KEYNOTES

Design and Analysis of Information Granules and Rule-based Architectures with Federated Learning

In data analytics, system modeling, and decision-making models, the aspects of interpretability and explainability are of paramount relevance, just to refer here to explainable Artificial Intelligence (XAI). They are especially timely in light of the increasing complexity of systems one has to cope with and ultimate concerns about privacy and security of data and models. With the omnipresence of mobile devices, distributed data, and security and privacy restrictions, federated learning becomes a suitable and practically viable alternative.

Information granules are building blocks forming the interpretable environment capturing the essence of data and key relationships existing there. Their emergence is supported by a systematic and focused analysis of data. At the same time, their initialization is specified by stakeholders or/and the owners and users of data.  We cover a comprehensive discussion of information granules-oriented design of information granules by engaging an innovative mechanism of federated unsupervised learning in which information granules are constructed and refined with the use of collaborate schemes of clustering. The development of fuzzy rule-based models engaging federated learning is also discussed.

Computing with Words in Decision Making: Its Application to Climate Policy

The concept of computing usually implies calculation processes either by mathematical means of numbers and symbols or by a computer. Paying attention to computing processes done by human beings, it is remarkable that they employ mostly words in computing and reasoning, arriving at results linguistically expressed from linguistic premises. Hence, Computing with Words (CW) applies the same view to their computing processes aiming at obtaining linguistic outcomes from linguistic inputs. Because words have fuzzy denotations when they are used by human beings, the paradigm of Computing with Words was clearly stated as a branch of fuzzy logic.

The complexity of decision-making problems implies uncertainty that on many occasions has been modelled by means of linguistic information, mainly based on fuzzy-based linguistic approaches. Several of these approaches have been successfully applied in decision-making to address challenges and resolve problems associated with environmental, energy, and climate policy. Therefore, it is important to point out the trends in CW and its application for climate policy for supporting policy makers together with the challenges ahead.

Emergent and Evolvable Ports: How this concept might be for Latin American Ports

The port model prevailing in Latin America (LATAM) is the result of the major reforms of the decades of the 1990s and 2000s. It allowed for substantial achievements as regards the reduction of foreign trade port costs. Changes in the economy may have been influenced by improvements in port performance and operations; though the improvement in seaport management and technology in LATAM has not been quite as impressive as in other regions around the world. The economic environment in which ports operate daily generate expectations that must address not only the creation of better infrastructure and technology but also its adaptability to challenges in the global environmental landscape. Port authorities in the Port LATAM region must re-define appropriate strategies to specify relevant core businesses and competencies to position ports for sustainability. Built on competitive grounds and thus finding the proper context/conditions opportunities for cooperation, collaboration and coordination among ports will emerge into the development for long-term future success. However, shortcomings in reliable data and information exchange often hamper collaborations and partnerships among the initiatives. The challenge goes beyond information technologies because intelligent business processes must be embedded in suitable solutions as the key factor for port competitive advantage. The set of integrated tools supporting businesses and IT users in managing process execution is known as Business Process Intelligence (BPI). The use of computational intelligence tools and methods involving knowledge discovery and computer-aided strategic decision-making are also relatively new endeavours in the port domain.

Artificial Hypothalamus: Artificial Intelligence and Mathematical Programming Integration

Today when we closely examine the supply and demand chain process of any industrial sector we find significant operational latency with poor visibility and the inability to fulfill demand resulting in lost revenue. To overcome these inefficiencies, we need to focus on the core fundamentals of the enterprise. At the nucleus of the enterprise, what we like to call the hypothalamus, it is extremely necessary to have ample knowledge and visibility of the entire chain process. A meticulous composition and relationship between raw materials, services, products, and customer-centric demands. This conference focuses on what is the hypothalamus of an enterprise and how is it built. Like the human hypothalamus key functions and drivers are universal. Yet each hypothalamus is different. They vary in acquiring knowledge through experience where some of us have more knowledge due to experience than others. Or in the case of an enterprise the acquisition and processing of information enables the discernibility of sharp insight that delivers impactful decisions when you most need them. The hypothalamus while bespoke to each enterprise and its business model needs to have a coherent information system. A tactical design throughout its entire process in order ensure it sees supplies and demands with the assurance to fulfill them and leverages the opportunity. The essential elements of this design must occupy a robust framework of algorithms comprised of artificial cognitive agents that model the world and solve problem like neurons in the brain. A carefully structured series of intelligent and autonomous algorithms establish a holistic view of a problem, self-learn, and solve it by using high-complexity mathematical modeling. The artificial hypothalamus supersedes a neural network made up of point-neurons. It is a dynamic and self-organizing sparse network culminated by the integration of multi-function cognitive agents operating simultaneously in real-time like the human hypothalamus. Envision a future not so distant where autonomous algorithms can plan and operate a manufacturing facility, seamlessly run smart cities, optimize net zero transportation fleets, or automate hospitals for better care. All to elevate quality of life while preserving the balance of nature. As our global population grows exponentially the supply chain relationship has become disproportional to demand. The need to incorporate an artificial hypothalamus goes beyond fulfilling demand, our survival and sustainability depends on it.

Organizers: Fernando Padrón, Edgar Pedraza and Carlos Llorente

Tecnológico Nacional de México / Instituto Tecnológico de Cd. Madero, Mexico

Business Analytics (BA) Systems are toolboxes that need user knowledge about algorithms, as well as mathematical and computer sciences theories. On the other hand, Business Intelligence (BI) systems have been evolving towards frameworks which show by diverse sophisticated visualization ways and dashboards, information about the decisión making problem, using specific business analytics techniques. BA and BI systems are not used by decisión makers, such as CEOs and entrepreneurs, the real owners of business ideas and knowledge about them, limiting the effective direct use of those systems in business decision-making real problems.

Eureka-Universe: The Universal Analyst is a unique BA and BI system which allows solving the following independent tasks and mixing them towards Decision Making Problems solutions, without needing any users knowledge about mathematics and computers sciences theories and algorithms; just their own knowledge about the business and common sense for the interaction with a graphical and based on natural language editor.

Eureka-Universe has been possible because of the scientific results of the cooperation into the framework of Eurekas Community, and its networks Eureka Iberoamerica and Eureka International. They have elaborated transdisciplinary theories based on Fuzzy Logic, which allows modeling and interpretation by natural and professional languages for the analysis of metacognitive systems and processes like evaluation, decisión making, reasoning, as well as logical and statistical inferences.

Knowledge Discovery is possible through Eureka-Universe from the Discovery, evaluation, and inference of fuzzy predicates. Eureka-Universe allows the creation of projects starting on a database.

PUBLICATIONS