The MUSE Machine -- an Architecture for Structured Data Flow Computation

Computers employing some degree of data flow organisation are now well established as providing a possible vehicle for concurrent computation. Although data-driven computation frees the architecture from the constraints of the single program counter, processor and global memory, inherent in the classic von Neumann computer, there can still be problems with the unconstrained generation of fresh result tokens if a pure data flow approach is adopted. The advantages of allowing serial processing for those parts of a program which are inherently serial, and of permitting a demand-driven, as well as data-driven, mode of operation are identified and described. The MUSE machine described here is a structured architecture supporting both serial and parallel processing which allows the abstract structure of a program to be mapped onto the machine in a logical way.

Capability Matrix for Open Data

Part 14: Interoperability and Integration

A Bayesian network dealing with measurements and residuals for system monitoring

International audience

Multivalent reuse of web data about temporary art exhibitions: The Exhibitium Project

Our proposal aims to display the analysis techniques, methodologies as well as the most relevant results expected within the Exhibitium project framework (http://www.exhibitium.com). Awarded by the BBVA Foundation, the Exhibitium project is being developed by an international consortium of several research groups . Its main purpose is to build a comprehensive and structured data repository about temporary art exhibitions, captured from the web, to make them useful and reusable in various domains through open and interoperable data systems.

An investigation of imputation methods for discrete databases and multi-variate time series

When it comes to information sets in real life, often pieces of the whole set may not be available. This problem can find its origin in various reasons, describing therefore different patterns. In the literature, this problem is known as Missing Data. This issue can be fixed in various ways, from not taking into consideration incomplete observations, to guessing what those values originally were, or just ignoring the fact that some values are missing. The methods used to estimate missing data are called Imputation Methods. The work presented in this thesis has two main goals. The first one is to determine whether any kind of interactions exists between Missing Data, Imputation Methods and Supervised Classification algorithms, when they are applied together. For this first problem we consider a scenario in which the databases used are discrete, understanding discrete as that it is assumed that there is no relation between observations. These datasets underwent processes involving different combina- tions of the three components mentioned. The outcome showed that the missing data pattern strongly influences the outcome produced by a classifier. Also, in some of the cases, the complex imputation techniques investigated in the thesis were able to obtain better results than simple ones. The second goal of this work is to propose a new imputation strategy, but this time we constrain the specifications of the previous problem to a special kind of datasets, the multivariate Time Series. We designed new imputation techniques for this particular domain, and combined them with some of the contrasted strategies tested in the pre- vious chapter of this thesis. The time series also were subjected to processes involving missing data and imputation to finally propose an overall better imputation method. In the final chapter of this work, a real-world example is presented, describing a wa- ter quality prediction problem. The databases that characterized this problem had their own original latent values, which provides a real-world benchmark to test the algorithms developed in this thesis.

Business Process Management and Process Mining within a Real Business Environment: An Empirical Analysis of Event Logs Data in a Consulting Project

Il presente elaborato esplora l’attitudine delle organizzazioni nei confronti dei processi di business che le sostengono: dalla semi-assenza di struttura, all’organizzazione funzionale, fino all’avvento del Business Process Reengineering e del Business Process Management, nato come superamento dei limiti e delle problematiche del modello precedente. All’interno del ciclo di vita del BPM, trova spazio la metodologia del process mining, che permette un livello di analisi dei processi a partire dagli event data log, ossia dai dati di registrazione degli eventi, che fanno riferimento a tutte quelle attività supportate da un sistema informativo aziendale. Il process mining può essere visto come naturale ponte che collega le discipline del management basate sui processi (ma non data-driven) e i nuovi sviluppi della business intelligence, capaci di gestire e manipolare l’enorme mole di dati a disposizione delle aziende (ma che non sono process-driven). Nella tesi, i requisiti e le tecnologie che abilitano l’utilizzo della disciplina sono descritti, cosi come le tre tecniche che questa abilita: process discovery, conformance checking e process enhancement. Il process mining è stato utilizzato come strumento principale in un progetto di consulenza da HSPI S.p.A. per conto di un importante cliente italiano, fornitore di piattaforme e di soluzioni IT. Il progetto a cui ho preso parte, descritto all’interno dell’elaborato, ha come scopo quello di sostenere l’organizzazione nel suo piano di improvement delle prestazioni interne e ha permesso di verificare l’applicabilità e i limiti delle tecniche di process mining. Infine, nell’appendice finale, è presente un paper da me realizzato, che raccoglie tutte le applicazioni della disciplina in un contesto di business reale, traendo dati e informazioni da working papers, casi aziendali e da canali diretti. Per la sua validità e completezza, questo documento è stata pubblicato nel sito dell'IEEE Task Force on Process Mining.

Network Construction and Graph Theoretical Analysis of Functional Language Networks in Pediatric Epilepsy

This dissertation introduces a new approach for assessing the effects of pediatric epilepsy on the language connectome. Two novel data-driven network construction approaches are presented. These methods rely on connecting different brain regions using either extent or intensity of language related activations as identified by independent component analysis of fMRI data. An auditory description decision task (ADDT) paradigm was used to activate the language network for 29 patients and 30 controls recruited from three major pediatric hospitals. Empirical evaluations illustrated that pediatric epilepsy can cause, or is associated with, a network efficiency reduction. Patients showed a propensity to inefficiently employ the whole brain network to perform the ADDT language task; on the contrary, controls seemed to efficiently use smaller segregated network components to achieve the same task. To explain the causes of the decreased efficiency, graph theoretical analysis was carried out. The analysis revealed no substantial global network feature differences between the patient and control groups. It also showed that for both subject groups the language network exhibited small-world characteristics; however, the patient’s extent of activation network showed a tendency towards more random networks. It was also shown that the intensity of activation network displayed ipsilateral hub reorganization on the local level. The left hemispheric hubs displayed greater centrality values for patients, whereas the right hemispheric hubs displayed greater centrality values for controls. This hub hemispheric disparity was not correlated with a right atypical language laterality found in six patients. Finally it was shown that a multi-level unsupervised clustering scheme based on self-organizing maps, a type of artificial neural network, and k-means was able to fairly and blindly separate the subjects into their respective patient or control groups. The clustering was initiated using the local nodal centrality measurements only. Compared to the extent of activation network, the intensity of activation network clustering demonstrated better precision. This outcome supports the assertion that the local centrality differences presented by the intensity of activation network can be associated with focal epilepsy.

Modelling and predictive control techniques for building heating systems

Model predictive control (MPC) has often been referred to in literature as a potential method for more efficient control of building heating systems. Though a significant performance improvement can be achieved with an MPC strategy, the complexity introduced to the commissioning of the system is often prohibitive. Models are required which can capture the thermodynamic properties of the building with sufficient accuracy for meaningful predictions to be made. Furthermore, a large number of tuning weights may need to be determined to achieve a desired performance. For MPC to become a practicable alternative, these issues must be addressed. Acknowledging the impact of the external environment as well as the interaction of occupants on the thermal behaviour of the building, in this work, techniques have been developed for deriving building models from data in which large, unmeasured disturbances are present. A spatio-temporal filtering process was introduced to determine estimates of the disturbances from measured data, which were then incorporated with metaheuristic search techniques to derive high-order simulation models, capable of replicating the thermal dynamics of a building. While a high-order simulation model allowed for control strategies to be analysed and compared, low-order models were required for use within the MPC strategy itself. The disturbance estimation techniques were adapted for use with system-identification methods to derive such models. MPC formulations were then derived to enable a more straightforward commissioning process and implemented in a validated simulation platform. A prioritised-objective strategy was developed which allowed for the tuning parameters typically associated with an MPC cost function to be omitted from the formulation by separation of the conflicting requirements of comfort satisfaction and energy reduction within a lexicographic framework. The improved ability of the formulation to be set-up and reconfigured in faulted conditions was shown.

Análisis de las barreras para la unificación de una Historia Clínica Electrónica -HCE- en Colombia

Objetivo: Identificar las barreras para la unificación de una Historia Clínica Electrónica –HCE- en Colombia. Materiales y Métodos: Se realizó un estudio cualitativo. Se realizaron entrevistas semiestructuradas a profesionales y expertos de 22 instituciones del sector salud, de Bogotá y de los departamentos de Cundinamarca, Santander, Antioquia, Caldas, Huila, Valle del Cauca. Resultados: Colombia se encuentra en una estructuración para la implementación de la Historia Clínica Electrónica Unificada -HCEU-. Actualmente, se encuentra en unificación en 42 IPSs públicas en el departamento de Cundinamarca, el desarrollo de la HCEU en el país es privado y de desarrollo propio debido a las necesidades particulares de cada IPS. Conclusiones: Se identificaron barreras humanas, financieras, legales, organizacionales, técnicas y profesionales en los departamentos entrevistados. Se identificó que la unificación de la HCE depende del acuerdo de voluntades entre las IPSs del sector público, privado, EPSs, y el Gobierno Nacional.

Synthetic portfolio for event studies: Estimating the effects of volatility call auctions

We propose a method denoted as synthetic portfolio for event studies in market microstructure that is particularly interesting to use with high frequency data and thinly traded markets. The method is based on Synthetic Control Method and provides a robust data driven method to build a counterfactual for evaluating the effects of the volatility call auctions. We find that SMC could be used if the loss function is defined as the difference between the returns of the asset and the returns of a synthetic portfolio. We apply SCM to test the performance of the volatility call auction as a circuit breaker in the context of an event study. We find that for Colombian Stock Market securities, the asynchronicity of intraday data reduces the analysis to a selected group of stocks, however it is possible to build a tracking portfolio. The realized volatility increases after the auction, indicating that the mechanism is not enhancing the price discovery process.

Classi cation of new electricity customers based on surveys and smart metering data

This paper proposes a process for the classifi cation of new residential electricity customers. The current state of the art is extended by using a combination of smart metering and survey data and by using model-based feature selection for the classifi cation task. Firstly, the normalized representative consumption profi les of the population are derived through the clustering of data from households. Secondly, new customers are classifi ed using survey data and a limited amount of smart metering data. Thirdly, regression analysis and model-based feature selection results explain the importance of the variables and which are the drivers of diff erent consumption profi les, enabling the extraction of appropriate models. The results of a case study show that the use of survey data signi ficantly increases accuracy of the classifi cation task (up to 20%). Considering four consumption groups, more than half of the customers are correctly classifi ed with only one week of metering data, with more weeks the accuracy is signifi cantly improved. The use of model-based feature selection resulted in the use of a signifi cantly lower number of features allowing an easy interpretation of the derived models.

Automatic detection of the inferior alveolar nerve in CT scans with CenterNet

The inferior alveolar nerve (IAN) lies within the mandibular canal, named inferior alveolar canal in literature. The detection of this nerve is important during maxillofacial surgeries or for creating dental implants. The poor quality of cone-beam computed tomography (CBCT) and computed tomography (CT) scans and/or bone gaps within the mandible increase the difficulty of this task, posing a challenge to human experts who are going to manually detect it and resulting in a time-consuming task.Therefore this thesis investigates two methods to automatically detect the IAN: a non-data driven technique and a deep-learning method. The latter tracks the IAN position at each frame leveraging detections obtained with the deep neural network CenterNet, fined-tuned for our task, and temporal and spatial information.

Hybrid Artificial Intelligence to Extract Patterns and Rules from Argumentative and Legal Texts

This Thesis is composed of a collection of works written in the period 2019-2022, whose aim is to find methodologies of Artificial Intelligence (AI) and Machine Learning to detect and classify patterns and rules in argumentative and legal texts. We define our approach “hybrid”, since we aimed at designing hybrid combinations of symbolic and sub-symbolic AI, involving both “top-down” structured knowledge and “bottom-up” data-driven knowledge. A first group of works is dedicated to the classification of argumentative patterns. Following the Waltonian model of argument and the related theory of Argumentation Schemes, these works focused on the detection of argumentative support and opposition, showing that argumentative evidences can be classified at fine-grained levels without resorting to highly engineered features. To show this, our methods involved not only traditional approaches such as TFIDF, but also some novel methods based on Tree Kernel algorithms. After the encouraging results of this first phase, we explored the use of a some emerging methodologies promoted by actors like Google, which have deeply changed NLP since 2018-19 — i.e., Transfer Learning and language models. These new methodologies markedly improved our previous results, providing us with best-performing NLP tools. Using Transfer Learning, we also performed a Sequence Labelling task to recognize the exact span of argumentative components (i.e., claims and premises), thus connecting portions of natural language to portions of arguments (i.e., to the logical-inferential dimension). The last part of our work was finally dedicated to the employment of Transfer Learning methods for the detection of rules and deontic modalities. In this case, we explored a hybrid approach which combines structured knowledge coming from two LegalXML formats (i.e., Akoma Ntoso and LegalRuleML) with sub-symbolic knowledge coming from pre-trained (and then fine-tuned) neural architectures.

Variational and learning models for image and time series inverse problems

Inverse problems are at the core of many challenging applications. Variational and learning models provide estimated solutions of inverse problems as the outcome of specific reconstruction maps. In the variational approach, the result of the reconstruction map is the solution of a regularized minimization problem encoding information on the acquisition process and prior knowledge on the solution. In the learning approach, the reconstruction map is a parametric function whose parameters are identified by solving a minimization problem depending on a large set of data. In this thesis, we go beyond this apparent dichotomy between variational and learning models and we show they can be harmoniously merged in unified hybrid frameworks preserving their main advantages. We develop several highly efficient methods based on both these model-driven and data-driven strategies, for which we provide a detailed convergence analysis. The arising algorithms are applied to solve inverse problems involving images and time series. For each task, we show the proposed schemes improve the performances of many other existing methods in terms of both computational burden and quality of the solution. In the first part, we focus on gradient-based regularized variational models which are shown to be effective for segmentation purposes and thermal and medical image enhancement. We consider gradient sparsity-promoting regularized models for which we develop different strategies to estimate the regularization strength. Furthermore, we introduce a novel gradient-based Plug-and-Play convergent scheme considering a deep learning based denoiser trained on the gradient domain. In the second part, we address the tasks of natural image deblurring, image and video super resolution microscopy and positioning time series prediction, through deep learning based methods. We boost the performances of supervised, such as trained convolutional and recurrent networks, and unsupervised deep learning strategies, such as Deep Image Prior, by penalizing the losses with handcrafted regularization terms.

Abusing data dominance in the digital market. The application of merger control and article 102 TFEU to data-related conducts.

The internet and digital technologies revolutionized the economy. Regulating the digital market has become a priority for the European Union. While promoting innovation and development, EU institutions must assure that the digital market maintains a competitive structure. Among the numerous elements characterizing the digital sector, users’ data are particularly important. Digital services are centered around personal data, the accumulation of which contributed to the centralization of market power in the hands of a few large providers. As a result, data-driven mergers and data-related abuses gained a central role for the purposes of EU antitrust enforcement. In light of these considerations, this work aims at assessing whether EU competition law is well-suited to address data-driven mergers and data-related abuses of dominance. These conducts are of crucial importance to the maintenance of competition in the digital sector, insofar as the accumulation of users’ data constitutes a fundamental competitive advantage. To begin with, part 1 addresses the specific features of the digital market and their impact on the definition of the relevant market and the assessment of dominance by antitrust authorities. Secondly, part 2 analyzes the EU’s case law on data-driven mergers to verify if merger control is well-suited to address these concentrations. Thirdly, part 3 discusses abuses of dominance in the phase of data collection and the legal frameworks applicable to these conducts. Fourthly, part 4 focuses on access to “essential” datasets and the indirect effects of anticompetitive conducts on rivals’ ability to access users’ information. Finally, Part 5 discusses differential pricing practices implemented online and based on personal data. As it will be assessed, the combination of an efficient competition law enforcement and the auspicial adoption of a specific regulation seems to be the best solution to face the challenges raised by “data-related dominance”.