how effectively and efficiently weighted constraints can be used for generating an ER clustering and for determining a propagation order across multiple entity types. Their experimental study shows that using weighted constraints can lead to improved ER quality and scalability.