Seminari - Dipartimento Computer Science Seminari - Dipartimento Computer Science validi dal 27.09.2023 al 27.09.2024. "Mining Healthcare Big Data - Concepts, Techniques and Practice" Relatore: Alex M. H. Kuo; Provenienza: University of Victoria, BC, Canada; Data inizio: 2023-10-10; Ora inizio: 10.30; Note orario: Sala Verde (presenza e remoto); Referente interno: Carlo Combi; Riassunto: Abstract: Big Data is different from the so called big data set in that it is a collection of data so large, so complex, so distributed, and growing so fast (or so called 5Vs). Big Data are not usable until they can be aggregated and integrated into a manner that computer can process to generate knowledge. Extracting useful knowledge from Big Data can be considered as a processing pipeline that involves multiple distinct configuration stages to achieve full utilization. Each stage faces several specific challenges. The objective of this presentation is to discuss the opportunities of mining health Big Data to improve healthcare as well as the challenges and solutions for health Big Data Analytics (BDA) ndash; the process of extracting knowledge from sets of Health Big Data. Short bio: Dr. Alex Kuo holds a PhD from the Department of Computer Science, University of Nottingham, UK. He is a Professor at the School of Health Information Science, University of Victoria, BC, Canada. He was a visiting scholar at the Electronic Commerce Resource Centre (ECRC), Georgia Tech. (1999-2000), and the Center for Expanded Data Annotation and Retrieval (CEDAR), School of Medicine, Stanford University (2015-2016). Now, he is the chair of the IEEE Big Data Education Track and the study group leader of Metadata Standard for Big Data Management at the IEEE Big Data Initiative (BDI). With over 20 years of programming and data analysis practical as well as research experience, he has over 140 peer-reviewed publications. His research interests include health Big Data analytics, health data interoperability, health database & data warehousing, data mining application in healthcare, e-health and clinical decision support system. Riunione in Zoom ID riunione: 969 5040 2304 Passcode: 972177 Referente: Carlo Combi . Tue, 10 Oct 2023 10:30:00 +0200 Single cell analysis Relatore: Luciano Cascione; Provenienza: SIB Swiss Institute of Bioinformatics; Data inizio: 2023-10-05; Ora inizio: 10.30; Referente interno: Rosalba Giugno; Riassunto: This minicourse aims to provide knowledge and understanding of algorithms and advanced programming tools for managing single-cell data. Unlock the potential of single-cell data analysis as we answer pressing biological questions where cell-specific changes in the transcriptome are crucial. These questions include discovering new or rare cell types, identifying differential cell composition between healthy and diseased tissues, or understanding cell differentiation during development. Throughout this course, you#39;ll explore a rich array of computational and statistical methods, all at your fingertips, ready to empower your data analysis. To make this experience even more exciting, we will apply your newfound knowledge to a real-world case study, bringing theory to life. Schedule Oct 5: 10:30-13:30, Sala Verde (Piramide) Oct 6: 08:30-11:30, Sala Verde (Piramide) Oct 12: 10:30-13:30, Sala Verde (Piramide) Oct 13: Room and hours to be defined. Thu, 5 Oct 2023 10:30:00 +0200 Machine Learning and Game Theory: Recent Novel Solutions for Recommender Systems and the Smart Grid Relatore: Georgios Chalkiadakis; Provenienza: Technical University of Crete; Data inizio: 2023-09-27; Ora inizio: 14.30; Note orario: Aula Tessari (presenza e da remoto); Referente interno: Alessandro Farinelli; Riassunto: Abstract In the first part of this talk, I will be describing recent research conducted in my group to tackle both the personalized and group recommendations problems, using Machine Learning and Social Choice Theory. More specifically, we build our (Bayesian) recommendation algorithms following the ldquo;you are what you consumerdquo; (Babas et al, 2013) principle, while a key novelty in our approach is the application of multiwinner voting rules to increase recommendationsrsquo; diversity and fairness. We conduct a systematic experimental evaluation of our approaches by applying them on a real-world dataset of Points of Interest (POIs) in the popular touristic destination of Agios Nikolaos, Crete; also exploiting information collected via questionnaires from actual tourists visiting the city of Agios Nikolaos. Our experimental results (i) highlight the ability of our systems to successfully produce personalized recommendations that match the specific interests of a single user; (ii) confirm that the employment of prior knowledge regarding the preferences of tourists, based on their demographics, guides our recommender to avoid the cold-start problem; (iii) demonstrate that the use of multiwinner mechanisms allows for diverse recommendations with respect to travel-related features, and increased system performance in the case of limited user-system interactions; and (iv) show that the use of multiwinner mechanisms allows for fair group recommendations with respect to the well-known m-PROPORTIONALITY and m-ENVY-FREENESS metrics. Last but not least, our personalized Bayesian recommendation algorithm is incorporated in a real-world mobile tour-planning application for Agios Nikolaos, Crete. In the second part of this talk, we shift our focus to a novel Smart Grid Flexibility Aggregation framework, encompassing a multiagent architecture and various types of mechanisms for the effective management and efficient integration of Distributed Energy Resources in the emerging Smart Grid. One critical component of our architecture is the Local Flexibility Estimators (LFEs) agents, which are key for offloading the Aggregator from serious or resource-intensive responsibilities--such as addressing privacy concerns and predicting the accuracy of DER statements regarding their offered demand response services. The proposed aggregation framework allows the formation of efficient and effective LFE cooperatives. To this end, we developed and deployed a variety of cooperative member selection mechanisms, including scoring rules, and (deep) reinforcement learning. We use data from the well-known PowerTAC simulator to systematically evaluate our framework in various scenarios based on Smart Grid settings, so the efficiency of the framework can be properly assessed. Our experiments verify its effectiveness for incorporating heterogeneous DERs into the Grid in an efficient manner--showing that the use of appropriate mechanisms results in higher payments for competent LFEs managed by the Aggregator. Short Bio Professor Georgios Chalkiadakis (PhD University of Toronto, 2007) joined the faculty of the Technical University of Crete in March 2011. His research interests lie mainly in the area of Multi-Agent Systems (MAS); and more specifically on decision making under uncertainty and machine learning in MAS environments, coalition formation and cooperation in MAS, and cooperative game theory. He is a co-author of the graduate level textbook ldquo; Computational Aspects of Cooperative Game Theoryrdquo; (Morgan and Claypool, 2011) . He has also co-authored four chapters in scientific collections, and more than 100 research papers in journals and conferences in his areas of expertise (several of which have received awards or award nominations). He has served in the Editorial Board of the Journal of Artificial Intelligence Research; while he has served in the Boards of Directors of the Hellenic AI Society-EETN (General Secretary: 2014-2016; 2016-2018), and of the European Association for Multi-Agent Systems-EURAMAS (Deputy Chair: 2018-2020; Chair: 2020-2022). More details on Georgios and his research can be found at: Aula Tessari (presenza e da remoto) link zoom: Referente: Alessandro Farinelli . Wed, 27 Sep 2023 14:30:00 +0200