University of Los Andes

About: University of Los Andes is a education organization based out in Bogotá, Colombia. It is known for research contribution in the topics: Population & Large Hadron Collider. The organization has 17616 authors who have published 25555 publications receiving 413463 citations.

Learning Accurate LSTM Models of Business Processes

Abstract: Deep learning techniques have recently found applications in the field of predictive business process monitoring. These techniques allow us to predict, among other things, what will be the next events in a case, when will they occur, and which resources will trigger them. They also allow us to generate entire execution traces of a business process, or even entire event logs, which opens up the possibility of using such models for process simulation. This paper addresses the question of how to use deep learning techniques to train accurate models of business process behavior from event logs. The paper proposes an approach to train recurrent neural networks with Long-Short-Term Memory (LSTM) architecture in order to predict sequences of next events, their timestamp, and their associated resource pools. An experimental evaluation on real-life event logs shows that the proposed approach outperforms previously proposed LSTM architectures targeted at this problem.

Heavy Oil−Water Interfacial Properties and Emulsion Stability: Influence of Dilution

Abstract: We have performed a study of dynamic interfacial tension and rheology of interfaces between water and either crude oil diluted in toluene or in a model oil consisting of toluene, in which pentane-extracted asphaltenes were redissolved. We show that the break point of the curve interfacial tension versus asphaltene concentration is not related to an eventual critical aggregation concentration but rather to the collapse of the asphaltene surface layer. As for other amphiphilic species that adsorb irreversibly at an oil−water interface, the corresponding emulsion stability increases with the amphiphile concentration, even above the collapse point. The behavior of diluted crude oil and model oil are quite similar. However, the surface layers evolve over longer time scales in the case of diluted crude oil, probably because of the contribution of resins that form less polar aggregates with asphaltenes. This results in less stable emulsions, especially before the collapse point.

Measurement of the anomalous like-sign dimuon charge asymmetry with 9 fb(-1) of p(p)over-bar collisions

Abstract: We present an updated measurement of the anomalous like-sign dimuon charge asymmetry A(sl)(b) for semileptonic b-hadron decays in 9.0 fb(-1) of p (p) over bar collisions recorded with the D0 detector at a center-of-mass energy of root s = 1.96 TeV at the Fermilab Tevatron collider. We obtain A(sl)(b) = (-0.787 +/- 0.172(stat) +/- 0.093(syst)%. This result differs by 3.9 standard deviations from the prediction of the standard model and provides evidence for anomalously large CP violation in semileptonic neutral B decay. The dependence of the asymmetry on the muon impact parameter is consistent with the hypothesis that it originates from semileptonic b-hadron decays.