Search

[EN] Cellulose nanocrystals (CNCs) were isolated for the first time from pine cones (PC) by alkali and bleaching treatments and subsequent sulfuric acid hydrolysis (64 %) at 45 ºC. The influence of the hydrolytic reaction time (30, 45 and 90 min) on the yield, chemical composition and structure, and thermal stability of CNCs was evaluated. The removal of non-cellulosic constituents during the alkaline and bleaching treatment resulted in high pure cellulosic fibres. The isolation of CNCs from these cellulosic fibres at different reaction times was verified by the nano-dimensions of the individual crystals (< 3 and, This work was supported by the Ministry of Economy and Competitiveness (MINECO) [MAT2014-59242-C2-1-R]. D. García-García wants to thanks the Spanish Ministry of Education, Culture and Sports for their financial support through an FPU Grant [FPU13/06011].

[EN] For the economically depressed communities such as of those in the Canton Guaranda, Ecuador, to generate their own energy, from organic waste is very important because they are sometimes insulated and its gas and electricity supply is very deficient. The aim of this research, was to determine the feasibility of anaerobic co-digestion of wasted cabbage from the town's market in Guaranda, Ecuador, and livestock manure. Two variables were studied: temperature of the process and the percentage of cabbage and livestock manure. Biogas quantity and kinetic parameters were evaluated. Kinetic model were analyzed by minimizing the mean percentage of error between the observed values (measured experimentally) and predicted, using the Runge-Kutta of order 4 for solving the system of differential equations obtained from mass balance. The results showed that a 50-50% ratio cabbage-manure at 30 degrees C temperature gave the highest production of biogas achieved is (389.47cm(3)N/g initial SV) with a composition of 61% methane. The kinetic parameters found were mu(max)=0.1053day(-1); Ks=0.1153mg/l; Y=0.00246g VSS /g COD and K-dec=0.001005day(-1)., This research work has been carried out inside the cooperation framework funded by the ADSIDEO program of the Centro de Cooperacion al Desarrollo (CCD) of Universidad Politecnica de Valencia (Spain), in collaboration with the Centro de Estudios de la Biomasa (CEB), Universidad Estatal de Bolivar, Guaranda, Ecuador. The participation of Dr. Sergio Perez in this work was possible thanks to funding from the Ecuadorian Government by means of the PROMETEO program, led by the Secretaria Nacional de Educacion Superior, Ciencia y Tecnologia (SENESCYT).

International audience; Network Control Systems (NAC) have been used in many industrial processes. They aim to reduce the human factor burden and efficiently handle the complex process and communication of those systems. Supervisory control and data acquisition (SCADA) systems are used in industrial, infrastructure and facility processes (e.g. manufacturing, fabrication, oil and water pipelines, building ventilation, etc.) Like other Internet of Things (IoT) implementations, SCADA systems are vulnerable to cyber-attacks, therefore, a robust anomaly detection is a major requirement. However, having an accurate anomaly detection system is not an easy task, due to the difficulty to differentiate between cyber-attacks and system internal failures (e.g. hardware failures). In this paper, we present a model that detects anomaly events in a water system controlled by SCADA. Six Machine Learning techniques have been used in building and evaluating the model. The model classifies different anomaly events including hardware failures (e.g. sensor failures), sabotage and cyber-attacks (e.g. DoS and Spoofing). Unlike other detection systems, our proposed work helps in accelerating the mitigation process by notifying the operator with additional information when an anomaly occurs. This additional information includes the probability and confidence level of event(s) occurring. The model is trained and tested using a real-world dataset.

[EN] Neighbor discovery represents a first step after the deployment of wireless ad hoc networks, since the nodes that form them are equipped with limited-range radio transceivers, and they typically do not know their neighbors. In this paper two randomized neighbor discovery approaches, called CDH and CDPRR, based on collision detection for static multi-hop wireless ad hoc networks, are presented. Castalia 3.2 simulator has been used to compare our proposed protocols against two protocols chosen from the literature and used as reference: the PRR, and the Hello protocol. For the experiments, we chose five metrics: the neighbor discovery time, the number of discovered neighbors, the energy consumption, the throughput and the number of discovered neighbors versus packets sent ratio. According to the results obtained through simulation, we can conclude that our randomized proposals outperform both Hello and PRR protocols in the presence of collisions regarding all five metrics, for both one-hop and multi-hop scenarios. As novelty compared to the reference protocols, both proposals allow nodes to discover all their neighbors with probability 1, they are based on collision detection and know when to terminate the neighbor discovery process. Furthermore, qualitative comparisons of the existing protocols and the proposals are available in this paper. Moreover, CDPRR presents better results in terms of time, energy consumption and number of discovered neighbors versus packets sent ratio. We found that both proposals achieve to operate under more realistic assumptions. Furthermore, CDH does not need to know the number of nodes in the network., This work has been partially supported by the "Ministerio de Economia y Competitividad" in the "Programa Estatal de Fomento de la Investigacion Cientifica y Tecnica de Excelencia, Subprograma Estatal de Generacion de Conocimiento" within the project under Grant TIN2017-84802-C2-1-P. This work has also been partially supported by European Union through the ERANETMED (Euromediterranean Cooperation through ERANET joint activities and beyond) project ERANETMED3-227 SMARTWATIR.

This paper presents a numerical procedure to couple shell to solid elements by using the Nitsche’s method. The continuity of displacements can be satisfied approximately with the penalty method, which is effective in setting the penalty parameter to a sufficiently large value. When the continuity of only displacements on the interface is applied between shell and solid elements, an unreasonable deformation may be observed near the interface. In this work, the continuity of the stress vector on the interface is considered by employing the Nitsche’s method, and hence a reasonable deformation can be obtained on the interface. The authors propose two types of shell elements coupled with solid elements in this paper. One of them is the conventional MITC4 shell element, which is one of the most popular elements in engineering applications. This approach shows the capability of discretizing the domain of the structure with the different types of elements. The other is the shell element with additional degrees of freedom to represent thickness–stretch developed by the authors. In this approach, the continuity of displacements including the deformation in the thickness direction on the interface can be considered. Several numerical examples are presented to examine the fundamental performance of the proposed procedure. The behavior of the proposed simulation model is compared with that of the whole domain discretized with only solid elements., This is a post-peer-review, pre-copyedit version of an article published in "Computational Mechanics". The final authenticated version is available online at: https://doi.org/10.1007/s00466-018-1585-6

[EN] Customer service measures are traditionally used to determine the performance or/and the control parameters of any inventory system. Among them, the fill rate is one of the most widely used in practice and is defined as the fraction of demand that is immediately met from shelf i.e. from the available on-hand stock. However, this definition itself set out several problems that lead to consider two different approaches to compute the fill rate: the traditional, which computes the fill rate in terms of units short; and the standard, which directly computes the expected satisfied demand. This paper suggest two expressions, the traditional and the standard, to compute the fill rate in the continuous reorder point, order quantity (s, Q) policy following these approaches. Experimental results shows that the traditional approach is biased since underestimate the real fill rate whereas the standard computes it accurately and therefore both approaches cannot be treated as equivalent. This paper focuses on the lost sales context and discrete distributed demands., This work was supported by the European Regional Development Fund and Spanish Government (MINECO/FEDER, UE) under the Project with reference DPI2015-64133-R.

[EN] According to the United Nations Office for Disaster Risk Reduction cities must take measures to anticipate disasters and mitigate their impact, protecting homes and cultural heritage, minimizing losses due to earthquakes and other threats. After the recent earthquakes in historical city centres, the evaluation of the seismic risk, even in regions of low or moderate seismicity, is imperative, being damage assessment and loss estimation essential for its reduction. In this paper the seismic risk analysis of the residential buildings in the Ciutat Vella District in Valencia, with great historical and architectural value, is presented. The vulnerability of Ciutat Vella has been assessed with the Vulnerability Index Method establishing the seismic quality of the residential buildings according to their structural typology, conservation status, age and geometry; identifying the most vulnerable buildings. Damage probability matrices have been obtained for the deterministic and probabilistic earthquake scenarios, estimating for each of them, the seismic risk in terms of direct social and economic losses, mapping the results using a GIS system tool. Results confirm that the vulnerability of the building stock in Ciutat Vella is high being the seismic risk significant. In the event of an earthquake, many residential buildings included in the Catalogue of Listed Buildings of the Special Protection Plan of Ciutat Vella will be damaged. The detailed analysis of the most vulnerable buildings will provide, in future research works, criteria for intervention to enhance their structural seismic response, safeguarding the architectural heritage and contributing to the reduction of human and material losses., The Authors wish to acknowledge the students Ana Perez Recatala, Beatriz Alborch Vidal and Blanca Salavert Pamblanco for their help in the building information retrieval under a collaboration grant program funded by the Ministry of Education, Culture and Sports. Finally, the authors want to thank P. McGowan for the review of the English version of the paper.

[EN] Augmented Reality (AR) has become a mainstream technology in the development of solutions for repair and maintenance operations. Although most of the AR solutions are still limited to specific contexts in industry, some consumer electronics companies have started to offer pre-packaged AR solutions as alternative to video-based tutorials (VT) for minor maintenance operations. In this paper, we present a comparative study of the acquired knowledge and user perception achieved with AR and VT solutions in some maintenance tasks of IT equipment. The results indicate that both systems help users to acquire knowledge in various aspects of equipment maintenance. Although no statistically significant differences were found between AR and VT solutions, users scored higher on the AR version in all cases. Moreover, the users explicitly preferred the AR version when evaluating three different usability and satisfaction criteria. For the AR version, a strong and significant correlation was found between the satisfaction and the achieved knowledge. Since the AR solution achieved similar learning results with higher usability scores than the video-based tutorials, these results suggest that AR solutions are the most effective approach to substitute the typical paper-based instructions in consumer electronics., This work has been supported by Spanish MINECO and EU ERDF programs under grant RTI2018-098156-B-C55.

[EN] The mixed no-idle flowshop scheduling problem arises in modern industries including integrated circuits, ceramic frit and steel production, among others, and where some machines are not allowed to remain idle between jobs. This paper describes an exact algorithm that uses Benders decomposition with a simple yet effective enhancement mechanism that entails the generation of additional cuts by using a referenced local search to help speed up convergence. Using only a single additional optimality cut at each iteration, and combined with combinatorial cuts, the algorithm can optimally solve instances with up to 500 jobs and 15 machines that are otherwise not within the reach of off-the-shelf optimization software, and can easily surpass ad-hoc existing metaheuristics. To the best of the authors' knowledge, the algorithm described here is the only exact method for solving the mixed no-idle permutation flowshop scheduling problem., This research project was partially supported by the Scientific and Technological Research Council of Turkey (TuBITAK) under Grant 1059B191600107. While writing this paper, Dr Hamzaday was a visiting researcher at the Southampton Business School at the University of Southampton. Ruben Ruiz is supported by the Spanish Ministry of Science, Innovation and Universities, under the Project 'OPTEP-Port Terminal Operations Optimization' (No. RTI2018-094940-B-I00) financed with FEDER funds. Thanks are due to two anonymous reviewers for their careful reading of the paper and helpful suggestions.

[EN] CO2-free urban logistics is one of the 10 objectives to reach by 2030 as part of transport policy. What technologies can help to accomplish it? In this paper, we discuss the very complex situation that today¿s big and modern cities are facing with a tremendous environment of many urban logistics companies running in the same city. In the majority of cases, there is less or none coordination among them worsening traffic congestions. We believe that intelligent techniques are one of the key approaches that can aid to support smart and sustainable urban logistic applications. There are large open problems in the field of cooperative urban logistics that can greatly improve with the help of artificial intelligence. Some solutions are cited in this paper, but the overall conclusion is that there is still much work to be done.

The role of prognostics and health management is ever more prevalent with advanced techniques of estimation methods. However, data processing and remaining useful life prediction algorithms are often very different. Some difficulties in accurate prediction can be tackled by redefining raw data parameters into more meaningful and comprehensive health level indicators that will then provide performance information. Proper data processing has a significant importance on remaining useful life predictions, for example, to deal with data limitations or/and multi-regime operating conditions. The framework proposed in this paper considers a similarity-based prognostic algorithm that is fed by the use of data normalisation and filtering methods for operational trajectories of complex systems. This is combined with a data-driven prognostic technique based on feed-forward neural networks with multi-regime normalisation. In particular, the paper takes a close look at how pre-processing methods affect algorithm performance. The work presented herein shows a conceptual prognostic framework that overcomes challenges presented by short-term test datasets and that increases the prediction performance with regards to prognostic metrics. Validerad;2019;Nivå 2;2019-04-12 (johcin)

The aim of the sparse analytic hierarchy process (SAHP) problem is to rank a set of alternatives based on their utility/importance; this task is accomplished by asking human decision-makers to compare selected pairs of alternatives and to specify relative preference information, in the form of ratios of utilities. However, such an information is often affected by subjective biases or inconsistencies. Moreover, there is no general consent on the best approach to accomplish this task, and in the literature several techniques have been proposed. Finally, when more than one decision-maker is involved in the process, there is a need to provide adequate methodologies to aggregate the available information. In this view, the contribution of this paper to the SAHP body of knowledge is twofold. From one side, it develops a novel methodology to aggregate sparse data given by multiple sources of information. From another side, the paper undertakes an experimental validation of the most popular techniques to solve the SAHP problem, discussing the strength points and shortcomings of the different methodology with respect to a real case study.

[EN] The transport of sensitive products is very important because their deterioration may cause the value lost and even the product rejection by the buyer. In addition, it is important to choose the optimal way to achieve this end. In a data network, the task of calculating the best routes is performed by routers. We can consider the optimal path as the one that provides a shortest route. However, if a real transport network is considered the shortest path can sometimes be affected by incidents and traffic jams that would make it inadvisable. On the other hand, when we need to come back, due to features that symmetry provides, it would be interesting to follow the same path in reverse sense. For this reason, in this paper we present a symmetric routing mechanism for cooperative monitoring system for the delivery of fresh products. The systems is based on a combination of fixed nodes and a mobile node that stores the path followed to be able of coming back following the same route in reverse sense. If this path is no longer available, the system will try to maintain the symmetry principle searching the route that provide the shortest time to the used in the initial trip. The paper shows the algorithm used by the systems to calculate the symmetric routes. Finally, the system is tested in a real scenario which combines different kind of roads. As the results shows, the energy consumption of this kind of nodes is highly influenced by the activity of sensors., This work has been supported by the "Ministerio de Economia y Competitividad", through the "Convocatoria 2014. Proyectos I+D -Programa Estatal de Investigacion Cientifica y Tecnica de Excelencia" in the "Subprograma Estatal de Generacion de Conocimiento", (project TIN2014-57991-C3-1- P) and the "programa para la Formacion de Personal Investigador - (FPI-2015-S2-884)" by the "Universitat Politecnica de Valencia".

[EN] In order to be reusable, history-based feature-based parametric CAD models must reliably allow for modifications while maintaining their original design intent. In this paper, we demonstrate that relations that fix the location of geometric entities relative to the reference system produce inflexible profiles that reduce model reusability. We present the results of an experiment where novice students and expert CAD users performed a series of modifications in different versions of the same 2D profile, each defined with an increasingly higher number of fix geometric constraints. Results show that the amount of fix constraints in a 2D profile correlates with the time required to complete reusability tasks, i.e., the higher the number of fix constraints in a 2D profile, the less flexible and adaptable the profile becomes to changes. In addition, a pilot software tool to automatically track this type of constraints was developed and tested. Results suggest that the detection of fix constraint overuse may result in a new metric to assess poor quality models with low reusability. The tool provides immediate feedback for preventing high semantic level quality errors, and assistance to CAD users. Finally, suggestions are introduced on how to convert fix constraints in 2D profiles into a negative metric of 3D model quality., The authors would like to thank Raquel Plumed for her support in the statistical analysis. This work has been partially funded by Grant UJI-A02017-15 (Universitat Jaume I) and DPI201784526-R (MINECO/AEI/FEDER, UE), project CAL-MBE. The authors also wish to thank the editor and reviewers for their valuable comments and suggestions that helped us improve the quality of the paper.

[EN] This paper presents and evaluates a system that provides TV and radio services in mobility using 4G communications. The system has mainly two blocks, one on the cloud and another on the mobile vehicle. On the cloud, a DVB (Digital Video Broadcasting) receiver obtains the TV/radio signal and prepares the contents to be sent through 4G. Specifically, contents are transcoded and packetized using the DASH (Dynamic Adaptive Streaming over HTTP) standard. Vehicles in mobility use their 4G connectivity to receive the flows transmitted by the cloud. The key element of the system is an on-board proxy that manages the received flows and offers them to the final users in the vehicle. The proxy contains a buffer that helps reduce the number of interruptions caused by hand over effects and lack of coverage. The paper presents a comparison between a live transmission using 4G connecting the clients directly with the cloud server and a near real-time transmission based on an on-board proxy. Results prove that the use of the proxy reduces the number of interruptions considerably and, thus, improves the Quality of Experience of users at the expense of slightly increasing the delay., This work is supported by the Centro para el Desarrollo Tecnologico Industrial (CDTI) from the Government of Spain under the project "Plataforma avanzada de conectividad en movilidad" (CDTI IDI-20150126) and the project "Desarrollo de nueva plataforma de entretenimiento multimedia para entornos nauticos" (CDTI TIC-20170102).

OpenCL has been proposed as a means of accelerating functional computation using FPGA and GPU accelerators. Although it provides ease of programmability and code portability, questions remain about the performance portability and underlying vendor’s compiler capabilities to generate efficient implementations without user-defined, platform specific optimizations. In this work, we systematically evaluate this by formalizing a design space exploration strategy using platform-independent micro-architectural and application-specific optimizations only. The optimizations are then applied across Altera FPGA, NVIDIA GPU and ARM Mali GPU platforms for three computing examples, namely matrix-matrix multiplication, binomial-tree option pricing and 3-dimensional finite difference time domain. Our strategy enables a fair comparison across platforms in terms of throughput and energy efficiency by using the same design effort. Our results indicate that FPGA provides better performance portability in terms of achieved percentage of device’s peak performance (68%) compared to NVIDIA GPU (20%) and also achieves better energy efficiency (up to 1.4\(\times \)) for some of the considered cases without requiring in-depth hardware design expertise.

International audience; This paper reports on a four-year project that aims to raise the abstraction level through the use of model-driven engineering (MDE) techniques in the development of scientific applications relying on high-performance computing. The development and maintenance of high-performance scientific computing software is reputedly a complex task. This complexity results from the frequent evolutions of supercomputers and the tight coupling between software and hardware aspects. Moreover, current parallel programming approaches result in a mixing of concerns within the source code. Our approach relies on the use of MDE and consists in defining domain-specific modeling languages targeting various domain experts involved in the development of HPC applications, allowing each of them to handle their dedicated model in a both user-friendly and hardware-independent way. The different concerns are separated thanks to the use of several models as well as several modeling viewpoints on these models. Depending on the targeted execution platforms, these abstract models are translated into executable implementations by means of model transformations. To make all of these effective, we have developed a tool chain that is also presented in this paper. The approach is assessed through a multi-dimensional validation that focuses on its applicability, its expressiveness and its efficiency. To capitalize on the gained experience, we analyze some lessons learned during this project.

[EN] Autonomous vehicle guidance and trajectory planning is one of the key technologies in the autonomous control system for intelligent vehicles. Firstly, the target pursuit model for intelligent vehicles was established and described in this text. Then, the research work for global motion planning was carried out based on Stackelberg Differential Game Theory, and the global optimal solution was obtained by using the survival type differential game. Finally, to overcome errors, we use a polynomial method to achieve the smooth motion planning. So, based on Terminal Sliding Mode method, the Active Front Steering controller design was used to calculate the desired active wheel angle for intelligent vehicle path tracking. The simulation and experiment results demonstrate the feasibility and effectiveness of this method for intelligent vehicles' path planning and tracking., This paper is supported by the Zhejiang Provincial Natural Science Foundation under Grant No. LY13E080010. The first author would like to appreciate Dr. Xuecai Yu and the reviewers for the valuable discussions to improve the quality and presentation of the paper.

The online shopping market is growing rapidly in the 21st century, leading to a huge amount of duplicate products being sold online. An important component for aggregating online products is duplicate detection, although this is a time consuming process. In this paper, we focus on reducing the amount of possible duplicates that can be used as an input for the Multi-component Similarity Method (MSM), a state-of-the-art duplicate detection solution. To find the candidate pairs, Locality Sensitive Hashing (LSH) is employed. A previously proposed LSH-based algorithm makes use of binary vectors based on the model words in the product titles. This paper proposes several extensions to this, by performing advanced data cleaning and additionally using information from the key-value pairs. Compared to MSM, the MSMP+ method proposed in this paper leads to a minor reduction by \(6\%\) in the \(F_1\)-measure whilst reducing the number of needed computations by \(95\%\).

[EN] Commendable amount of work has been attempted in the field of Sentiment Analysis or Opinion Mining from natural language texts and Twitter texts. One of the main goals in such tasks is to assign polarities (positive or negative) to a piece of text. But, at the same time, one of the important as well as difficult issues is how to assign the degree of positivity or negativity to certain texts. The answer becomes more complex when we perform a similar task on figurative language texts collected from Twitter. Figurative language devices such as irony and sarcasm contain an intentional secondary or extended meaning hidden within the expressions. In this paper we present a novel approach to identify the degree of the sentiment (fine grained in an 11-point scale) for the figurative language texts. We used several semantic features such as sentiment and intensifiers as well as we introduced sentiment abruptness, which measures the variation of sentiment from positive to negative or vice versa. We trained our systems at multiple levels to achieve the maximum cosine similarity of 0.823 and minimum mean square error of 2.170., The work reported in this paper is supported by a grant from the project “CLIA System Phase II” funded by Department of Electronics and Information Technology (DeitY), Ministry of Communications and Information Technology (MCIT), Government of India. The work of the fourth author is also supported by the SomEMBED TIN2015-71147-C2-1-P MINECO research project and by the Generalitat Valenciana under the grant ALMAPATER (PrometeoII/2014/030).

[EN] Significant changes in the Jucar River Basin District's hydrology in the Mediterranean side of Spain, have been observed during last decades. A statistical change-point in the year 1980 was detected in the basins' hydrological series in the main upper river, Jucar and Tuna basins. In the study scope are, the North Atlantic Oscillation (NAO) is linked with the winter precipitations in the Upper Basins, which are here responsible for the major part of streamflow. So changes in the rainfall has an important effect in the natural river flows. The statistical analysis detected a change at NAO's seasonal pattern, what means a considerable reduction of winter rainfalls in the Upper River basins located in the inland zone which is simultaneously the water collection and reservoirs area (a - 40% of water resources availability since 1980). Hydro-meteorological data and a Water Balance Model, Patrical, have been used to assess these water resources' reduction. Results points out to the change in the Basin's precipitation pattern in the inland areas (upper basins), associated to Atlantic weather patterns, as the main cause, while it has not been detected in the coastal areas. All these changes implies water stress for water resources planning, management and allocation, where more than 5.2 million people and irrigation of 390,000 ha are served, joint to the time variability, an important territorial imbalance exists between resources and demands. Thus, in the main upper basins, with the biggest streamflow's reductions, locate the largest reservoirs in terms of water resources collection and reserves., The authors would like to thank the Jucar RBD (Spanish Ministry of Environment) and the Confederacion Hidrografica del Jucar (Jucar River Basin Authority - RBA) for their cooperation in the compilation of this paper. The language revision of this paper was funded by the Universitat Politecnica de Valencia, Spain.

[EN] In this paper, in order to reliably meet the homogeneity required by customers, a fuzzy model is proposed to support the promising process in LHP contexts after taking into account uncertainty in planned homoge- neous sublots. The fuzzy model is translated into an alpha- parametric equivalent crisp model. Here, it is important to highlight another important novelty of the paper related to the proposed methodology to analyse the suitability of the minimum degree of possibility (the a-cut), by an adapted TOPSIS-based fuzzy procedure. Finally, the experimental design, which is inspired in the ceramic sector, proves both the validity of the model and a better performance of the fuzzy model compared to the deterministic one, in several executions with forecasts of the real size of homogeneous sublots., This research is partly supported by: The Ministry of Science, Technology and Telecommunications of the of Costa Rica Government (MICITT), through the Programme of Innovation and Human Capital for Competitiveness (PINN)(Contract No. PED-019-2015-1); and the Spanish Ministry of Economy and Competitiveness Projects "Methods and models for operations planning and order management in supply chains characterised by uncertainty in production due to the lack of product uniformity'' (PLANGES-FHP) (Ref. DPI2011-23597) and "Operations design and Management of Global Supply Chains'' (GLOBOP) (Ref. DPI2012-38061-C02-01).

[EN] Modern data-driven spoken language systems (SLS) require manual semantic annotation for training spoken language understanding parsers. Multilingual porting of SLS demands significant manual effort and language resources, as this manual annotation has to be replicated. Crowdsourcing is an accessible and cost-effective alternative to traditional methods of collecting and annotating data. The application of crowdsourcing to simple tasks has been well investigated. However, complex tasks, like cross-language semantic annotation transfer, may generate low judgment agreement and/or poor performance. The most serious issue in cross-language porting is the absence of reference annotations in the target language; thus, crowd quality control and the evaluation of the collected annotations is difficult. In this paper we investigate targeted crowdsourcing for semantic annotation transfer that delegates to crowds a complex task such as segmenting and labeling of concepts taken from a domain ontology; and evaluation using source language annotation. To test the applicability and effectiveness of the crowdsourced annotation transfer we have considered the case of close and distant language pairs: Italian-Spanish and Italian-Greek. The corpora annotated via crowdsourcing are evaluated against source and target language expert annotations. We demonstrate that the two evaluation references (source and target) highly correlate with each other; thus, drastically reduce the need for the target language reference annotations., This research is partially funded by the EU FP7 PortDial Project No. 296170, FP7 SpeDial Project No. 611396, and Spanish contract TIN2014-54288-C4-3-R. The work presented in this paper was carried out while the author was affiliated with Universitat Politecnica de Valencia.

We consider the problem of classifying business process instances based on structural features derived from event logs. The main motivation is to provide machine learning based techniques with quick response times for interactive computer assisted root cause analysis. In particular, we create structural features from process mining such as activity and transition occurrence counts, and ordering of activities to be evaluated as potential features for classification. We show that adding such structural features increases the amount of information thus potentially increasing classification accuracy. However, there is an inherent trade-off as using too many features leads to too long run-times for machine learning classification models. One way to improve the machine learning algorithms' run-time is to only select a small number of features by a feature selection algorithm. However, the run-time required by the feature selection algorithm must also be taken into account. Also, the classification accuracy should not suffer too much from the feature selection. The main contributions of this paper are as follows: First, we propose and compare six different feature selection algorithms by means of an experimental setup comparing their classification accuracy and achievable response times. Second, we discuss the potential use of feature selection results for computer assisted root cause analysis as well as the properties of different types of structural features in the context of feature selection., Comment: Extended version of a paper published in the proceedings of the BPM 2017 workshops

[EN] In this paper, we retrieve data about the frequent users of electronic commerce during the period 2011-2016 from the Spanish National Institute of Statistics. These data,coming from surveys, have intrinsic uncertainty that we describe using appropriate random variables. Then, we propose a stochastic model to study the dynamics of frequent users of electronic commerce. The goal of this paper is to solve the inverse problem that consists of determining themodel parameters as suitable parametric random variables, in such a way the model output be capable of capturing the data uncertainty, at the time instantswhere sample data are available, via adequate probability density functions. To achieve the aforementioned goal, we propose a computational procedure that involves building a nonlinear objective function, based on statistical moment measures, to be minimized using a variation of the particle swarm optimization algorithm., This work has been supported by the Ministerio de Economía, Industria y Competitividad Grant MTM2017-89664-P.

[EN] In this paper we summarise the content of the keynote that will be given at the 5th International Conference on Statistical Language and Speech Processing (SLSP) in Le Mans, France in October 23¿25, 2017. In the keynote we will address the importance of inferring demographic information for marketing and security reasons. The aim is to model how language is shared in gender and age groups taking into account its statistical usage. We will see how a shallow discourse analysis can be done on the basis of a graph-based representation in order to extract information such as how complicated the discourse is (i.e., how connected the graph is), how much interconnected grammatical categories are, how far a grammatical category is from others, how different grammatical categories are related to each other, how the discourse is modelled in different structural or stylistic units, what are the grammatical categories with the most central use in the discourse of a demographic group, what are the most common connectors in the linguistic structures used, etc. Moreover, we will see also the importance to consider emotions in the shallow discourse analysis and the impact that this has. We carried out some experiments for identifying gender and age, both in Spanish and in English, using PAN-AP-13 and PAN-PC-14 corpora, obtaining comparable results to the best performing systems of the PAN Lab at CLEF., The research work described in this paper was partially carried out in the framework of the SomEMBED project (TIN2015-71147-C2-1-P), funded by the Spanish Ministry of Economy, Industry and Competitiveness (MINECO).

High-Utility Itemset Mining (HUIM) is a relevant data mining task. The goal is to discover recurrent combinations of items characterized by high profit from transactional datasets. HUIM has a wide range of applications among which market basket analysis and service profiling. Based on the observation that items can be clustered into domain-specific categories, a parallel research issue is generalized itemset mining. It entails generating correlations among data items at multiple abstraction levels. The extraction of multiple-level patterns affords new insights into the analyzed data from different viewpoints. This paper aims at discovering a novel pattern that combines the expressiveness of generalized and High-Utility itemsets. According to a user-defined taxonomy items are first aggregated into semantically related categories. Then, a new type of pattern, namely the Generalized High-utility Itemset (GHUI), is extracted. It represents a combinations of items at different granularity levels characterized by high profit (utility). While profitable combinations of item categories provide interesting high-level information, GHUIs at lower abstraction levels represent more specific correlations among profitable items. A single-phase algorithm is proposed to efficiently discover utility itemsets at multiple abstraction levels. The experiments, which were performed on both real and synthetic data, demonstrate the effectiveness and usefulness of the proposed approach.

Concurrent Constraint Programming (CCP) is a declarative model for concurrency where agents interact by telling and asking constraints (pieces of information) in a shared store. Some previous works have developed (approximated) declarative debuggers for CCP languages. However, the task of debugging concurrent programs remains difficult. In this paper we define a dynamic slicer for CCP and we show it to be a useful companion tool for the existing debugging techniques. Our technique starts by considering a partial computation (a trace) that shows the presence of bugs. Often, the quantity of information in such a trace is overwhelming, and the user gets easily lost, since she cannot focus on the sources of the bugs. Our slicer allows for marking part of the state of the computation and assists the user to eliminate most of the redundant information in order to highlight the errors. We show that this technique can be tailored to timed variants of CCP. We also develop a prototypical implementation freely available for making experiments., Comment: Pre-proceedings paper presented at the 26th International Symposium on Logic-Based Program Synthesis and Transformation (LOPSTR 2016), Edinburgh, Scotland UK, 6-8 September 2016 (arXiv:1608.02534)

International audience; When two people engage in a conversation, knowingly or unknowingly, they are playing a game. Players of such games have diverse objectives, or winning conditions: an applicant trying to convince her potential employer of her eligibility over that of a competitor, a prosecutor trying to convict a defendant, a politician trying to convince an electorate in a political debate, and so on. We argue that infinitary games offer a natural model for many structural characteristics of such conversations. We call such games message exchange games, and we compare them to existing game theoretic frameworks used in linguistics—for example, signaling games—and show that message exchange games are needed to handle non-cooperative conversation. In this paper, we concentrate on conversational games where players’ interests are opposed. We provide a taxonomy of conversations based on their winning conditions, and we investigate some essential features of winning conditions like consistency and what we call rhetorical cooperativity. We show that these features make our games decomposition sensitive, a property we define formally in the paper. We show that this property has far-reaching implications for the existence of winning strategies and their complexity. There is a class of winning conditions (decomposition invariant winning conditions) for which message exchange games are equivalent to Banach- Mazur games, which have been extensively studied and enjoy nice topological results. But decomposition sensitive goals are much more the norm and much more interesting linguistically and philosophically.

In the near future the number of equipment connected to the Internet will greatly increase, so that further development of applications meant to verify their operations will be required. Monitoring represents an important factor in improving the quality of the services provided in cloud computing, given the fact that it allows scaling resource utilization in an adaptive manner. This paper aims to provide a solution for the monitoring of network devices and services, allowing administrator to verify connectivity of the equipment, their performances and network security. The main contribution of the paper consists in proposing an integrated solution that is deployed in the cloud for monitoring all the network components. Finally, the paper discusses the main findings and advantages for a reference implementation of the monitoring system using a simulated network.

© Springer International Publishing AG 2016. This paper improves the understanding of linear cryptanalysis by highlighting some previously overlooked aspects. It shows that linear hulls are sometimes formed already in a single round, and that overlooking such hulls may lead to a wrong estimation of the linear correlation, and thus of the data complexity. It shows how correlation matrices can be used to avoid this, and provides a tutorial on how to use them properly. By separating the input and output masks from the key mask it refines the formulas for computing the expected correlation and the expected linear potential. Finally, it shows that when the correlation of a hull is not properly estimated (e.g., by using the correlation of a single trail as the correlation of the hull), the success probability of Matsui’s Algorithm 1 drops, sometimes drastically. It also shows that when the trails composing the hull are properly accounted for, more than a single key bit can be recovered using Algorithm 1. All the ideas presented in this paper are followed by examples comparing previous methods to the corrected ones, and verified experimentally with reduced-round versions of Simon32/64. ispartof: pages:269-286 ispartof: Lecture Notes in Computer Science vol:10095 pages:269-286 ispartof: INDOCRYPT 2016 location:Kolkata, INDIA date:11 Dec - 14 Dec 2016 status: published

International audience; In the video analysis domain, automatic detection of actions performed in a recorded video represents an important scientific and industrial challenge. This paper presents a new method to approximate the boundaries of actions performed by a person while interacting with his environment (such as moving objects). This method relies on a Codebook quantization method to analyze the rough evolution of each pixel and then decide whether this evolution corresponds to an action or not; this decision is taken by an automated system. Statistics are then produced - at the scale of the whole frame - to estimate the start and the end of an action. According to our proposed evaluation protocol, this method produces interesting results on both real and simulated videos. This statistic-based protocol is discussed at the end of this paper. The interpretation of this evaluation protocol nominates this method to be a solid base to localize the exact boundaries of actions or - in the framework of this research activity - to associate prescriptive text with a visual content.

© International Association for Cryptologic Research 2015. LowMC is a collection of block cipher families introduced at Eurocrypt 2015 by Albrecht et al. Its design is optimized for instantiations of multi-party computation, fully homomorphic encryption, and zero-knowledge proofs. A unique feature of LowMC is that its internal affine layers are chosen at random, and thus each block cipher family contains a huge number of instances. The Eurocrypt paper proposed two specific block cipher families of LowMC, having 80-bit and 128-bit keys. In this paper, we mount interpolation attacks (algebraic attacks introduced by Jakobsen and Knudsen) on LowMC, and show that a practically significant fraction of 2−38 of its 80-bit key instances could be broken 223 times faster than exhaustive search. Moreover, essentially all instances that are claimed to provide 128-bit security could be broken about 1000 times faster. In order to obtain these results we optimize the interpolation attack using several new techniques. In particular, we present an algorithm that combines two main variants of the interpolation attack, and results in an attack which is more efficient than each one. ispartof: pages:535-560 ispartof: ADVANCES IN CRYPTOLOGY - ASIACRYPT 2015, PT II vol:9453 pages:535-560 ispartof: ASIACRYPT 2015 location:Auckland, NEW ZEALAND date:27 Nov - 3 Dec 2015 status: published

[EN] Agri-food production must increase while food waste needs to be reduced for improving the position of farmers. To do so it is necessary to sustainably manage agri-food supply chains beginning with the crop planning decisions. Although the centralized approach has usually been adopted for this purpose, it can lead to unfair solutions due to inequitable distribution of profits among farmers causing their unwillingness to collaborate in the implementation of decisions made. To solve this, in this paper a novel centralized multi-objective mathematical programming model is proposed to support the sustainable crop planning definition for a region that jointly optimize three objectives aligned to the sustainability aspects: supply chain profits maximization (economic objective), waste minimization (environmental objective) and unfairness among farmers minimization (social objective), being the last two objectives novel in the crop planning literature. It has also shown the conflicting nature of the three objectives finding trade-offs among them. Other novelties of this proposal are: (1) anticipation of operative decisions (such as harvest, transport, sale, clearance sale, waste and unmet demand) when defining the crop planning, (2) possibility of clearing the oversupply of crops as a means of increasing the farmers' profits and reducing waste, and (3) the modelling of a agri-food supply chain characterized by the lack of intermediaries between farmers and retailers, fostering the freshest product delivery and farmers' power position. The model is solved by applying the weighted sum method concluding that the crop waste generated along the chain and the unfairness among farmers can be considerably reduced by little decreasing the optimal SC profits., We acknowledge the support of the Project 691249, RUCAPS: "Enhancing and implementing knowledge based ICT solutions within high risk and uncertain conditions for agriculture production systems", funded by the European Union's research and innovation programme under the H2020 Marie Sklodowska-Curie Actions.

[EN] In this paper, an optimization tool based on a MILP model to support the teaching assignment process is proposed. It considers not only hierarchical issues among lecturers but also their preferences to teach a particular subject, the non-regular time schedules throughout the academic year, different type of credits, number of groups and other specific characteristics. Besides, it adds restrictions based on the time compatibility among the different subjects, the lecturers' availability, the maximum number of subjects per lecturer, the maximum number of lecturers per subject as well as the maximum and minimum saturation level for each lecturer, all of them in order to increase the teaching quality. Schedules heterogeneity and other features regarding the operation of some universities justify the usefulness of this model since no study that deals with all of them has been found in the literature review. Model validation has been performed with two real data sets collected from one academic year schedule at the Spanish University Universitat Politecnica de Valencia.

[EN] Internet of Things (IoT) has introduced new applications and environments. Smart Home provides new ways of communication and service consumption. In addition, Artificial Intelligence (AI) and deep learning have improved different services and tasks by automatizing them. In this field, reinforcement learning (RL) provides an unsupervised way to learn from the environment. In this paper, a new intelligent system based on RL and deep learning is proposed for Smart Home environments to guarantee good levels of QoE, focused on multimedia services. This system is aimed to reduce the impact on user experience when the classifying system achieves a low accuracy. The experiments performed show that the deep learning model proposed achieves better accuracy than the KNN algorithm and that the RL system increases the QoE of the user up to 3.8 on a scale of 10., This work has been partially supported by the "Ministerio de Economia y Competitividad" in the "Programa Estatal de Fomento de la Investigacion Cientifica y Tecnica de Excelencia, Subprograma Estatal de Generacion de Conocimiento" within the project under Grant TIN2017-84802-C2-1-P. This work has also been partially founded by the Universitat Polite`cnica de Vale`ncia through the postdoctoral PAID-10-20 program.

[EN] One of the problems when conducting research in mathematical programming models for operations planning is having an adequate database of experiments that can be used to verify advances and developments with enough factors to understand different consequences. This paper presents a test bed generator and instances database for a rolling horizons analysis for multiechelon planning, multiproduct with alternatives processes, multistroke, multicapacity with different stochastic demand patterns to be used with a stroke-like bill of materials considering production costs, setup, storage and delays for operations management. From the analysis of the operations planning obtained from this test bed, it is concluded that a product structure with an alternative process obtains the lowest total cost and the highest service level. In addition, decreasing seasonal demand could present a lower total cost than constant demand, but would generate a worse service level. This test bed will allow researchers further investigation so as to verify improvements in forecast methods, rolling horizons parameters, employed software, etc.

[EN] In this paper we go further with the study initiated by Behle, Krebs and Reifferscheid (in: Proceedings CAI 2011, Lecture Notes in Computer Science, vol 6742, pp 97-114, 2011), who gave an Eilenberg-type theorem for non-regular languages via typed monoids. We provide a new extension of that result, inspired by the one carried out by Pin in the regular case in 1995, who considered classes of languages not necessarily closed under complement. We introduce the so-called positively typed monoids, and give a correspondence between varieties of such algebraic structures and positive varieties of possibly non-regular languages. We also prove a similar result for classes of languages with weaker closure properties, The third author is supported by Proyecto PGC2018-096872-B-100-AR, Agencia Estatal de Investigacion (Spain), and by Proyecto Prometeo/2017/057, Generalitat Valenciana (Spain).

[EN] Virtual learning environments are growing in importance as fast as e-learning is becoming highly demanded by universities and students all over the world. This paper investigates how to automatically evaluate User eXperience in this domain using sentiment analysis techniques. For this purpose, a corpus with the opinions given by a total of 583 users (107 English speakers and 476 Spanish speakers) about three learning management systems in different courses has been built. All the collected opinions were manually labeled with polarity information (positive, negative or neutral) by three human annotators, both at the whole opinion and sentence levels. We have applied our state-of-the-art sentiment analysis models, trained with a corpus of a different semantic domain (a Twitter corpus), to study the use of cross-domain models for this task. Cross-domain models based on deep neural networks (convolutional neural networks, transformer encoders and attentional BLSTM models) have been tested. In order to contrast our results, three commercial systems for the same task (MeaningCloud, Microsoft Text Analytics and Google Cloud) were also tested. The obtained results are very promising and they give an insight to keep going the research of applying sentiment analysis tools on User eXperience evaluation. This is a pioneering idea to provide a better and accurate understanding on human needs in the interaction with virtual learning environments and a step towards the development of automatic tools that capture the feed-back of user perception for designing virtual learning environments centered in user's emotions, beliefs, preferences, perceptions, responses, behaviors and accomplishments that occur before, during and after the interaction., Partially supported by the Spanish MINECO and FEDER founds under Project TIN2017-85854-C4-2-R. Work of J.A. Gonzalez is financed under Grant PAID-01-17

[EN] Adapting the user interface of a software system to the requirements of the context of use continues to be a major challenge, particularly when users become more demanding in terms of adaptation quality. A considerable number of methods have, over the past three decades, provided some form of modelling with which to support user interface adaptation. There is, however, a crucial issue as regards in analysing the concepts, the underlying knowledge, and the user experience afforded by these methods as regards comparing their benefits and shortcomings. These methods are so numerous that positioning a new method in the state of the art is challenging. This paper, therefore, defines a conceptual reference framework for intelligent user interface adaptation containing a set of conceptual adaptation properties that are useful for model-based user interface adaptation. The objective of this set of properties is to understand any method, to compare various methods and to generate new ideas for adaptation. We also analyse the opportunities that machine learning techniques could provide for data processing and analysis in this context, and identify some open challenges in order to guarantee an appropriate user experience for end-users. The relevant literature and our experience in research and industrial collaboration have been used as the basis on which to propose future directions in which these challenges can be addressed., This work is supported by the Spanish Ministry of Science, Innovation, and Universities under Grant No.: TIN2017-84550-R, Adapt@Cloud Project and by the Generalitat Valenciana under Grant No.: AICO/2020/113, UX-Adapt Project. Arthur Sluyters is funded by the "Fonds de la Recherche Scientifique - FNRS" under Grant n40001931.

[EN] Program slicing is an analysis technique that has a wide range of applications, ranging from compilers to clone detection software, and that has been applied to practically all programming languages. Most program slicing techniques are based on a widely extended program representation, the System Dependence Graph (SDG). However, in the presence of unconditional jumps, there exist some situations where most SDG-based slicing techniques are not as accurate as possible, including more code than strictly necessary. In this paper, we identify one of these scenarios, pointing out the cause of the inaccuracy, and describing the initial solution to the problem proposed in the literature, together with an extension, which solves the problem completely. These solutions modify both the SDG generation and the slicing algorithm. Additionally, we propose an alternative solution, that solves the problem by modifying only the SDG generation, leaving the slicing algorithm untouched., This work has been partially supported by the EU (FEDER) and the Spanish MCI/AEI under grants TIN2016-76843-C4-1-R and PID2019-104735RB-C41, by the Generalitat Valenciana under grant Prometeo/2019/098 (DeepTrust), and by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No 952215

[EN] Sustainability practice within supply chains remains in an early development phase. Enterprises still need tools that support the integration of sustainability strategy into their activity, and to align their sustainability strategy with the supplier selection process. This paper proposes a methodology using a multi-criteria technique to support supplier selection decisions by taking two groups of inputs that integrate sustainability performance: supply chain performance and supplier assessment criteria. With the proposed methodology, organisations will have a tool to select suppliers based on their development towards sustainability and on their alignment with the supply chain strategy towards sustainability. The methodology is applied to an agri-food supply chain to assess sustainability in the supplier selection process., The authors of this publication acknowledge the contribution of Project GV/2017/065 'Development of a decision support tool for the management and improvement of sustainability in supply chains', funded by the Regional Valencian Government. Also, the authors acknowledge Project 691249, RUC-APS: Enhancing and implementing knowledge-based ICT solutions within high risk and uncertain conditions for agriculture production systems (www.ruc-aps.eu), funded by the European Union according to funding scheme H2020-MSCA-RISE-2015.

[EN] This paper introduces an open-source platform to support serverless computing for scientific data-processing workflow-based applications across the Cloud continuum (i.e. simultaneously involving both on-premises and public Cloud platforms to process data captured at the edge). This is achieved via dynamic resource provisioning for FaaS platforms compatible with scale-to-zero approaches that minimise resource usage and cost for dynamic workloads with different elasticity requirements. The platform combines the usage of dynamically deployed auto-scaled Kubernetes clusters on on-premises Clouds and automated Cloud bursting into AWS Lambda to achieve higher levels of elasticity. A use case in public health for smart cities is used to assess the platform, in charge of detecting people not wearing face masks from captured videos. Faces are blurred for enhanced anonymity in the on-premises Cloud and detection via Deep Learning models is performed in AWS Lambda for this data-driven containerised workflow. The results indicate that hybrid workflows across the Cloud continuum can efficiently perform local data processing for enhanced regulations compliance and perform Cloud bursting for increased levels of elasticity., The authors would like to thank the European Union for the project "Artificial Intelligence in Secure PRIvacy-preserving computing coNTinuum" (AI-SPRINT), with code 101016577, funded under the H2020 Framework Programme and also the regional government of the Comunitat Valenciana (Spain) for the project IDIFEDER/2018/032 (High-Performance Algorithms for the Modeling, Simulation and early Detection of diseases in Personalized Medicine), co-funded by the European Union ERDF funds (European Regional Development Fund) of the Comunitat Valenciana 2014-2020.

[EN] Organisations are concerned about measuring the performance of the product/service they deliver to their customers. In all types of organisations, if a proper performance assessment is to be developed, it should be measured in different dimensions. At University, the new study programs include the development and assessment of transversal competences due to their importance in enhancing the abilities and improving the employability of students. The achievement of transversal competences can be assessed in different levels/stages; for example, the 1st and 2nd years of a Bachelor's degree; the 3rd and 4th years of a Bachelor's degree and at Master's level. The purpose of this paper is to integrate the research into performance management in organisations to develop an approach consisting of four components (a methodology, a solid and integrated performance management framework, graphical diagrams and quantitative techniques) to assess and manage the achievement of transversal competences through the different levels of study using a consolidated approach. The proposal uses the Analytic Network Process (ANP) to model dependences and feedback among the elements of the competences., This work has been developed within the research projects "Use of software applications for training and assessing the UPV transversal competence "effective communication" in bachelor and master" (Ref. PIME/2016/A/16) and "Development of a social network for the design, follow-up, assessment and multi-disciplinary coordination of the TFG and TFM" (Ref.: PIME/2017/A/018) funded by the Vice-Rectorate for Studies, Quality and Accreditation at Universitat Politecnica de Valencia.

[EN] This paper presents FENOCOP, a game-theoretic approach for solving non-cooperative planning problems that involve a set of self-interested agents. Each agent wants to execute its own plan in a shared environment but the plans may be rendered infeasible by the appearance of potential conflicts; agents are willing to coordinate their plans in order to avoid conflicts during a joint execution. In order to attain a conflict-free combination of plans, agents must postpone the execution of some of their actions, which negatively affects their individual utilities. FENOCOP is a two-level game approach: the General Game selects a Nash equilibrium among several combinations of plans, and the Scheduling Game generates, for a combination of plans, an executable outcome by introducing delays in the agents¿ plans. For the Scheduling Game, we developed two algorithms that return a Pareto optimal and fair equilibrium from which no agent would be willing to deviate., This work is supported by the Spanish MINECO project TIN2017-88476-C2-1-R. Jaume Jordan is funded by grant APOSTD/2018/010 of Generalitat Valenciana - Fondo Social Europeo and by UPV PAID-06-18 project.

[EN] The problem of automatic text classification is an essential part of text analysis. The improvement of text classification can be done at different levels such as a preprocessing step, network implementation, etc. In this paper, we focus on how the combination of different methods of text encoding may affect classification accuracy. To do this, we implemented a multi-input neural network that is able to encode input text using several text encoding techniques such as BERT, neural embedding layer, GloVe, skip-thoughts and ParagraphVector. The text can be represented at different levels of tokenised input text such as the sentence level, word level, byte pair encoding level and character level. Experiments were conducted on seven datasets from different language families: English, German, Swedish and Czech. Some of those languages contain agglutinations and grammatical cases. Two out of seven datasets originated from real commercial scenarios: (1) classifying ingredients into their corresponding classes by means of a corpus provided by Northfork; and (2) classifying texts according to the English level of their corresponding writers by means of a corpus provided by ProvenWord. The developed architecture achieves an improvement with different combinations of text encoding techniques depending on the different characteristics of the datasets. Once the best combination of embeddings at different levels was determined, different architectures of multi-input neural networks were compared. The results obtained with the best embedding combination and best neural network architecture were compared with state-of-the-art approaches. The results obtained with the dataset used in the experiments were better than the state-of-the-art baselines., This work is partially supported by MINECO under Grant DI-15-08169 and by Sciling under its R+D program. The authors would like to thank NVIDIA for their donation of a Titan Xp GPU, which allowed us to conduct this research

[EN] Assessing the ability of applicants to repay their loans is generally recognized as a critical task in credit risk management. Credit managers rely on financial and market information, usually in the form of ratios, to estimate the quality of credit applicants. However, there is no guarantee that a given set of ratios contains the information needed for credit classification. Decision rules under strict uncertainty aim to mitigate this drawback. In this paper, we propose the use of a moderate pessimism decision rule combined with dimensionality reduction techniques and compromise programming. Moderate pessimism ensures that neither extreme optimistic nor pessimistic decisions are taken. Dimensionality reduction from a set of ratios facilitates the extraction of the relevant information. Compromise programming allows to find a balance between quality of debt and risk concentration. Our model produces two critical outputs: a quality assessment and the optimum allocation of funds. To illustrate our multicriteria approach, we include a case study on 29 firms listed in the Spanish stock market. Our results show that dimensionality reduction contributes to avoid redundancy and that quality-diversification optimization is able to produce budget allocations with a reduced number of firms.

[EN] To satisfy their main goal, namely providing quality water to consumers, water distribution networks (WDNs) need to be suitably monitored. Only well designed and reliable monitoring data enables WDN managers to make sound decisions on their systems. In this belief, water utilities worldwide have invested in monitoring and data acquisition systems. However, good monitoring needs optimal sensor placement and presents a multi-objective problem where cost and quality are conflicting objectives (among others). In this paper, we address the solution to this multi-objective problem by integrating quality simulations using EPANET-MSX, with two optimization techniques. First, multi-objective optimization is used to build a Pareto front of non-dominated solutions relating contamination detection time and detection probability with cost. To assist decision makers with the selection of an optimal solution that provides the best trade-off for their utility, a multi-criteria decision-making technique is then used with a twofold objective: 1) to cluster Pareto solutions according to network sensitivity and entropy as evaluation parameters; and 2) to rank the solutions within each cluster to provide deeper insight into the problem when considering the utility perspectives.The clustering process, which considers features related to water utility needs and available information, helps decision makers select reliable and useful solutions from the Pareto front. Thus, while several works on sensor placement stop at multi-objective optimization, this work goes a step further and provides a reduced and simplified Pareto front where optimal solutions are highlighted. The proposed methodology uses the NSGA-II algorithm to solve the optimization problem, and clustering is performed through ELECTRE TRI. The developed methodology is applied to a very well-known benchmarking WDN, for which the usefulness of the approach is shown. The final results, which correspond to four optimal solution clusters, are useful for decision makers during the planning and development of projects on networks of quality sensors. The obtained clusters exhibit distinctive features, opening ways for a final project to prioritize the most convenient solution, with the assurance of implementing a Pareto-optimal solution.

In Panama City, the water supply sources for the population come mainly from surface waters, but in times of drought and extreme natural phenomena, the need to resort to underground sources is explored. This paper concerns an exploratory investigation with the aim of assessing the quality of groundwater in the southeastern part of the Province of Panama and to determine its possible use as drinking water during extreme natural phenomena. Monthly monitoring over a period of 15 months was carried out. The study included the sampling and analysis of groundwater by different physic-chemical, hydrochemical and biological parameters. The results showed that most of the parameters in the analysed groundwater conformed to the Panamanian drinking water norm and the water potability diagram, this is a convenient source of raw water for purification. As for other uses for this water source, it should be used with caution since it represents a medium risk of soil salinization for irrigation water; and it is a moderately hard water for industrial uses.

Forecasting is an important task for intelligent agents involved in dynamical processes. A specific application domain concerns district heating networks, in which the future heating load generated by centralized power plants and distributed to buildings must be optimized for better plant maintenance, energy consumption and environmental impact. In this paper we present XM_HeatForecast a Python tool designed to support district heating network operators. The tool provides an integrated architecture for i) generating and updating in real-time predictive models of heating load, ii) supporting the analysis of prediction performance and errors, iii) inspecting model parameters and analyzing the historical dataset from which models are trained. A case study is presented in which the software is used on a synthetic dataset of heat loads and weather forecast from which a regression model is generated and updated every 24 h, while predictions of load in the next 48 h are performed every hour.