Stronger Lempel-Ziv Based Compressed Text Indexing

Given a text T[1..u] over an alphabet of size σ, the full-text search problem consists in finding the occ occurrences of a given pattern P[1..m] in T. In indexed text searching we build an index on T to improve the search time, yet increasing the space requirement. The current trend in indexed text searching is that of compressed full-text self-indices, which replace the text with a more space-efficient representation of it, at the same time providing indexed access to the text. Thus, we can provide efficient access within compressed space. The Lempel-Ziv index (LZ-index) of Navarro is a compressed full-text self-index able to represent T using 4uH _k (T)+o(ulog?σ) bits of space, where H _k (T) denotes the k-th order empirical entropy of T, for any k=o(log? _σ u). This space is about four times the compressed text size. The index can locate all the occ occurrences of a pattern P in T in O(m ³log?σ+(m+occ)log?u) worst-case time. Although this index has proven very competitive in practice, the O(m ³log?σ) term can be excessive for long patterns. Also, the factor 4 in its space complexity makes it larger than other state-of-the-art alternatives. In this paper we present stronger Lempel-Ziv based indices (LZ-indices), improving the overall performance of the original LZ-index. We achieve indices requiring (2+ε)uH _k (T)+o(ulog?σ) bits of space, for any constant ε>0, which makes them the smallest existing LZ-indices. We simultaneously improve the search time to O(m ²+(m+occ)log?u), which makes our indices very competitive with state-of-the-art alternatives. Our indices support displaying any text substring of length ? in optimal O(?/log? _σ u) time. In addition, we show how the space can be squeezed to (1+ε)uH _k (T)+o(ulog?σ) to obtain a structure with O(m ²) average search time for m≥2log? _σ u. Alternatively, the search time of LZ-indices can be improved to O((m+occ)log?u) with (3+ε)uH _k (T)+o(ulog?σ) bits of space, which is much less than the space needed by other Lempel-Ziv-based indices achieving the same search time. Overall our indices stand out as a very attractive alternative for space-efficient indexed text searching.     

Probabilistic proximity search: Fighting the curse of dimensionality in metric spaces

Proximity searches become very difficult on “high dimensional” metric spaces, that is, those whose histogram of distances has a large mean and/or a small variance. This so-called “curse of dimensionality”, well known in vector spaces, is also observed in metric spaces. The search complexity grows sharply with the dimension and with the search radius. We present a general probabilistic framework applicable to any search algorithm and whose net effect is to reduce the search radius. The higher the dimension, the more effective the technique. We illustrate empirically its practical performance on a particular class of algorithms, where large improvements in the search time are obtained at the cost of a very small error probability.     

Applying EuroWordNet to Cross-Language Text Retrieval

We discuss ways in which EuroWordNet (EWN) can be used in multilingual information retrieval activities, focusing on two approaches to Cross-Language Text Retrieval that use the EWN database as a large-scale multilingual semantic resource. The first approach indexes documents and queries in terms of the EuroWordNet Inter-Lingual-Index, thus turning term weighting and query/document matching into language-independent tasks. The second describes how the information in the EWN database could be integrated with a corpus-based technique, thus allowing retrieval of domain-specific terms that may not be present in our multilingual database. Our objective is to show the potential of EuroWordNet as a promising alternative to existing approaches to Cross-Language Text Retrieval.     

Structural studies of pulsed-laser deposited nanocomposite metal-oxide films

Pulsed laser deposition in vacuum has been used to develop metal-oxide nanocomposite films with well controlled structural quality. Results for the copper–aluminium oxide (Cu:Al₂O₃) system are used to illustrate the main morphological and structural features of these films. High resolution transmission electron microscopy (TEM) analysis shows that the films consist of Cu nanocrystals with average dimensions that can be controlled between 2 nm and 10 nm embedded in an amorphous Al₂O₃ matrix. It is observed that the in-plane shape of the nanocrystals evolves from circular to elongated, and the number of nanocrystals per unit area decreases as their size increases. This evolution is explained in terms of nucleation at the substrate surface and coalescence during the later stages of growth. The thermal stability of the films has been studied by in situ TEM annealing and no transformation could be observed up to about 800 °C when partial crystallization of the Al₂O₃ starts.     

A differential inclusion approach for modeling and analysis of dynamical systems under uncertainty

In this paper we deal with the application of differential inclusions to modeling nonlinear dynamical systems under uncertainty in parameters. In this case, differential inclusions seem to be better suited to modeling practical situations under uncertainty and imprecision than formulations by means of fuzzy differential equations. We develop a practical algorithm to approximate the reachable sets of a class of nonlinear differential inclusion, which eludes the computational problems of a previous set-valued version of the Heun’s method. Our algorithm is based on a complete discretization (time and state space) of the differential inclusion and it suits hardware features, handling the memory used by the method in a controlled fashion during all iterations. As a case of study, we formulate a differential inclusion to model an epidemic outbreak of dengue fever under Cuban conditions. The model takes into account interaction of human and mosquito populations as well as vertical transmission in the mosquito population. It is studied from the theoretical point of view to apply the Practical Algorithm. Also, we estimate the temporal evolution of the different human and mosquito populations given by the model in the Dengue 3 epidemic in Havana 2001, through the computation of the reachable sets using the Practical Algorithm.     

Challenges for integrated design (ID) in sustainable buildings

It is often argued that integrated design (ID) is a powerful way to enhance collaboration in construction projects. This collaboration is seen as a way of improving innovation to create more sustainable buildings. Contrary to the traditional silo-type and linear design process, ID is based on upfront stakeholder involvement and a holistic approach to project decision-making. Although ID’s premises are theoretically-founded, a close empirical look at its practices shows that numerous challenges compromise its results and efficiency. This study examines the ID process through an iterative process that includes the construction of a conceptual framework and its empirical validation. We examine three green construction projects in Canada. Based on the analysis of 26 interviews with key project stakeholders and more than 198 construction documents, the study assembles – and ultimately applies – a multi-lens framework based on four themes: the fragmented nature of construction; risk perception; stakeholders’ commitment; and efficiency in the design process. Results show that three tensions arise in ID practices: between collaboration and process efficiency, between short-term and long-term goals, and between integrated methods and traditional ones. The study sheds light on ID limits and reveals how stakeholders can improve their interactions to design more sustainable buildings.     

Use of β-hydroxyacyl-CoA-dehydrogenase (HADH) activity to differentiate frozen from unfrozen fish and shellfish

 Further work on an enzymic method to differentiate frozen from unfrozen fish and shellfish is reported. The method is based on the release of the β-hydroxyacyl-CoA-dehydrogenase (HADH) from mitochondria during freezing. Enzymic activity was evaluated in fresh and frozen thawed samples from sole (Solea solea), sea bream (Pagellus centrodontus), hake (Merluccius merluccius), gilt headed bream (Sparus aurata), sea bass (Dicentrarchus labrax), salmon (Salmo salar), prawn (Penaeus japonicus) and Norwegian lobster (Nephrops norvegicus). Changes in the HADH activity of fresh and frozen thawed samples were compared after freezing at –196  °C for 15 min. Two values were obtained: U (by dividing: HADH activity of samples frozen at –196  °C, then thawed/HADH activity of unfrozen samples) and F (by dividing: HADH activity of samples frozen at –18  °C, thawed, then frozen at –196  °C /HADH activity of samples frozen at –18  °C, then thawed). Statistical analysis showed significant differences (P≤0.05) between both quotients for gilt headed bream, salmon, sea bream, sole and prawn, and an arbitrary limit was set at 2 to differentiate frozen thawed from unfrozen samples. The application of this limit made it possible to discriminate the unfrozen from the frozen thawed state of around 90% of the total samples analysed. Best results were obtained for prawn (100% of samples differentiated). In the present paper, a laboratory routine is proposed based on the comparison of the HADH activity of a sample analysed straight away and that of a sample frozen at –196  °C and then thawed. The reported method is simple and fast. The entire laboratory procedure can be performed in 45 min. Received: 20 July 1998 / Revised version: 2 November 1998     

Convolutional Neural Networks for Mechanistic Driver Detection in Atrial Fibrillation

The maintaining and initiating mechanisms of atrial fibrillation (AF) remain controversial. Deep learning is emerging as a powerful tool to better understand AF and improve its treatment, which remains suboptimal. This paper aims to provide a solution to automatically identify rotational activity drivers in endocardial electrograms (EGMs) with convolutional recurrent neural networks (CRNNs). The CRNN model was compared with two other state-of-the-art methods (SimpleCNN and attention-based time-incremental convolutional neural network (ATI-CNN)) for different input signals (unipolar EGMs, bipolar EGMs, and unipolar local activation times), sampling frequencies, and signal lengths. The proposed CRNN obtained a detection score based on the Matthews correlation coefficient of 0.680, an ATI-CNN score of 0.401, and a SimpleCNN score of 0.118, with bipolar EGMs as input signals exhibiting better overall performance. In terms of signal length and sampling frequency, no significant differences were found. The proposed architecture opens the way for new ablation strategies and driver detection methods to better understand the AF problem and its treatment.