Search

Recommender systems are among the most widely used applications of artificial intelligence. Since they are so widely used, it is important that we, as practitioners and researchers, think about the impact these systems may have on users, society, and other stakeholders. To that effect, the NORMalize workshop seeks to introduce normative thinking, to consider the norms and values that underpin recommender systems in the recommender systems community. The objective of NORMalize is to bring together a growing community of researchers and practitioners across disciplines who want to think about the norms and values that should be considered in the design and evaluation of recommender systems; and further educate them on how to reflect on, prioritise, and operationalise such norms and values. NORMalize offers a comprehensive program designed to cater to both the norm-curious and the norm-active. The morning session is on-site and features a lecture on normative thinking and an interactive workshop. The afternoon is a hybrid program focused on the dissemination of results. NORMalize publishes proceedings, as well as a technical report that summarises the outcomes of the interactive morning session.

Denitrifying woodchip bioreactors (DWBs) have proven to be an efficient nature-based solution for nitrate removal. Modeling DWBs is required for improving their design and operation, but is hindered by the complexity of the modeled system where numerous chemical species and model parameters are needed. Reactions inside the woodchips are different from those at the edges, causing chemical localization (i.e., apparent simultaneous occurrence of incompatible reactions). We used the Multi Rate Mass Transfer (MRMT) approach to overcome these problems when simulating reactive transport processes in a DWB located at Kiruna, Sweden. Besides denitrification, other nitrogen-cycling processes (e.g., nitrification, dissimilatory nitrate reduction to ammonium, anammox) and alternative electron donors (e.g. oxygen, sulfate) were also considered. Biomass concentration is incorporated into the biochemical reaction rates, including growth and decay, to characterize microbial catalysis. We found that the MRMT model: 1) can account for the heterogeneity of the porous woodchips; 2) was capable of reproducing the nitrogen species evolution in the DWB with kinetic parameters from the literature; and 3) allows reproducing localized biochemical reactions (e.g., aerobic reactions on the woodchip edges, near the DWB entrance and anaerobic reactions inside); and 4) reproduces the full denitrification reactions sequence, but with the different reactions occurring in different portions of the woodchip (e.g., nitrate to nitrite near the edges and nitrite to nitrous oxide further inside). The latter observation suggests that increasing woodchip size may reduce the outflow of these undesired species., The authors acknowledge the financial support of projects NITREM, RESTORA (Agència Catalana de l'Aigua, ACA210/18/0040), and ConMimo (Spanish Research Agency, AEI, number TED2021-131188B-C31). NITREM (project number 17013) has received funding from the European Institute of Innovation and Technology (EIT). This body of the European Union received support from the European Union's Horizon 2020 research and innovation programme. Additional funding was obtained from the Generalitat de Catalunya (2017 SGR1485) and the Spanish Ministry of Science and Innovation (Centre of Excellence Severo-Ochoa, CSIC-IDAEA, CEX2018-000794-S)., Peer Reviewed, Postprint (published version)

Many actual images contain partial body camera shots, in which a significant part of the body is not visible. This issue is especially prevalent in film images, where less than 10% are full-body shots. Most 2D human pose estimation methods return incomplete poses when applied to partial body images. This lack of completeness becomes a problem in some situations, for example, when the 2D pose is converted to a 3D pose by a two-stage 3D human pose estimation method since most of these methods require a complete pose to work. This article proposes a new technique, called CompletePose, consisting of completing the missing keypoints when 2D human pose estimation methods are applied to images with partial body camera shots. A Conditional Generative Adversarial Network is used to obtain a complete and plausible pose, realistic enough to predict a 3D pose with a two-stage 3D human pose estimation method. A complete empirical validation has been carried out with the Human3.6 M dataset and a new dataset, called CHARADE, specially built and made public for reproducibility and benchmarking for this research. Quantitative evaluation employing the Fréchet distance shows that the approach manages to approximate the actual data distribution. The qualitative evaluation shows that the completed poses enable obtaining plausible 3D poses from images previously intractable., This work is partially supported by the Spanish Ministry of Science and Innovation under contracts PID2019-107255GB, PLEC2021-007850 and RTI2018-095232-B-C21, and by the SGR programme of the Catalan Government under contracts 2021-SGR00478, 2017-SGR-1414 and 2017-SGR-1742., Peer Reviewed, Postprint (author's final draft)

Denitrifying woodchip bioreactors (DWBs) have proven to be an efficient nature-based solution for nitrate removal. Modeling DWBs is required for improving their design and operation, but is hindered by the complexity of the modeled system where numerous chemical species and model parameters are needed. Reactions inside the woodchips are different from those at the edges, causing chemical localization (i.e., apparent simultaneous occurrence of incompatible reactions). We used the Multi Rate Mass Transfer (MRMT) approach to overcome these problems when simulating reactive transport processes in a DWB located at Kiruna, Sweden. Besides denitrification, other nitrogen-cycling processes (e.g., nitrification, dissimilatory nitrate reduction to ammonium, anammox) and alternative electron donors (e.g. oxygen, sulfate) were also considered. Biomass concentration is incorporated into the biochemical reaction rates, including growth and decay, to characterize microbial catalysis. We found that the MRMT model: 1) can account for the heterogeneity of the porous woodchips; 2) was capable of reproducing the nitrogen species evolution in the DWB with kinetic parameters from the literature; and 3) allows reproducing localized biochemical reactions (e.g., aerobic reactions on the woodchip edges, near the DWB entrance and anaerobic reactions inside); and 4) reproduces the full denitrification reactions sequence, but with the different reactions occurring in different portions of the woodchip (e.g., nitrate to nitrite near the edges and nitrite to nitrous oxide further inside). The latter observation suggests that increasing woodchip size may reduce the outflow of these undesired species.

Recommender systems are among the most widely used applications of artificial intelligence. Because of their widespread use, it is important that practitioners and researchers think about the impact they may have on users, society, and other stakeholders. To that effect, the NORMalize workshop seeks to introduce normative thinking, to consider the norms and values that underpin recommender systems in the recommender systems community. The objective of NORMalize is to bring together a growing community of researchers and practitioners across disciplines who want to think about the norms and values that should be considered in the design and evaluation of recommender systems, and further educate them on how to reflect on, prioritise, and operationalise such norms and values. This document is a report on the first workshop, co-located with ACM RecSys'23 in Singapore.

Phase-field models describe the motion of multiphase flows using smooth interfaces across which the composition changes continuously. The phase-field variable represents a measure of phase as it quantifies relative differences or fractions of the fluid s components. The Cahn-Hilliard equation was originally proposed to model spinodal decomposition and coarsening in binary alloys. To this day, it has become broad ranged in its applicability. This thesis focuses on solving the Cahn-Hilliard equation numerically. A review of the mathematical modelling is made in order to develop numerical methods. Different numerical simulations in two dimensions are implemented to study the numerical and physical properties. Two realistic physical examples are also numerically solved.

[CATALÀ] S'han implementat mètodes d'SDC en quatre filtres de privacitat pel software MOA. Els algorismes han estat adaptats de solucions conegudes per habilitar el seu ús en entorns de processament de fluxos. Finalment, han estat avaluats en termes del risc de revelació i la pèrdua d'informació., [ANGLÈS] Four MOA privacy-preserving filters have been developed to implement some SDC methods. The algorithms have been adapted from well-known solutions to enable their use in streaming settings. Finally, they have been benchmarked to assess their quality in terms of disclosure risk and information loss.

Statistical Disclosure Control (SDC, for short) studies the problem of privacy-preserving data publishing in cases where the data is expected to be used for statistical analysis. An original dataset T containing sensitive information is transformed into a sanitized version T' which is released to the public. Both utility and privacy aspects are very important in this setting. For utility, T' must allow data miners or statisticians to obtain similar results to those which would have been obtained from the original dataset T. For privacy, T' must significantly reduce the ability of an adversary to infer sensitive information on the data subjects in T. One of the main a-posteriori measures that the SDC community has considered up to now when analyzing the privacy offered by a given protection method is the Distance-Based Record Linkage (DBRL) risk measure. In this work, we argue that the classical DBRL risk measure is insufficient. For this reason, we introduce the novel Global Distance-Based Record Linkage (GDBRL) risk measure. We claim that this new measure must be evaluated alongside the classical DBRL measure in order to better assess the risk in publishing T' instead of T. After that, we describe how this new measure can be computed by the data owner and discuss the scalability of those computations. We conclude by extensive experimentation where we compare the risk assessments offered by our novel measure as well as by the classical one, using well-known SDC protection methods. Those experiments validate our hypothesis that the GDBRL risk measure issues, in many cases, higher risk assessments than the classical DBRL measure. In other words, relying solely on the classical DBRL measure for risk assessment might be misleading, as the true risk may be in fact higher. Hence, we strongly recommend that the SDC community considers the new GDBRL risk measure as an additional measure when analyzing the privacy offered by SDC protection algorithms., Postprint (author's final draft)

[ANGLÈS] Nowadays, social network services play a very important role in the way people interact with each other and with the world. This generates big amounts of data that can be used to study social relationships and extract useful information about preferences and trends. When analysing this information, two main problems emerge: The need to aggregate dif- ferent data coming from multiple sources, and hardware limitations due to the incapability traditional systems have to deal with large amounts of data. In order to solve the problems mentioned before, this project aims to implement a distributed, scalable social media analysis tool, ready to connect and gather data from multiple sources and show the aggregated results in real-time., [CASTELLÀ] Hoy en día, las redes sociales juegan un papel muy importante en la manera como las personas interactúa entre ellos y con el mundo. Esto genera grandes volúmenes de inforación que pueden ser utilizados para estudiar las relaciones sociales y extraer información útil acerca de gustos y tendencias. Cuando se analiza esta información, surgen dos problemas principales: La necesidad de agregar diferentes datos provenientes de múltiples fuentes, y las limitaciones hardware por la incapacidad de los sistemas tradicionales de manejar grandes cantidades de datos. Para poder solventar estos problemas, este proyecto propone implementar una herramienta de análisis de redes sociales distribuida y escalable, preparada para conectarse y recolectar datos de múltiples fuentes y mostrar los resultados agregados en tiempo real., [CATALÀ] Avui en dia, les xarxes socials juguen un paper molt important en la manera com les persones interactua entre ells i amb el mon. Això genera grans quantitats de dades que poden ser utilitzats per estudiar les relacions socials i extreure informació útil sobre gustos i tendències. 2 Quan s’analitza aquesta informació, sorgeixen dos problemes principals: La necessitat de agregar diferents dades provinents de múltiples fonts, i les limitacions hardware per la incapaci- tat dels sistemes tradicionals de gestionar grans quantitats de dades. Per poder solucionar aque- sts problemes, aquest projecte proposa implementar una eina d’anàlisi de xarxes socials dis- tribuïda i escalable, preparada per connectar-se i recol·lectar dades de múltiples fonts i mostrar els resultats agregats en temps real.

[CASTELLÀ] Foodstagram es una aplicación que permite al usuario visualizar en un mapa fotografías de restaurantes de Cataluña. Las fotos que se muestran son capturadas en tiempo real de las enviadas por los usuarios de Instagram. Estas fotos pasarán una serie de filtros para ser seleccionadas., [ANGLÈS] Foodstagram is a appication that allows the user view on a map photos from restaurants of Catalonia. Shown photos are captured in real time from the Instagram users. These photos will go through a series of filters for be selected.

The increasing adoption of crowdsourcing for commercial and industrial purposes rises the need for creating sophisticated mechanisms in crowd-based digital platforms for efficient worker management. One of the main challenges in this area is worker motivation and skill set control and its impact on the output quality. The quality delivered by the workers in the crowd depends on different aspects such as their skills, experience, commitment, etc. The lack of generic and detailed proposals to incentive workers and the need for creating ad-hoc solutions depending on the domain make it difficult to evaluate the best rewarding functions in each scenario. In this paper, we make a step further in this direction and propose the use of aggregation functions to evaluate the professional skills of crowd-workers based on the quality of their past tasks. Additionally, we present a real industrial crowdsourcing solution for software localisation in which the proposed solutions are put into practice with real text translations quality measures., Peer Reviewed, Postprint (author’s final draft)

Public transport optimisation is becoming everyday a more di cult and challenging task, because of the increasing number of transportation options as well as the exponential increase of users. Many research contributions about this issue have been recently published under the umbrella of the smart cities research. In this work, we sketch a possible framework to optimize the tourist bus in the city of Barcelona. Our framework will extract information from Twitter and other web services, such as Foursquare to infer not only the most visited places in Barcelona, but also the trajectories and routes that tourist follow. After that, instead of using complex geospatial or trajectory clustering methods, we propose to use simpler clustering techniques as k-means or DBScan but using a real sequence of symbols as a distance measure to incorporate in the clustering process the trajectory information., This work is partially supported by the Ministry of Science and Technology of Spain under contract TIN2012-34557 and by the BSC-CNS Severo Ochoa program (SEV-2011-00067) and with the support of ACC1Ó, the Catalan Agency to promote applied research and innovation; and by the Spanish Centre for Development of Industrial Technology under the INNPRONTA program, project IPT-20111006, \CIUDAD2020"., Postprint (author’s final draft)

In the last years, with the increase of the available data from social networks and the rise of big data technologies, social data has emerged as one of the most profitable market for companies to increase their benefits. Besides, social computation scientists see such data as a vast ocean of information to study modern human societies. Nowadays, enterprises and researchers are developing their own mining tools in house, or they are outsourcing their social media mining needs to specialised companies with its consequent economical cost. In this paper, we present the first cloud computing service to facilitate the deployment of social media analytics applications to allow data practitioners to use social mining tools as a service. The main advantage of this service is the possibility to run different queries at the same time and combine their results in real time. Additionally, we also introduce twearch, a prototype to develop twitter mining algorithms as services in the cloud., Peer Reviewed, Postprint (author’s final draft)

[CATALÀ] Desenvolupament d'una aplicació web que serveixi com a eina d'anàlisi sentimental a través de la xarxa social Twitter per comprendre d'una manera general quina és la opinió del públic sobre un producte, una marca, un esdeveniment o qualsevol concepte., [ANGLÈS] Development of a web application that serves as a sentiment analysis tool using Twitter data, aiming to give a better understanding of what the users think of a product, brand, event or any other subject.

Nowadays, a huge amount of high resolution satellite images are freely available. Such images allow researchers in environmental sciences to study the different natural habitats and farming practices in a remote way. However, satellite images content strongly depends on the season of the acquisition. Due to the periodicity of natural and agricultural dynamics throughout seasons, sequential patterns arise as a new opportunity to model the behaviour of these environments. In this paper, we describe some preliminary results obtained with a new framework for studying spatiotemporal evolutions over natural and agricultural areas using k-partite graphs and sequential patterns extracted from segmented Landsat images., Postprint (author’s final draft)

Microaggregation is one of the most commonly employed microdata protection methods. The basic idea of microaggregation is to anonymize data by aggregating original records into small groups of at least k elements and, therefore, preserving k -anonymity. Usually, in order to avoid information loss, when records are large, i.e., the number of attributes of the data set is large, this data set is split into smaller blocks of attributes and microaggregation is applied to each block, successively and independently. This is called multivariate microaggregation. By using this technique, the information loss after collapsing several values to the centroid of their group is reduced. Unfortunately, with multivariate microaggregation, the k -anonymity property is lost when at least two attributes of different blocks are known by the intruder, which might be the usual case. In this work, we present a new microaggregation method called one dimension microaggregation ( Mic1D-k ). With Mic1D-k , the problem of k -anonymity loss is mitigated by mixing all the values in the original microdata file into a single non-attributed data set using a set of simple pre-processing steps and then, microaggregating all the mixed values together. Our experiments show that, using real data, our proposal obtains lower disclosure risk than previous approaches whereas the information loss is preserved., Postprint (author’s final draft)

Anonymization techniques that provide k-anonymity suffer from loss of quality when data dimensionality is high. Microaggregation techniques are not an exception. Given a set of records, attributes are grouped into non-intersecting subsets and microaggregated independently. While this improves quality by reducing the loss of information, it usually leads to the loss of the k-anonymity property, increasing entity disclosure risk. In spite of this, grouping attributes is still a common practice for data sets containing a large number of records. Depending on the attributes chosen and their correlation, the amount of information loss and disclosure risk vary. However, there have not been serious attempts to propose a way to find the best way of grouping attributes. In this paper, we present GOMM, the Genetic Optimizer for Multivariate Microaggregation which, as far as we know, represents the first proposal using evolutionary algorithms for this problem. The goal of GOMM is finding the optimal, or near-optimal, attribute grouping taking into account both information loss and disclosure risk. We propose a way to map attribute subsets into a chromosome and a set of new mutation operations for this context. Also, we provide a comprehensive analysis of the operations proposed and we show that, after using our evolutionary approach for different real data sets, we obtain better quality in the anonymized data comparing it to previously used ad-hoc attribute grouping techniques. Additionally, we provide an improved version of GOMM called D-GOMM where operations are dynamically executed during the optimization process to reduce the GOMM execution time., Postprint (author’s final draft)

Let us consider the following situation: t entities (e.g., hospitals) hold different databases containing different records for the same type of confidential (e.g., medical) data. They want to deliver a protected version of this data to third parties (e.g., pharmaceutical researchers), preserving in some way both the utility and the privacy of the original data. This can be done by applying a statistical disclosure control (SDC) method. One possibility is that each entity protects its own database individually, but this strategy provides less utility and privacy than a collective strategy where the entities cooperate, by means of a distributed protocol, to produce a global protected dataset. In this paper, we investigate the problem of distributed protocols for SDC protection methods. We propose a simple, efficient and secure distributed protocol for the specific SDC method of rank shuffling. We run some experiments to evaluate the quality of this protocol and to compare the individual and collective strategies for solving the problem of protecting a distributed database. With respect to other distributed versions of SDC methods, the new protocol provides either more security or more efficiency, as we discuss through the paper., Peer Reviewed, Postprint (published version)

Digital watermarking is the process of embedding information into a noise-tolerant digital signal such as image or audio data. Such information is embedded in a way difficult to be removed in order to easily identify relevant information for many different purposes, such as the copyright ownership of the media, source tracking, piracy deterrence, etc. There is an extensive literature about watermarking algorithms and methods as well as possible attack techniques. In this work we collect a part of this vast literature in order to make easier for a non-expert reader about watermarking to have a high-level overview on new trends and technologies related to multimedia watermark algorithms and attacks., Peer Reviewed, Postprint (author’s final draft)

We present an application of the Hungarian Method, an optimal assignment graph theory algorithm, to record linkage in order to improve the disclosure risk assessment. We should note that Hungarian Method has O(n^3) complexity; three different methods are presented to reduce its computational cost.

Abiquo Provisioning Portal millora la gestió dels entorns cloud privats i permet gestionar centres de dades de forma dinàmica i escalable, facilitant l'aprovisionament d'infraestructures virtuals privades controlades pels clients finals, llestes per treballar en producció, i del software requerit.

Aplicació de geolocalització per a dispositius iOS i Android que permet a l'usuari trobar establiments propers on comprar pa amb Indicació Geográfica Protegida (IGP) Pa de Pagés Català.