Photo: Heriberto Manuel González Dominguez
FIWARE is a technological platform of Future Internet, based on Cloud Computing, using OpenStack (open cloud computing) where software development and hardware integrate components such as accelerators or enablers that serve to implement applications in different areas, particularly FIWARE Mexico Is promoting the use of this technology for developments in smart cities, video surveillance and security, health and energy. The components, technologies and applications developed around FIWARE can be opened or marketed in a closed way.
The course is aimed at developers, integrators and hardware and software vendors interested in generating innovative solutions using FIWARE, a public and open technology platform that provides a sustainable ecosystem for the development of cloud-based applications, Big Data and Internet. things.
Upon completion of the training, participants will be able to understand the potential of the FIWARE platform and gain theoretical / practical skills to develop cloud computing and Internet applications of the Objects using the Orion Context Broker component of the FIWARE platform and the FIWARE cloud "FIWARE Lab".
Read more about the contents and prerequisites for the workshop.
In companies, the use of the so-called business intelligence and knowledge engineering is commonly found. By essence, companies are using computer tools which tend to include them in the knowledge society; from a mathematical point of view, knowledge representation can be made through rules and logics. The question is: what could be done in this direction for urban planning? The big challenge is to deal with urban and environmental features which are usually described, stored and manipulated via computational geometry and spatial analysis. However, those disciplines cannot easily be combined with logics.
The workshop address to show how knowledge engineering can be the foundation of a new type of urban planning, i.e. urban planning based on knowledge. Geographic knowledge bunches are usually described through geographic objects, relations, structures, ontologies, gazetteers, rules and mathematical models. After having explained in details those bunches of knowledge, the structure of a geographic inference engine is sketched so to renovate urban planning. Then beyond Spatial knowledge Infrastructure, we explain that some geographic knowledge infrastructure could be the basis of a new generation of tools for urban planning.
Bayesian models are one of the more successful technologies used for the management of smart cities environment. This paradigm allows to efficiently exploiting prior knowledge and sensed data in order to make tasks such as prediction, diagnosis, control and autonomous decision making among many others. Bayesian models are very successful to develop diverse applications such as energy consumption forecasting, autonomous video surveillance, human mobility and city dynamics anticipation, traffic light control, smart buildings, etc.
The aim of this workshop is to assimilate and master the basics concepts of Bayesian models and Bayesian programming. We present the different types of Bayesian models, their use cases and their implementation using the Python ProBT© library. The workshop is composed of two parts: a theoretical session and a practical session.
Theoretical session. In the theoretical session, participants can concentrate in the modeling framework and basic theory without worrying of the computer implementation details. Participants can examine the principles of Bayesian Programming as well as the use examples without seeing a single line of code. A good balance between basic theory, practical examples, and tutorial-style organization are what makes the course very accessible.
Practical sessions. In the practical session we teach a second level of expertise: how to effectively develop and implement a computer program that includes Bayesian computation. Programs will be developed in the Python ProBT© Library a multi-platform library used to automate probabilistic calculus. The ProBT® library has two main components: (i) a friendly Application Program Interface (API) for building Bayesian models and (ii) a high-performance Bayesian Inference Engine (BIE) allowing execution of the probability calculus in exact or approximate modes.
Requirements. Laptop with Python 2.7 and Python ProBT library installed. Participants only require basic programming and mathematics skills; there is no need to know Python language programming.
About the workshop speaker. Dr. Juan Manuel Ahuactzin Larios is the Innovation Program Manager at T-Systems Mexico. He has over 20 years’ experience driving top scientific and technological research projects, and over 10 years’ experience in systems analysis, design and industry development. He has a solid background as a teacher and mentor in higher education. He is experienced in Bayesian technology, data mining, machine learning, optimization techniques and geometric reasoning. Dr. Ahuactzin has extensive and applicable knowledge in subjects such as robotics, diagnosis and prediction under uncertainty, decision and risk analysis, health informatics and pattern matching. He is author and co-author of several journals articles and book chapters mainly in the areas of robotics and applied Bayesian reasoning. He is co-author of the book “Bayesian Programming” published by CRC Press and cofounder of the French Company ProbaYes now part of the group La Poste. Dr. Ahuactzin is a passionate of academia; has collaborated with different universities in Mexico, Europe and North America such as Instituto Tecnológico y de Estudios Superiores de Monterrey, Universidad de las Américas Puebla, Instituto Nacional de Astrofísica Óptica y Electrónica, École Polytechnique Fédérale de Lausanne, Max-Planck-Gesellschaft, Universidade de Coimbra and .Simon Fraser University.