Self-supervised information retrieval: a novel approach based on Deep Metric Learning and Neural Language Models

Most of the existing open-source search engines, utilize keyword or tf-idf based techniques to find relevant documents and web pages relative to an input query. Although these methods, with the help of a page rank or knowledge graphs, proved to be effective in some cases, they often fail to retrieve relevant instances for more complicated queries that would require a semantic understanding to be exploited. In this Thesis, a self-supervised information retrieval system based on transformers is employed to build a semantic search engine over the library of Gruppo Maggioli company. Semantic search or search with meaning can refer to an understanding of the query, instead of simply finding words matches and, in general, it represents knowledge in a way suitable for retrieval. We chose to investigate a new self-supervised strategy to handle the training of unlabeled data based on the creation of pairs of ’artificial’ queries and the respective positive passages. We claim that by removing the reliance on labeled data, we may use the large volume of unlabeled material on the web without being limited to languages or domains where labeled data is abundant.

Methods and tools for analysis and management of risks and regulatory compliance in the healthcare sector: the hospital at home – HaH

Changing or creating an organisation means creating a new process. Each process involves many risks that need to be identified and managed. The main risks considered here are procedural and legal risks. The former are related to the risks of errors that may occur during processes, while the latter are related to the compliance of processes with regulations. Managing the risks implies proposing changes to the processes that allow the desired result: an optimised process. In order to manage a company and optimise it in the best possible way, not only should the organisational aspect, risk management and legal compliance be taken into account, but it is important that they are all analysed simultaneously with the aim of finding the right balance that satisfies them all. This is the aim of this thesis, to provide methods and tools to balance these three characteristics, and to enable this type of optimisation, ICT support is used. This work isn’t a thesis in computer science or law, but rather an interdisciplinary thesis. Most of the work done so far is vertical and in a specific domain. The particularity and aim of this thesis is not to carry out an in-depth analysis of a particular aspect, but rather to combine several important aspects, normally analysed separately, which however have an impact and influence each other. In order to carry out this kind of interdisciplinary analysis, the knowledge base of both areas was involved and the combination and collaboration of different experts in the various fields was necessary. Although the methodology described is generic and can be applied to all sectors, the case study considered is a new type of healthcare service that allows patients in acute disease to be hospitalised to their home. This provide the possibility to perform experiments using real hospital database.

Methods for integrating machine learning and constrained optimization

In the framework of industrial problems, the application of Constrained Optimization is known to have overall very good modeling capability and performance and stands as one of the most powerful, explored, and exploited tool to address prescriptive tasks. The number of applications is huge, ranging from logistics to transportation, packing, production, telecommunication, scheduling, and much more. The main reason behind this success is to be found in the remarkable effort put in the last decades by the OR community to develop realistic models and devise exact or approximate methods to solve the largest variety of constrained or combinatorial optimization problems, together with the spread of computational power and easily accessible OR software and resources. On the other hand, the technological advancements lead to a data wealth never seen before and increasingly push towards methods able to extract useful knowledge from them; among the data-driven methods, Machine Learning techniques appear to be one of the most promising, thanks to its successes in domains like Image Recognition, Natural Language Processes and playing games, but also the amount of research involved. The purpose of the present research is to study how Machine Learning and Constrained Optimization can be used together to achieve systems able to leverage the strengths of both methods: this would open the way to exploiting decades of research on resolution techniques for COPs and constructing models able to adapt and learn from available data. In the first part of this work, we survey the existing techniques and classify them according to the type, method, or scope of the integration; subsequently, we introduce a novel and general algorithm devised to inject knowledge into learning models through constraints, Moving Target. In the last part of the thesis, two applications stemming from real-world projects and done in collaboration with Optit will be presented.

Temporary teams learning and performance: the case of 3D printing technology in orthopaedic surgery

This dissertation contributes to the scholarly debate on temporary teams by exploring team interactions and boundaries.The fundamental challenge in temporary teams originates from temporary participation in the teams. First, as participants join the team for a short period of time, there is not enough time to build trust, share understanding, and have effective interactions. Consequently, team outputs and practices built on team interactions become vulnerable. Secondly, as team participants move on and off the teams, teams’ boundaries become blurred over time. It leads to uncertainty among team participants and leaders about who is/is not identified as a team member causing collective disagreement within the team. Focusing on the above mentioned challenges, we conducted this research in healthcare organisations since the use of temporary teams in healthcare and hospital setting is prevalent. In particular, we focused on orthopaedic teams that provide personalised treatments for patients using 3D printing technology. Qualitative and quantitative data were collected using interviews, observations, questionnaires and archival data at Rizzoli Orthopaedic Institute, Bologna, Italy. This study provides the following research outputs. The first is a conceptual study that explores temporary teams’ literature using bibliometric analysis and systematic literature review to highlight research gaps. The second paper qualitatively studies temporary relationships within the teams by collecting data using group interviews and observations. The results highlighted the role of short-term dyadic relationships as a ground to share and transfer knowledge at the team level. Moreover, hierarchical structure of the teams facilitates knowledge sharing by supporting dyadic relationships within and beyond the team meetings. The third paper investigates impact of blurred boundaries on temporary teams’ performance. Using quantitative data collected through questionnaires and archival data, we concluded that boundary blurring in terms of fluidity, overlap and dispersion differently impacts team performance at high and low levels of task complexity.

Applied deep learning and data science with a human-centric and data-centric approach

The term Artificial intelligence acquired a lot of baggage since its introduction and in its current incarnation is synonymous with Deep Learning. The sudden availability of data and computing resources has opened the gates to myriads of applications. Not all are created equal though, and problems might arise especially for fields not closely related to the tasks that pertain tech companies that spearheaded DL. The perspective of practitioners seems to be changing, however. Human-Centric AI emerged in the last few years as a new way of thinking DL and AI applications from the ground up, with a special attention at their relationship with humans. The goal is designing a system that can gracefully integrate in already established workflows, as in many real-world scenarios AI may not be good enough to completely replace its humans. Often this replacement may even be unneeded or undesirable. Another important perspective comes from, Andrew Ng, a DL pioneer, who recently started shifting the focus of development from “better models” towards better, and smaller, data. He defined his approach Data-Centric AI. Without downplaying the importance of pushing the state of the art in DL, we must recognize that if the goal is creating a tool for humans to use, more raw performance may not align with more utility for the final user. A Human-Centric approach is compatible with a Data-Centric one, and we find that the two overlap nicely when human expertise is used as the driving force behind data quality. This thesis documents a series of case-studies where these approaches were employed, to different extents, to guide the design and implementation of intelligent systems. We found human expertise proved crucial in improving datasets and models. The last chapter includes a slight deviation, with studies on the pandemic, still preserving the human and data centric perspective.

Methods and tools to foster territorial approaches for the assessment of food supply chains

The challenges of the current global food systems are often framed around feeding the world's growing population while meeting sustainable development for future generations. Globalization has brought to a fragmentation of food spaces, leading to a flexible and mutable supply chain. This poses a major challenge to food and nutrition security, affecting also rural-urban dynamics in territories. Furthermore, the recent crises have highlighted the vulnerability to shocks and disruptions of the food systems and the eco-system due to the intensive management of natural, human and economic capital. Hence, a sustainable and resilient transition of the food systems is required through a multi-faceted approach that tackles the causes of unsustainability and promotes sustainable practices at all levels of the food system. In this respect, a territorial approach becomes a relevant entry point of analysis for the food system’s multifunctionality and can support the evaluation of sustainability by quantifying impacts associated with quantitative methods and understanding the territorial responsibility of different actors with qualitative ones. Against this background the present research aims to i) investigate the environmental, costing and social indicators suitable for a scoring system able to measure the integrated sustainability performance of food initiatives within the City/Region territorial context; ii) develop a territorial assessment framework to measure sustainability impacts of agricultural systems; and iii) define an integrated methodology to match production and consumption at a territorial level to foster a long-term vision of short food supply chains. From a methodological perspective, the research proposes a mixed quantitative and qualitative research method. The outcomes provide an in-depth view into the environmental and socio-economic impacts of food systems at the territorial level, investigating possible indicators, frameworks, and business strategies to foster their future sustainable development.

Computational creativity: an interdisciplinary approach to sequential learning and creative generations

Creativity seems mysterious; when we experience a creative spark, it is difficult to explain how we got that idea, and we often recall notions like ``inspiration" and ``intuition" when we try to explain the phenomenon. The fact that we are clueless about how a creative idea manifests itself does not necessarily imply that a scientific explanation cannot exist. We are unaware of how we perform certain tasks, such as biking or language understanding, but we have more and more computational techniques that can replicate and hopefully explain such activities. We should understand that every creative act is a fruit of experience, society, and culture. Nothing comes from nothing. Novel ideas are never utterly new; they stem from representations that are already in mind. Creativity involves establishing new relations between pieces of information we had already: then, the greater the knowledge, the greater the possibility of finding uncommon connections, and the more the potential to be creative. In this vein, a beneficial approach to a better understanding of creativity must include computational or mechanistic accounts of such inner procedures and the formation of the knowledge that enables such connections. That is the aim of Computational Creativity: to develop computational systems for emulating and studying creativity. Hence, this dissertation focuses on these two related research areas: discussing computational mechanisms to generate creative artifacts and describing some implicit cognitive processes that can form the basis for creative thoughts.

Red flesh fruit in european pear (Pyrus communis): genetic sources, QTLs, candidate genes and tools for breeding

Red flesh fruit is a character which interest is increasing in several commercial species. Following a review of the research on the biosynthesis and accumulation of anthocyanin in pears (Chapter 1) the general aim of the project is reported in Chapter 2. Chapter 3 reports the results of a molecular analysis of 33 red-fleshed pear accessions, genotyped with 18 SSR markers with the aim of improving germplasm conservation strategies to support ongoing breeding programs. The molecular profiles revealed both cases of synonymy and homonymy and 6 unique genotypes were identified. The S-allele were established to highlight the genetic relationships among these landraces. Four of the unique genotypes have been clustered based on pomological data. In the Chapter 4, the work was directed to identify the putative genomic regions involved in the appearance of this character in pear fruit. A crossing population (‘Carmen’ x ‘Cocomerina Precoce’) segregating for the trait was phenotyped for 2 consecutive years and used for QTL analysis. A strong QTL was identified in a small genomic region related to the red flesh fruit trait at 27 Mb from the start of LG5. Two candidate genes were detected in this genomic region: ‘PcMYB114’ and ‘PcABCC2’. SSR marker SSR114 was found able to detect the red flesh phenotype segregation in all the red-fleshed pear accessions and segregating progenies tested. Chapter 5 focuses on examining the trend of anthocyanin synthesis and accumulation during the fruit development, from fruit set to ripening time. Three different trials were planned: qPCR and HPLC methods were performed to correlate the genes expression with the anthocyanin accumulation in ‘Cocomerina Precoce’ and six progenies. Total transcriptome sequencing was used to compare the differential genes expression between red and white-fleshed fruit. Chapter 6 reviews and analyses all the earlier study findings while providing new potential future perspectives.

Sustainable futuring. Learning and unlearning the past A critical discourse study of ‘fashion sustainability’-related discursive formations in Vogue Italia (1965-2021)

Recent scholarly works on the relationship between ‘fashion’ and ‘sustainability’ have identified a need for a systemic transition towards fashion media ‘for sustaianbility’. Nevertheless, the academic research on the topic is still limited and rather circumscribed to the analysis of marketing practices, while only recently some more systemic and critical analyses of the symbolic production of sustainability through fashion media have been undertaken. Responding to this need for an in-depth investigation of ‘sustainability’-related media production, my research focuses on the ‘fashion sustainability’-related discursive formations in the context of one of the most influential fashion magazines today – Vogue Italia. In order to investigate the ways in which the ‘sustainability’ discourse was formed and has evolved, the study considered the entire Vogue Italia archive from 1965 to 2021. The data collection was carried out in two phases, and the individualised relevant discursive units were then in-depth and critically analysed to allow for a grounded assessment of the media giant’s position. The Discourse-Historical Approach provided a methodological base for the analysis, which took into consideration the various levels of context: the immediate textual and intertextual, but also the broader socio-cultural context of the predominant, over-production oriented and capital-led fashion system. The findings led to a delineation of the evolution of the ‘fashion sustainability’ discourse, unveiling how despite Vogue Italia’s auto-determination as attentive to ‘sustainability’-related topics, the magazine is systemically employing discursive strategies which significantly mitigate the meaning of the ‘sustainable commitment’ and thus the meaning of ‘fashion sustainability’.

Deep learning and spatial transcriptomics to investigate thyroid tumors

There are many diseases that affect the thyroid gland, and among them are carcinoma. Thyroid cancer is the most common endocrine neoplasm and the second most frequent cancer in the 0-49 age group. This thesis deals with two studies I conducted during my PhD. The first concerns the development of a Deep Learning model to be able to assist the pathologist in screening of thyroid cytology smears. This tool created in collaboration with Prof. Diciotti, affiliated with the DEI-UNIBO "Guglielmo Marconi" Department of Electrical Energy and Information Engineering, has an important clinical implication in that it allows patients to be stratified between those who should undergo surgery and those who should not. The second concerns the application of spatial transcriptomics on well-differentiated thyroid carcinomas to better understand their invasion mechanisms and thus to better comprehend which genes may be involved in the proliferation of these tumors. This project specifically was made possible through a fruitful collaboration with the Gustave Roussy Institute in Paris. Studying thyroid carcinoma deeply is essential to improve patient care, increase survival rates, and enhance the overall understanding of this prevalent cancer. It can lead to more effective prevention, early detection, and treatment strategies that benefit both patients and the healthcare system.

Dynamic Human Robot Interaction Framework Using Deep Learning and Robot Operating System (ROS): a practical approach

Trying to explain to a robot what to do is a difficult undertaking, and only specific types of people have been able to do so far, such as programmers or operators who have learned how to use controllers to communicate with a robot. My internship's goal was to create and develop a framework that would make that easier. The system uses deep learning techniques to recognize a set of hand gestures, both static and dynamic. Then, based on the gesture, it sends a command to a robot. To be as generic as feasible, the communication is implemented using Robot Operating System (ROS). Furthermore, users can add new recognizable gestures and link them to new robot actions; a finite state automaton enforces the users' input verification and correct action sequence. Finally, the users can create and utilize a macro to describe a sequence of actions performable by a robot.

Machine learning and convolutional neural networks for melanomas identification

Il machine learning negli ultimi anni ha acquisito una crescente popolarità nell’ambito della ricerca scientifica e delle sue applicazioni. Lo scopo di questa tesi è stato quello di studiare il machine learning nei suoi aspetti generali e applicarlo a problemi di computer vision. La tesi ha affrontato le difficoltà del dover spiegare dal punto di vista teorico gli algoritmi alla base delle reti neurali convoluzionali e ha successivamente trattato due problemi concreti di riconoscimento immagini: il dataset MNIST (immagini di cifre scritte a mano) e un dataset che sarà chiamato ”MELANOMA dataset” (immagini di melanomi e nevi sani). Utilizzando le tecniche spiegate nella sezione teorica si sono riusciti ad ottenere risultati soddifacenti per entrambi i dataset ottenendo una precisione del 98% per il MNIST e del 76.8% per il MELANOMA dataset

Strumenti e metodi per lo sviluppo prodotto di lenti oftalmiche

The recent introduction of free form NC machining in the ophthalmic field involved a full review of the former product development process both from the design and the manufacturing viewpoint. Aim of the present work is to investigate and to set up innovative methods and tools supporting the product development, particularly for lenses characterized by free form geometry, as e.g. progressive lenses. In the design stage, the research addressed geometric modeling of complex lens shapes and relevant analysis tools for the optical-geometrical characterization of the produced models. In the manufacturing stage, the main interest was focused on the set-up of the fabrication process, particularly on the NC machining process for which an integration CADCAM software was developed for the generation and the simulation of the machining cycle. The methodologies and tools made available by the present work are currently used in the development of new complex geometry product typologies as, e.g. progressive lenses.

Mobile and web tools for participative learning

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa, para a obtenção do grau de Mestre em Engenharia Informática

Machine learning for geospatial data : algorithms, software tools and case studies

Résumé Cette thèse est consacrée à l'analyse, la modélisation et la visualisation de données environnementales à référence spatiale à l'aide d'algorithmes d'apprentissage automatique (Machine Learning). L'apprentissage automatique peut être considéré au sens large comme une sous-catégorie de l'intelligence artificielle qui concerne particulièrement le développement de techniques et d'algorithmes permettant à une machine d'apprendre à partir de données. Dans cette thèse, les algorithmes d'apprentissage automatique sont adaptés pour être appliqués à des données environnementales et à la prédiction spatiale. Pourquoi l'apprentissage automatique ? Parce que la majorité des algorithmes d'apprentissage automatiques sont universels, adaptatifs, non-linéaires, robustes et efficaces pour la modélisation. Ils peuvent résoudre des problèmes de classification, de régression et de modélisation de densité de probabilités dans des espaces à haute dimension, composés de variables informatives spatialisées (« géo-features ») en plus des coordonnées géographiques. De plus, ils sont idéaux pour être implémentés en tant qu'outils d'aide à la décision pour des questions environnementales allant de la reconnaissance de pattern à la modélisation et la prédiction en passant par la cartographie automatique. Leur efficacité est comparable au modèles géostatistiques dans l'espace des coordonnées géographiques, mais ils sont indispensables pour des données à hautes dimensions incluant des géo-features. Les algorithmes d'apprentissage automatique les plus importants et les plus populaires sont présentés théoriquement et implémentés sous forme de logiciels pour les sciences environnementales. Les principaux algorithmes décrits sont le Perceptron multicouches (MultiLayer Perceptron, MLP) - l'algorithme le plus connu dans l'intelligence artificielle, le réseau de neurones de régression généralisée (General Regression Neural Networks, GRNN), le réseau de neurones probabiliste (Probabilistic Neural Networks, PNN), les cartes auto-organisées (SelfOrganized Maps, SOM), les modèles à mixture Gaussiennes (Gaussian Mixture Models, GMM), les réseaux à fonctions de base radiales (Radial Basis Functions Networks, RBF) et les réseaux à mixture de densité (Mixture Density Networks, MDN). Cette gamme d'algorithmes permet de couvrir des tâches variées telle que la classification, la régression ou l'estimation de densité de probabilité. L'analyse exploratoire des données (Exploratory Data Analysis, EDA) est le premier pas de toute analyse de données. Dans cette thèse les concepts d'analyse exploratoire de données spatiales (Exploratory Spatial Data Analysis, ESDA) sont traités selon l'approche traditionnelle de la géostatistique avec la variographie expérimentale et selon les principes de l'apprentissage automatique. La variographie expérimentale, qui étudie les relations entre pairs de points, est un outil de base pour l'analyse géostatistique de corrélations spatiales anisotropiques qui permet de détecter la présence de patterns spatiaux descriptible par une statistique. L'approche de l'apprentissage automatique pour l'ESDA est présentée à travers l'application de la méthode des k plus proches voisins qui est très simple et possède d'excellentes qualités d'interprétation et de visualisation. Une part importante de la thèse traite de sujets d'actualité comme la cartographie automatique de données spatiales. Le réseau de neurones de régression généralisée est proposé pour résoudre cette tâche efficacement. Les performances du GRNN sont démontrées par des données de Comparaison d'Interpolation Spatiale (SIC) de 2004 pour lesquelles le GRNN bat significativement toutes les autres méthodes, particulièrement lors de situations d'urgence. La thèse est composée de quatre chapitres : théorie, applications, outils logiciels et des exemples guidés. Une partie importante du travail consiste en une collection de logiciels : Machine Learning Office. Cette collection de logiciels a été développée durant les 15 dernières années et a été utilisée pour l'enseignement de nombreux cours, dont des workshops internationaux en Chine, France, Italie, Irlande et Suisse ainsi que dans des projets de recherche fondamentaux et appliqués. Les cas d'études considérés couvrent un vaste spectre de problèmes géoenvironnementaux réels à basse et haute dimensionnalité, tels que la pollution de l'air, du sol et de l'eau par des produits radioactifs et des métaux lourds, la classification de types de sols et d'unités hydrogéologiques, la cartographie des incertitudes pour l'aide à la décision et l'estimation de risques naturels (glissements de terrain, avalanches). Des outils complémentaires pour l'analyse exploratoire des données et la visualisation ont également été développés en prenant soin de créer une interface conviviale et facile à l'utilisation. Machine Learning for geospatial data: algorithms, software tools and case studies Abstract The thesis is devoted to the analysis, modeling and visualisation of spatial environmental data using machine learning algorithms. In a broad sense machine learning can be considered as a subfield of artificial intelligence. It mainly concerns with the development of techniques and algorithms that allow computers to learn from data. In this thesis machine learning algorithms are adapted to learn from spatial environmental data and to make spatial predictions. Why machine learning? In few words most of machine learning algorithms are universal, adaptive, nonlinear, robust and efficient modeling tools. They can find solutions for the classification, regression, and probability density modeling problems in high-dimensional geo-feature spaces, composed of geographical space and additional relevant spatially referenced features. They are well-suited to be implemented as predictive engines in decision support systems, for the purposes of environmental data mining including pattern recognition, modeling and predictions as well as automatic data mapping. They have competitive efficiency to the geostatistical models in low dimensional geographical spaces but are indispensable in high-dimensional geo-feature spaces. The most important and popular machine learning algorithms and models interesting for geo- and environmental sciences are presented in details: from theoretical description of the concepts to the software implementation. The main algorithms and models considered are the following: multi-layer perceptron (a workhorse of machine learning), general regression neural networks, probabilistic neural networks, self-organising (Kohonen) maps, Gaussian mixture models, radial basis functions networks, mixture density networks. This set of models covers machine learning tasks such as classification, regression, and density estimation. Exploratory data analysis (EDA) is initial and very important part of data analysis. In this thesis the concepts of exploratory spatial data analysis (ESDA) is considered using both traditional geostatistical approach such as_experimental variography and machine learning. Experimental variography is a basic tool for geostatistical analysis of anisotropic spatial correlations which helps to understand the presence of spatial patterns, at least described by two-point statistics. A machine learning approach for ESDA is presented by applying the k-nearest neighbors (k-NN) method which is simple and has very good interpretation and visualization properties. Important part of the thesis deals with a hot topic of nowadays, namely, an automatic mapping of geospatial data. General regression neural networks (GRNN) is proposed as efficient model to solve this task. Performance of the GRNN model is demonstrated on Spatial Interpolation Comparison (SIC) 2004 data where GRNN model significantly outperformed all other approaches, especially in case of emergency conditions. The thesis consists of four chapters and has the following structure: theory, applications, software tools, and how-to-do-it examples. An important part of the work is a collection of software tools - Machine Learning Office. Machine Learning Office tools were developed during last 15 years and was used both for many teaching courses, including international workshops in China, France, Italy, Ireland, Switzerland and for realizing fundamental and applied research projects. Case studies considered cover wide spectrum of the real-life low and high-dimensional geo- and environmental problems, such as air, soil and water pollution by radionuclides and heavy metals, soil types and hydro-geological units classification, decision-oriented mapping with uncertainties, natural hazards (landslides, avalanches) assessments and susceptibility mapping. Complementary tools useful for the exploratory data analysis and visualisation were developed as well. The software is user friendly and easy to use.