University of Luxembourg

About: University of Luxembourg is a education organization based out in Luxembourg, Luxembourg. It is known for research contribution in the topics: Context (language use) & Computer science. The organization has 4744 authors who have published 22175 publications receiving 381824 citations.

Example-dependent cost-sensitive decision trees

Abstract: Example-dependent cost-sensitive tree algorithm.Each example is assumed to have different financial cost.Application on credit card fraud detection, credit scoring and direct marketing.Focus on maximizing the financial savings instead of accuracy.Code is open source and available at albahnsen.com/CostSensitiveClassification. Several real-world classification problems are example-dependent cost-sensitive in nature, where the costs due to misclassification vary between examples. However, standard classification methods do not take these costs into account, and assume a constant cost of misclassification errors. State-of-the-art example-dependent cost-sensitive techniques only introduce the cost to the algorithm, either before or after training, therefore, leaving opportunities to investigate the potential impact of algorithms that take into account the real financial example-dependent costs during an algorithm training. In this paper, we propose an example-dependent cost-sensitive decision tree algorithm, by incorporating the different example-dependent costs into a new cost-based impurity measure and a new cost-based pruning criteria. Then, using three different databases, from three real-world applications: credit card fraud detection, credit scoring and direct marketing, we evaluate the proposed method. The results show that the proposed algorithm is the best performing method for all databases. Furthermore, when compared against a standard decision tree, our method builds significantly smaller trees in only a fifth of the time, while having a superior performance measured by cost savings, leading to a method that not only has more business-oriented results, but also a method that creates simpler models that are easier to analyze.

Hybrid software and system development in practice: waterfall, scrum, and beyond

Abstract: Software and system development faces numerous challenges of rapidly changing markets. To address such challenges, companies and projects design and adopt specific development approaches by combining well-structured comprehensive methods and flexible agile practices. Yet, the number of methods and practices is large, and available studies argue that the actual process composition is carried out in a fairly ad-hoc manner. The present paper reports on a survey on hybrid software development approaches. We study which approaches are used in practice, how different approaches are combined, and what contextual factors influence the use and combination of hybrid software development approaches. Our results from 69 study participants show a variety of development approaches used and combined in practice. We show that most combinations follow a pattern in which a traditional process model serves as framework in which several fine-grained (agile) practices are plugged in. We further show that hybrid software development approaches are independent from the company size and external triggers. We conclude that such approaches are the results of a natural process evolution, which is mainly driven by experience, learning, and pragmatism.

From meta-omics to causality: experimental models for human microbiome research

Abstract: Large-scale ‘meta-omic’ projects are greatly advancing our knowledge of the human microbiome and its specific role in governing health and disease states. A myriad of ongoing studies aim at identifying links between microbial community disequilibria (dysbiosis) and human diseases. However, due to the inherent complexity and heterogeneity of the human microbiome, cross-sectional, case–control and longitudinal studies may not have enough statistical power to allow causation to be deduced from patterns of association between variables in high-resolution omic datasets. Therefore, to move beyond reliance on the empirical method, experiments are critical. For these, robust experimental models are required that allow the systematic manipulation of variables to test the multitude of hypotheses, which arise from high-throughput molecular studies. Particularly promising in this respect are microfluidics-based in vitro co-culture systems, which allow high-throughput first-pass experiments aimed at proving cause-and-effect relationships prior to testing of hypotheses in animal models. This review focuses on widely used in vivo, in vitro, ex vivo and in silico approaches to study host-microbial community interactions. Such systems, either used in isolation or in a combinatory experimental approach, will allow systematic investigations of the impact of microbes on the health and disease of the human host. All the currently available models present pros and cons, which are described and discussed. Moreover, suggestions are made on how to develop future experimental models that not only allow the study of host-microbiota interactions but are also amenable to high-throughput experimentation.

Codebook-Based Hybrid Precoding for Millimeter Wave Multiuser Systems

Abstract: In millimeter-wave (mmWave) systems, antenna architecture limitations make it difficult to apply conventional fully digital precoding techniques but call for low-cost analog radio frequency (RF) and digital baseband hybrid precoding methods. This paper investigates joint RF-baseband hybrid precoding for the downlink of multiuser multiantenna mmWave systems with a limited number of RF chains. Two performance measures, maximizing the spectral efficiency and the energy efficiency of the system, are considered. We propose a codebook-based RF precoding design and obtain the channel state information via a beam sweep procedure. Via the codebook-based design, the original system is transformed into a virtual multiuser downlink system with the RF chain constraint. Consequently, we are able to simplify the complicated hybrid precoding optimization problems to joint codeword selection and precoder design (JWSPD) problems. Then, we propose efficient methods to address the JWSPD problems and jointly optimize the RF and baseband precoders under the two performance measures. Finally, extensive numerical results are provided to validate the effectiveness of the proposed hybrid precoders.

PrankWeb: a web server for ligand binding site prediction and visualization.

Abstract: PrankWeb is an online resource providing an interface to P2Rank, a state-of-the-art method for ligand binding site prediction. P2Rank is a template-free machine learning method based on the prediction of local chemical neighborhood ligandability centered on points placed on a solvent-accessible protein surface. Points with a high ligandability score are then clustered to form the resulting ligand binding sites. In addition, PrankWeb provides a web interface enabling users to easily carry out the prediction and visually inspect the predicted binding sites via an integrated sequence-structure view. Moreover, PrankWeb can determine sequence conservation for the input molecule and use this in both the prediction and result visualization steps. Alongside its online visualization options, PrankWeb also offers the possibility of exporting the results as a PyMOL script for offline visualization. The web frontend communicates with the server side via a REST API. In high-throughput scenarios, therefore, users can utilize the server API directly, bypassing the need for a web-based frontend or installation of the P2Rank application. PrankWeb is available at http://prankweb.cz/, while the web application source code and the P2Rank method can be accessed at https://github.com/jendelel/PrankWebApp and https://github.com/rdk/p2rank, respectively.