No abstract

Recommending database queries is an emerging and promising field of research and is of particular interest in the domain of OLAP systems, where the user is left with the tedious process of navigating large datacubes. In this paper, the authors present a framework for a recommender system for OLAP users that leverages former users’ investigations to enhance discovery-driven analysis. This framework recommends the discoveries detected in former sessions that investigated the same unexpected data as the current session. This task is accomplished by (1) analysing the query log to discover pairs of cells at various levels of detail for which the measure values differ significantly, and (2) analysing a current query to detect if a particular pair of cells for which the measure values differ significantly can be related to what is discovered in the log. This framework is implemented in a system that uses the open source Mondrian server and recommends MDX queries. Preliminary experiments were conducted to assess the quality of the recommendations in terms of precision and recall, as well as the efficiency of their on-line computation.

Analytical queries are queries with numerical aggregators: computing the average number of objects per property, identifying the most frequent subjects, etc. Such queries are essential to monitor the quality and the content of the Linked Open Data (LOD) cloud. Many analytical queries cannot be executed directly on the SPARQL endpoints, because the fair use policy cuts off expensive queries. In this paper, we show how to rewrite such queries into a set of queries that each satisfy the fair use policy. We then show how to execute these queries in such a way that the result provably converges to the exact query answer. Our algorithm is an anytime algorithm, meaning that it can give intermediate approximate results at any time point. Our experiments show that the approach converges rapidly towards the exact solution, and that it can compute even complex indicators at the scale of the LOD cloud.

International audienceData mining is the study of how to extract information from data and express it as useful knowledge. One of its most important subfields, pattern mining, involves searching and enumerating interesting patterns in data. Various aspects of pattern mining are studied in the theory of computation and statistics. In the last decade, the pattern mining community has witnessed a sharp shift from efficiency-based approaches to methods which can extract more meaningful patterns. Recently, new methods adapting results from studies of economic efficiency and multi-criteria decision analyses such as Pareto efficiency, or skylines, have been studied. Within pattern mining, this novel line of research allows the easy expression of preferences according to a dominance relation. This approach is useful from a user-preference point of view and tends to promote the use of pattern mining algorithms for non-experts. We present a significant extension of our previous work [1] and [2] on the discovery of skyline patterns (or “skypatterns”) based on the theoretical relationships with condensed representations of patterns. We show how these relationships facilitate the computation of skypatterns and we exploit them to propose a flexible and efficient approach to mine skypatterns using a dynamic constraint satisfaction problems (CSP) framework.We present a unified methodology of our different approaches towards the same goal. This work is supported by an extensive experimental study allowing us to illustrate the strengths and weaknesses of each approach

Many applications rely on distributed databases. However, only few discovery methods exist to extract patterns without centralizing the data. In fact, this centralization is often less expensive than the communication of extracted patterns from the different nodes. To circumvent this difficulty, this paper revisits the problem of pattern mining in distributed databases by benefiting from pattern sampling. Specifically, we propose the algorithm DDSampling that randomly draws a pattern from a distributed database with a probability proportional to its interest. We demonstrate the soundness of DDSampling and analyze its time complexity. Finally, experiments on benchmark datasets highlight its low communication cost and its robustness. We also illustrate its interest on real-world data from the Semantic Web for detecting outlier entities in DBpedia and Wikidata.

No abstract

No abstract

Knowledge bases (KBs) such as DBpedia, Wikidata, and YAGO contain a huge number of entities and facts. Several recent works induce rules or calculate statistics on these KBs. Most of these methods are based on the assumption that the data is a representative sample of the studied universe. Unfortunately, KBs are biased because they are built from crowdsourcing and opportunistic agglomeration of available databases. This paper aims at approximating the representativeness of a relation within a knowledge base. For this, we use the generalized Benford's law, which indicates the distribution expected by the facts of a relation. We then compute the minimum number of facts that have to be added in order to make the KB representative of the real world. Experiments show that our unsupervised method applies to a large number of relations. For numerical relations where ground truths exist, the estimated representativeness proves to be a reliable indicator.