In this talk I will present a general framework for building intelligent collectives. A concrete task of prediction market will be described as an example. It will be shown that two aspects (diversity and independency ) play an important role in making proper decisions. I will show that the positive impact of diversity and independence on collective performance. With some restrictions, the hypothesis "The more diverse the collective, the higher the collective performance" is formally proved. The second part of this presentation will concern the application of collective intelligent methods for sentiment analysis. It will be shown that the set of user opinions (sentiments) can be treated as a collective and choosing a proper integration mechanism is important for achieving high quality of the analysis.
Prof. Ngoc-Thanh Nguyen is currently a full Professor of Wroclaw University of Science and Technology in Poland and the Head of Department of Applied Informatics. He is now a member of the State Council of Scientific Excellence of Poland and a member of the Committee on Informatics of the Polish Academy of Sciences. He was an invited and visiting professor in many universities, such as Friedrich-Alexander University (Germany), Vrije Universiteit in Amsterdam (Netherlands), Japan Advanced Institute of Science and Technology (Japan), Yeungnam University (Korea), University of Technology Sydney (Australia), National Kaohsiung University (Taiwan) and Universidad Autónoma de Madrid (Spain). He is the Chair of IEEE SMC Technical Committee on Computational Collective Intelligence, he received the title of Distinguished Scientist and Distinguished Speaker of ACM, and Distinguished Visitor of IEEE. He is the Editor-in-Chief of Transactions on Computational Collective Intelligence (Springer), Vietnam Journal of Computer Science (World Scientific) and Journal of Information and Telecommunication (Taylor&Francis). Prof. Ngoc Thanh Nguyen is an author of 2 European patents in information systems. He has headed a number of research projects funded by EU programs like COST, DAAD, Polonor and National Foundation of Research and Development. He serves as an expert of National Center of Research and Development and of European Commission in evaluating research projects in such programs as MSCA-IF, REA-FET, EUREKA and COST. He has given 24 plenary and keynote speeches for international conferences, and more than 45 invited lectures in many countries. He has published 46 books and more than 450 journal and conference papers. Prof. Nguyen is on the TOP 2% list of scientists whose publications are most cited by other authors, compiled by Stanford University and Elsevier. He is a recipient of the Gold Cross of Merit awarded by the President of Poland in 2022.
Disadvantaged communities across the world are disproportionately affected by a lack of access to healthcare with high societal and individual costs. Conditions such as congenital heart disease and brain seizures can have devastating effects if not detected in time. The main issue is that there is often a lack of trained personnel and/or expensive equipment to detect these conditions. Advances in the Internet of Things and AI technologies can provide reliable, low-cost solutions to detect these conditions without disrupting existing hospital protocols and operating within a secure framework. In our approach, solutions that involve the communion between the medical professional and AI are of special interest. These solutions must be explainable to gain wider acceptance in a medical setting. The presentation will discuss some implementations using EDGE and IoT platforms for an AI-driven auscultation for congenital heart disease and AI-driven sonification of brainwaves. Some aspects of IoT/EDGE security will also be briefly discussed.
Special thanks to my colleagues: Prof. Andriy Temko; Dr. Andreea Factor; Prof Viktoria Shelevytska, Prof. Volodymyr Sarana, , Dr. Sergi Gomez Quintana, Feargal O’Sullivan, Lavanya Pampana, Dominic Lightbody, Duc-Minh Ngo, Tien Van Nguyen, Giuseppe Carracciolo, Leah Twomey, Adam Creed, as well as our funders and supporters.
We are expecting the era of Trillion Sensor Universe(TSU), when trillion of sensors are distributed to environment. Since most of the distributed sensors needs to operate for long time without maintenance, the sensors should not utilize power of batteries. The combination of energy harvesting power supplies and low power sensor nodes is the key technology to realize the TSU. In this keynote talk, I would like to introduce challenges of energy harvesting technologies, especially focusing on RF energy harvesting and Wireless Power Transfer (WPT) technics for power supply of sensors for IoT. Then I will show characteristics of energy harvesting BEAT Sensors which operate for 24 hours a day without batteries, and achieve the communication range of more than 5 km. Energy saving lighting system using energy harvesting illumination Beat Sensors and the system operation will also be introduced.