Departamento de Ciencia de la Computación

In spite of being the standard query language for RDF, SPARQL is not efficient. In fact, the problem of evaluating SPARQL patterns is PSPACE-complete. In practice, however, the features that make this problem so hard appear to be used in very restricted ways. Establishing these restrictions should lead to low complexity bounds for the average case. In this article we propose a detailed study of real-world RDF databases, which will provide us with the background to extend the semantics of SPARQL to handle the aforementioned features. From there, we expect to be able to develop and test (efficient) query processing algorithms in a pilot implementation of an RDF data management system.

Blank nodes are defined in RDF as 'existential variables' in the same way that has been used before in mathematical logic. However, evidence suggests that actual usage of RDF does not follow this definition. In this paper we thoroughly cover the issue of blank nodes, from incomplete information in database theory, over different treatments of blank nodes across the W3C stack of RDF-related standards, to empirical analysis of RDF data publicly available on the Web.

In Computer Science, the area called “Formal Languages” is concerned about sets of words, or strings, and their properties. Every programming language is a particular case of a formal language, so applications arising from this area are almost everywhere, including, for example, every personal computer, web page and mobile phone.

In Computer Science, the area called “Formal Languages” is concerned about sets of words, or strings, and their properties. Every programming language is a particular case of a formal language, so applications arising from this area are almost everywhere, including, for example, every personal computer, web page and mobile phone. Although the theory of formal languages is a subject of study on its own, and is usually taught as a self-contained undergraduate course for Computer Science/Informatics students, connections with other ...

Jointly with the release of RDF in 1999 as recommendation of the W3C, the natural problem of querying RDF data was raised. Since then, several designs and implementations of RDF query languages have been proposed (see [9] and [7] for detailed comparisons of RDF query languages). In 2004, the RDF Data Access Working Group released a first public working draft of a query language for RDF, called SPARQL [13]. Since then, SPARQL has been rapidly adopted as the standard for querying Semantic Web data. In fact, SPARQL became a W3C Recommendation in January 2008. In spite of being the standard query language for RDF, the design of SPARQL was made to keep the efficiency of the language considering the current database technology. In this direction, the current definition of the semantics of SPARQL does not consider the combined treatment of two of the distinctive features of RDF graphs, namely the semantics of blank nodes and RDFS vocabulary recommended by the W3C in the definition of RDF [10]. In fact, the semantics of SPARQL does not match in some constructions the semantics for blank nodes recommended by the W3C in [10]. To see that this is the case, consider the RDF graphs G 1 and G 2 shown in Figure 1. In these graphs, :b1, :b2 and :b3 are blank nodes, which are used to represent objects that are owned by John and Peter. According to the semantics for blank nodes proposed by the W3C [10, 8], these two graphs are equivalent as they can be mapped into each other 4. Thus, one would expect that the answer to any SPARQL query over G 1 is the same as over G 2. However, this is not the case for the following SPARQL query Q: SELECT DISTINCT ?X WHERE { ?X owns ?Y. ?X owns ?Z. FILTER (?Y != ?Z) } According to the semantics of SPARQL, the answer to Q over G 1 is John since :b1 and :b2 are distinct values, while the answer to Q over G 2 is the empty set. 4 Function f defined as f (:b1) = f (:b2) = :b3 is a homomorphism from G1 into G2, while function g defined as g(:b3) = :b1 is a homomorphism from G2 into G1.

In spite of being the standard query language for RDF, SPARQL is not efficient. In fact, the problem of evaluating SPARQL patterns is PSPACEcomplete. In practice, however, the features that make this problem so hard appear to be used in very restricted ways. Establishing these restrictions should lead to low complexity bounds for the average case. In this article we propose a detailed study of real-world RDF databases, which will provide us with the background to extend the semantics of SPARQL to handle the aforementioned features. From there, we expect to be able to develop and test (efficient) query processing algorithms in a pilot implementation of an RDF data management system.

In this paper we thoroughly cover the issue of blank nodes, which have been defined in RDF as 'existential variables'. We first introduce the theoretical precedent for existential blank nodes from first order logic and incomplete information in database theory. We then cover the different (and sometimes incompatible) treatment of blank nodes across the W3C stack of RDF-related standards. We present an empirical survey of the blank nodes present in a large sample of RDF data published on the Web (the BTC-2012 dataset), where we find that 25.7% of unique RDF terms are blank nodes, that 44.9% of documents and 66.2% of domains featured use of at least one blank node, and that aside from one Linked Data domain whose RDF data contains many "blank node cycles", the vast majority of blank nodes form tree structures that are efficient to compute simple entailment over. With respect to the RDF-merge of the full data, we show that 6.1% of blank-nodes are redundant under simple entailment. The vast majority of non-lean cases are isomorphisms resulting from multiple blank nodes with no discriminating information being given within an RDF document or documents being duplicated in multiple Web locations. Although simple entailment is NP-complete and leanness-checking is coNP-complete, in computing this latter result, we demonstrate that in practice, real-world RDF graphs are sufficiently "rich" in ground information for problematic cases to be avoided by non-naive algorithms.

Business Process Management has gained importance within organizations due to the need to streamline their operations. Nevertheless, despite the existence of process modeling standards such as BPMN, nowadays it is difficult to specify complex temporal constraints and relationships among tasks of a given process, which prevents the specification and subsequent automation of processes where these restrictions are relevant. To solve the exposed difficulty, we have resorted to the project planning and management field, developing a BPMN equivalency of all temporal constraints and relationships that can be specified in a standard project planning tool: Microsoft Project. This not only enables a simple interface for specifying complex temporal restrictions in business processes, but also defines an execution semantic for the models developed in the field of project planning, allowing their later automation through process execution engines.

Today quality and safety are major drivers for the food industry. In this context, computer vision technologies might prove highly relevant, providing objective measurements of relevant visual attributes related to food quality and safety, such as, the shape or color of a given good. As a major producer of food products, this type of technologies are highly relevant for Latin America. In this paper we analyze and share our experience developing computer vision technologies for the food industry in a Latin American country. In particular, as a case study, we describe VisionQ-Lab, an application of computer vision technologies to classify seeds, beans, and grains. To demonstrate the capabilities of VisionQ-Lab, we present the results of testing its operation using real data corresponding to different types of rice grains and lentils provided by a food company. The results of our test indicate an average accuracy in the classification close to 95%.

I. Lillo, J. Plett, M. Guarini, R. Mir, P. Irarrazaval Electrical Engineering, Pontificia Universidad Catolica de Chile, Santiago, Chile INTRODUCTION. It has been shown that signals or images having a sparse or compressible representation in a given domain can be reconstructed from a few randomly drawn samples, obtained from a domain whose components are linear combinations of the pursued image. The process minimizes the L1 norm of the sparse domain coefficients, keeping the consistency between the acquired samples and the resulting image [1]. Phantoms usually allow an exact reconstruction because, in general, their images have a sparse representation, hence, are easily compressed. However, it has been difficult to achieve good results with in vivo images. To tackle this problem, we hypothesize that a windowed 2D-DCT (2DDiscrete Cosine Transform) of in vivo images, JPEG like, allows concentration of the image content in a few coefficients with a good combination of localization (win...