Search

Reports of long-lasting coronavirus disease 2019 (COVID-19) symptoms, the so-called ‘long COVID’, are rising but little is known about prevalence, risk factors or whether it is possible to predict a protracted course early in the disease. We ana- lyzed data from 4,182 incident cases of COVID-19 in which individuals self-reported their symptoms prospectively in the COVID Symptom Study app1. A total of 558 (13.3%) partici- pants reported symptoms lasting ≥28 days, 189 (4.5%) for ≥8 weeks and 95 (2.3%) for ≥12 weeks. Long COVID was characterized by symptoms of fatigue, headache, dyspnea and anosmia and was more likely with increasing age and body mass index and female sex. Experiencing more than five symptoms during the first week of illness was associated with long COVID (odds ratio = 3.53 (2.76–4.50)). A simple model to distinguish between short COVID and long COVID at 7 days (total sample size, n = 2,149) showed an area under the curve of the receiver operating characteristic curve of 76%, with replication in an independent sample of 2,472 individuals who were positive for severe acute respiratory syndrome coronavi- rus 2. This model could be used to identify individuals at risk of long COVID for trials of prevention or treatment and to plan education and rehabilitation services.

Recent technological advances have radically changed the way we communicate. Today’s communication has become ubiquitous and it has fostered the need for information that is easier to create, spread and consume. As a consequence, we have experienced the shortening of text messages in mediums ranging from electronic mailing, instant messaging to microblogging. Moreover, the ubiquity and fast-paced nature of these mediums have promoted their use for unthinkable tasks. For instance, reporting real-world events was classically carried out by news reporters, but, nowadays, most interesting events are first disclosed on social networks like Twitter by eyewitness through short text messages. As a result, the exploitation of the thematic content in short text has captured the interest of both research and industry. Topic models are a type of probability models that have traditionally been used to explore this thematic content, a.k.a. topics, in regular text. Most popular topic models fall into the sub-class of LVMs (Latent Variable Models), which include several latent variables at the corpus, document and word levels to summarise the topics at each level. However, classical LVM-based topic models struggle to learn semantically meaningful topics in short text because the lack of co-occurring words within a document hampers the estimation of the local latent variables at the document level. To overcome this limitation, pooling and hierarchical Bayesian strategies that leverage on contextual information have been essential to improve the quality of topics in short text. In this thesis, we study the problem of learning semantically meaningful and predictive representations of text in two distinct phases: • In the first phase, Part I, we investigate the use of LVM-based topic models for the specific task of event detection in Twitter. In this situation, the use of contextual information to pool tweets together comes naturally. Thus, we first extend an existing clustering algori, Els avenços tecnològics han canviat radicalment la forma que ens comuniquem. Avui en dia, la comunicació és ubiqua, la qual cosa fomenta l’ús de informació fàcil de crear, difondre i consumir. Com a resultat, hem experimentat l’escurçament dels missatges de text en diferents medis de comunicació, des del correu electrònic, a la missatgeria instantània, al microblogging. A més de la ubiqüitat, la naturalesa accelerada d’aquests medis ha promogut el seu ús per tasques fins ara inimaginables. Per exemple, el relat d’esdeveniments era clàssicament dut a terme per periodistes a peu de carrer, però, en l’actualitat, el successos més interessants es publiquen directament en xarxes socials com Twitter a través de missatges curts. Conseqüentment, l’explotació de la informació temàtica del text curt ha atret l'interès tant de la recerca com de la indústria. Els models temàtics (o topic models) són un tipus de models de probabilitat que tradicionalment s’han utilitzat per explotar la informació temàtica en documents de text. Els models més populars pertanyen al subgrup de models amb variables latents, els quals incorporen varies variables a nivell de corpus, document i paraula amb la finalitat de descriure el contingut temàtic a cada nivell. Tanmateix, aquests models tenen dificultats per aprendre la semàntica en documents curts degut a la manca de coocurrència en les paraules d’un mateix document, la qual cosa impedeix una correcta estimació de les variables locals. Per tal de solucionar aquesta limitació, l’agregació de missatges segons el context i l’ús d’estratègies jeràrquiques Bayesianes són essencials per millorar la qualitat dels temes apresos. En aquesta tesi, estudiem en dos fases el problema d’aprenentatge d’estructures semàntiques i predictives en documents de text: En la primera fase, Part I, investiguem l’ús de models temàtics amb variables latents per la detecció d’esdeveniments a Twitter. En aquest escenari, l’ús del context per agregar tweets sorgeix de forma, Postprint (published version)

The geographical identification of content in Social Networks have enabled to bridge the gap between online social platforms and the physical world. Although vast amounts of data in such networks are due to breaking news or global occurrences, local events witnessed by users in situ are also present in these streams and of great importance for many city entities. Nowadays, unsupervised machine learning techniques, such as Tweet-SCAN, are able to retrospectively detect these local events from tweets. However, these approaches have limited abilities to reason about unseen observations in a principled way due to the lack of a proper probabilistic foundation. Probabilistic models have also been proposed for the task, but their event identification capabilities are far from those of Tweet-SCAN. In this paper, we identify two key factors which, when combined, boost the accuracy of such models. As a first key factor, we notice that the large amount of meaningless social data requires explicitly modeling non-event observations.Therefore, we propose to incorporate a background model that captures spatio-temporal fluctuations of non-event tweets. As a second key factor, we observe that the shortness of tweets hampers the application of traditional topic models. Thus, we integrate event detection and topic modeling, assigning topic proportions to events instead of assigning them to individual tweets. As a result, we propose Warble, a new probabilistic model and learning scheme for retrospective event detection that incorporates these two key factors. We evaluate Warble in a data set of tweets located in Barcelona during its festivities. The empirical results show that the model outperforms other state-of-the-art techniques in detecting various types of events while relying on a principled probabilistic framework that enables to reason under uncertainty., This work is partially supported by Obra Social “la Caixa”, by the Spanish Ministry of Science and Innovation under contract (TIN2015-65316), by the Severo Ochoa Program (SEV2015-0493), by SGR programs of the Catalan Government (2014-SGR-1051, 2014-SGR-118), Collectiveware (TIN2015-66863-C2-1-R) and BSC/UPC NVIDIA GPU Center of Excellence.We would also like to thank the reviewers for their constructive feedback., Peer Reviewed, Postprint (author's final draft)

Twitter has become one of the most popular Location-based Social Networks (LBSNs) that bridges physical and virtual worlds. Tweets, 140-character-long messages, are aimed to give answer to the What’s happening? question. Occurrences and events in the real life (such as political protests, music concerts, natural disasters or terrorist acts) are usually reported through geo-located tweets by users on site. Uncovering event-related tweets from the rest is a challenging problem that necessarily requires exploiting different tweet features. With that in mind, we propose Tweet-SCAN, a novel event discovery technique based on the popular density-based clustering algorithm called DBSCAN. Tweet-SCAN takes into account four main features from a tweet, namely content, time, location and user to group together event-related tweets. The proposed technique models textual content through a probabilistic topic model called Hierarchical Dirichlet Process and introduces Jensen–Shannon distance for the task of neighborhood identification in the textual dimension. As a matter of fact, we show Tweet-SCAN performance in two real data sets of geo-located tweets posted during Barcelona local festivities in 2014 and 2015, for which some of the events were identified by domain experts beforehand. Through these tagged data sets, we are able to assess Tweet-SCAN capabilities to discover events, justify using a textual component and highlight the effects of several parameters., Peer Reviewed, Postprint (author's final draft)

With the advent of social networks and the rise of mobile technologies, users have become ubiquitous sensors capable of monitoring various real-world events in a crowd-sourced manner. Location-based social networks have proven to be faster than traditional media channels in reporting and geo-locating breaking news, i.e. Osama Bin Laden’s death was first confirmed on Twitter even before the announcement from the communication department at the White House. However, the deluge of user-generated data on these networks requires intelligent systems capable of identifying and characterizing such events in a comprehensive manner. The data mining community coined the term, event detection , to refer to the task of uncovering emerging patterns in data streams . Nonetheless, most data mining techniques do not reproduce the underlying data generation process, hampering to self-adapt in fast-changing scenarios. Because of this, we propose a probabilistic machine learning approach to event detection which explicitly models the data generation process and enables reasoning about the discovered events. With the aim to set forth the differences between both approaches, we present two techniques for the problem of event detection in Twitter : a data mining technique called Tweet-SCAN and a machine learning technique called Warble. We assess and compare both techniques in a dataset of tweets geo-located in the city of Barcelona during its annual festivities. Last but not least, we present the algorithmic changes and data processing frameworks to scale up the proposed techniques to big data workloads., This work is partially supported by Obra Social “la Caixa”, by the Spanish Ministry of Science and Innovation under contract (TIN2015-65316), by the Severo Ochoa Program (SEV2015-0493), by SGR programs of the Catalan Government (2014-SGR-1051, 2014-SGR-118), Collectiveware (TIN2015-66863-C2-1-R) and BSC/UPC NVIDIA GPU Center of Excellence.We would also like to thank the reviewers for their constructive feedback., Peer Reviewed, Postprint (author's final draft)

The increasing use of mobile social networks has lately transformed news media. Real-world events are nowadays reported in social networks much faster than in traditional channels. As a result, the autonomous detection of events from networks like Twitter has gained lot of interest in both research and media groups. DBSCAN-like algorithms constitute a well-known clustering approach to retrospective event detection. However, scaling such algorithms to geographically large regions and temporarily long periods present two major shortcomings. First, detecting real-world events from the vast amount of tweets cannot be performed anymore in a single machine. Second, the tweeting activity varies a lot within these broad space-time regions limiting the use of global parameters. Against this background, we propose to scale DBSCAN-like event detection techniques by parallelizing and distributing them through a novel density-aware MapReduce scheme. The proposed scheme partitions tweet data as per its spatial and temporal features and tailors local DBSCAN parameters to local tweet densities. We implement the scheme in Apache Spark and evaluate its performance in a dataset composed of geo-located tweets in the Iberian peninsula during the course of several football matches. The results pointed out to the benefits of our proposal against other state-of-the-art techniques in terms of speed-up and detection accuracy., Peer Reviewed, Postprint (author's final draft)

All right sreserved. We present GeoSRS, a hybrid recommender system for a popular location-based social network (LBSN), in which users are able to write short reviews on the places of interest they visit. Using state-of-the-art text mining techniques, our system recommends locations to users using as source the whole set of text reviews in addition to their geographical location. To evaluate our system, we have collected our own data sets by crawling the social network Foursquare. To do this efficiently, we propose the use of a parallel version of the Quadtree technique, which may be applicable to crawling/exploring other spatially distributed sources. Finally, we study the performance of GeoSRS on our collected data set and conclude that by combining sentiment analysis and text modeling, GeoSRS generates more accurate recommendations. The performance of the system improves as more reviews are available, which further motivates the use of large-scale crawling techniques such as the Quadtree., Preprint

Premi al millor PFC en l'Àrea de Sistemes de la informació d'Enginyeria de Telecomunicació o d'Enginyeria Electrònica de l'ETSETB-UPC (curs 2013-2014). Atorgat per Cátedra Red.es, Social Recommender Systems were born with the goal to mitigate the current information overload caused by the birth of Social Networks among other causes. They have enabled Internet actors (e.g. users, web browsers, sensors, actuators, etc.) to make more informed decisions based on the information that is been shown to them, up to the point that some actors even blindly trust the recommendation generated by these systems. Within this scenario, this thesis proposes a novel Hybrid Social Recommender System purely based on the text reviews typed by users. The proposed engine treats the review content and sentiment separately and finally, combines both into a single recommendation. Very little scientific research has been published on mining text reviews with the aim of performing item recommendation. Moreover, among all Hybrid Recommendation Systems in the literature, none use the above-mentioned review features into a collaborative and content-based recommender. With the purpose in mind of assessing the platform effectiveness, we present a methodology that goes from the process of extracting the data directly from a Social Network, cleaning and pre-processing the text data, building the predictive model with different state-of-the art machine learning techniques, up to the point of evaluating the system in terms of several key metrics. The data extraction process gains our attention due to the challenges imposed by most social platforms in obtaining all the geo-positioned data generated in a bounded region. To overcome the platform limitations, we introduce the use of the Quadtree algorithm with the goal of crawling all the geo-positioned reviews. The algorithm is enhanced with a module that copes with the time dynamics and captures the time-stamped data as well. Moreover, we study the effectiveness of the Quadtree partition method to crawl any type of spatial data, which tends to be softly distributed in the area. This thesis draws several conclusions from the availa, Award-winning

Projecte fet en col.laboració amb Georgia Institute of Technology, Català: En aquest projecte estudiem la comunicació entre diversos nano-dispositius que configuren una Nanoxarxa Electromagnètica sense cables. Tenint en compte les peculiaritats de la capa física, proposem un mètode d'accés múltiple, Rate Division Time Spread On-Off (RD TS-OOK), basat en la transmissió de polsos de femto-segons de duració, els quals són enviats a diferents temps entre símbols per a cada usuari. A sobre d'aquest mètode d'accés múltiple construïm un nou protocol MAC, interference-Aware Adaptive Asynchronous (i-AAA), el qual assegura una entrega justa del paquests de manera asíncrona, simple, però robusta. Finalment, analitzem el rendiment dels mecanismes proposats i optimitzem els seus paràmetres de manera que funcionin de la manera més eficient. La longitud dels paquets de dades té un impacte especial sobre el rendiment del protocols. És per això, que optimitzem el seu valor en base a una mètrica què té en compte l'eficiència energètica del protocol., Castellano: En este proyecto estudiamos la comunicación entre varios nano-dispositivos que configuran una Nanored electromágnetica sin cables. Teniendo en cuenta las peculiaridades de la capa física, proponemos un método de acceso múltiple, Rate Division Time Spread On-Off (RD TS-OOK), basado en la transmisión de pulsos de femto-segundos de duración, los cuales son enviados a diferente tiempo entre símbolos para cada usuario. En base a este método de acceso múltiple construimos un nuevo protocolo MAC, interference-Aware Adaptive Asynchronous (i-AAA), el cual asegura la entrega de los paquetes de forma asíncrona, simple, pero robusta. Finalmente, analizamos el rendimiento de los mecanimos propuestos y optimizamos sus parámetros de modo que funcionen de manera más eficiente. La longitud de los paquetes de datos tiene un impacto especial en el rendimiento de los protocolos. Es por eso, que optimizamos su valor en base a una métrica que tiene en cuenta la eficiencia energetic del protocolo., English: In this work, we study the communication among many nano-devices configuring electromagnetic Nanonetworks. Taking into account the underlying physical layer, we propose a multiple access method, Rate Division Time Spread On-Off Keying (RD TS-OOK), which is based on the radiation of femtosecond-long pulses with different intersymbol durations for each user. On top of this multiple access method, we build up a new MAC protocol, interference-Aware Adaptive Asynchronous (i-AAA), that ensures fair packet delivery in an asynchronous, simple but robust way. Finally, the performances of the proposed mechanisms are analytically studied and their parameters optimized in order to efficiently adjust them. Packet size has an special impact on the protocol performances; this is why, we lastly optimize its value based on an energy efficiency metric.

It has been widely observed that adult men of all ages are at higher risk of developing serious complications from COVID-19 when compared with women. This study aimed to investigate the association of COVID-19 positivity and severity with estrogen exposure in women, in a population based matched cohort study of female users of the COVID Symptom Study application in the UK. Analyses included 152,637 women for menopausal status, 295,689 women for exogenous estrogen intake in the form of the combined oral contraceptive pill (COCP), and 151,193 menopausal women for hormone replacement therapy (HRT). Data were collected using the COVID Symptom Study in May-June 2020. Analyses investigated associations between predicted or tested COVID-19 status and menopausal status, COCP use, and HRT use, adjusting for age, smoking and BMI, with follow-up age sensitivity analysis, and validation in a subset of participants from the TwinsUK cohort. Menopausal women had higher rates of predicted COVID-19 (P = 0.003). COCP-users had lower rates of predicted COVID-19 (P = 8.03E-05), with reduction in hospital attendance (P = 0.023). Menopausal women using HRT or hormonal therapies did not exhibit consistent associations, including increased rates of predicted COVID-19 (P = 2.22E-05) for HRT users alone. The findings support a protective effect of estrogen exposure on COVID-19, based on positive association between predicted COVID-19 with menopausal status, and negative association with COCP use. HRT use was positively associated with COVID-19, but the results should be considered with caution due to lack of data on HRT type, route of administration, duration of treatment, and potential unaccounted for confounders and comorbidities.

It has been widely observed that adult men of all ages are at higher risk of developing serious complications from COVID-19 when compared with women. This study aimed to investigate the association of COVID-19 positivity and severity with estrogen exposure in women, in a population based matched cohort study of female users of the COVID Symptom Study application in the UK. Analyses included 152,637 women for menopausal status, 295,689 women for exogenous estrogen intake in the form of the combined oral contraceptive pill (COCP), and 151,193 menopausal women for hormone replacement therapy (HRT). Data were collected using the COVID Symptom Study in May-June 2020. Analyses investigated associations between predicted or tested COVID-19 status and menopausal status, COCP use, and HRT use, adjusting for age, smoking and BMI, with follow-up age sensitivity analysis, and validation in a subset of participants from the TwinsUK cohort. Menopausal women had higher rates of predicted COVID-19 (P = 0.003). COCP-users had lower rates of predicted COVID-19 (P = 8.03E-05), with reduction in hospital attendance (P = 0.023). Menopausal women using HRT or hormonal therapies did not exhibit consistent associations, including increased rates of predicted COVID-19 (P = 2.22E-05) for HRT users alone. The findings support a protective effect of estrogen exposure on COVID-19, based on positive association between predicted COVID-19 with menopausal status, and negative association with COCP use. HRT use was positively associated with COVID-19, but the results should be considered with caution due to lack of data on HRT type, route of administration, duration of treatment, and potential unaccounted for confounders and comorbidities.