One of our customers is a company providing tablet education to primary schools in the Netherlands and abroad. Their solution is technology driven and offers an adaptive learning platform to maximize the learning progress of each individual pupil.

In this presentation I will talk about machine learning algorithms in the context of adaptive learning. These algorithms leverage the collected data (exercises made by pupils) to further improve adaptive learning. As such, these algorithms materially contribute to the realized learning progress of each pupil during primary school.

After an introduction to this case, I will zoom in on the technical side and also give you some insight in the development of real-life machine learning applications like this.

We are now at the beginning of the fourth industrial revolution. The third industrial revolution, also called the digital revolution, concerned the development of computers and IT systems. The fourth industrial revolution builds on this, but is considered a new era in view of the speed, the scope and the enormous impact it has.

The speed with which technological breakthroughs present themselves is unprecedented. There is an exponential growth instead of a linear growth. These developments will lead to radical transformations all over the world. Complete production, management and governance systems will be adapted.

An important development is the Internet of Things (IoT). In smart industry, manufacturing companies use more and more sensors to measure such dimensions as temperature, humidity, pressure, water quality, gases and chemicals, smoke, infrared, fluid levels, images, and movement and acceleration. The Internet of Things is growing very fast and it is expected that by 2020 30 billion objects will be part of this network.

Cisco expects that in 2020 these objects will generate about 2.5 quintillion bytes of data every day (that’s 2.5 followed by 18 zeros). AI is needed to extract useful information from all this data. The information collected through sensors in machines by manufacturers create opportunities to better support their customers. Services are playing an ever-greater role in the business model of manufacturing companies. This migration is called servitization.

During her talk Els van de Kar will give examples of her experience on managing ICT in Practice as an entrepreneur in this era. Next, she will give examples from the IoT projects ‘IoT4MKB’ and ‘IoT in the Fitbox’ which illustrate how companies in the new era are working on their future business models by implementing IoT.

AI creates great opportunities but also great challenges. Challenges for the future but also for the present. If the brilliant Gary Kasparov could not beat a primitive chess computer, can we beat an AI that is designed to grab our attention? If we have these big problems understanding ourselves, how do we deal with an AI that has no problems understanding us? These are important questions that are already taking shape around us.

If we want to have a full life in this world filled with advanced technology, we will have to learn to relate to this new world. This is a skill and it has to be learned. Fontys University is running a program called Technophilosophy to do just that. The program learns students that relating to technology is very important and can also be fun!

The best results in fast and practical deployment of AI systems usually involve a smart combination of human intelligence and machine intelligence. We present a project (‘Van A naar Eerder’) going live in the Netherlands where an AI system will save millions of traffic jam costs through faster detection of accidents. From conception to implementation within six months, and a clear roadmap to double the gains in the near future.

Starting a data science project is always challenging: you never know what data is available, what the results will be if any, and how useful or insightful the answers are. For that reason, we developed an approach to design, develop and test potential results within a short timeframe.

Our approach has a foundation in concepts from Design Thinking, Data Science and the Lean Startup to maximize results. The idea is to use validated learning as a process to develop a data science solution by means of prototyping with the end users involved. It incorporates data hypotheses, GDPR analyses, user interface design and a valid and pragmatic user validation study.

At the end of the Design Sprint, you know what a potential solution could look like, what data sources you could use, how useful they are for answering your business challenge, what privacy implications are involved and what other aspects you should take into account when you want to move forward. We will illustrate this in-depth with a practical example project.

How can technology help to create more resilient communities?

Society depends on active citizens doing good to make their communities a better place. Though many people feel the need to contribute to their community, many people lack time or money to become active for causes they care about. They get stuck in their job - or looking for a job - while daily costs and commitments such as taking care of family members, prevent them from investing their precious time in social issues they really care about.

What if we could remove these barriers and empower people and their communities? In the end, we all want a healthier, more social and sustainable world.

During this talk I will present the Possible Pilot Platform that we are currently testing in Eindhoven together with Fontys University and local volunteering groups. This platform not only matches social initiatives and volunteers, but also contains an advanced volunteer recognition system based on blockchain. Inspired by Greek timebanks that make people more resilient in times of crisis, the platform records volunteering hours and values time with Community Coins. These coins can be used for discounts at retailers participating to lower daily costs and stimulate the local economy. On top of that people can build their social resume or earn badges in social projects. This way, the platform encourages more people to contribute their time and skills to the benefit of local communities, while their transaction histories on blockchain help to make social impact measurable and transparent.

- The pace of change will never be this slow again -

AI is the buzzword, but is it actually changing our cities, society and even mankind? It is and it’s happening at dazzling speed. The potential to improve the quality of life is huge, but the downside and ethical challenges we face are becoming more visible every day. How do we cope with this? We need to re-invent ourselves as society; not in a few years but now. And AI can help us with that.

But AI is one ingredient in this societal disruption. With the worldwide implementation of 5G we are at a point in time in which the digital disruption will start accelerating. It’s happening, so let's imagine and define in which city we want to life.

This talk provides an overview of AI techniques for perception and action selection in robots. Modern autonomous mobile robots, like Autonomous Guided Vehicles (AGVs), face difficulties when acting in an unstructured world, where they have to cooperate with humans. Key problems are intelligent perception and action selection. AI provides solutions to tackle these complex problems. This talk provides an overview of various AI approaches, such as behavior trees and deep (reinforcement) learning. Examples are provided from our research on autonomous mobile robots.

(Pre-built, Customized & Custom AI)

In this session we will look at demystifying AI and the need to get started with experimenting. Applying AI can be done in different stages/levels which will be covered in this talk. From pre-built AI (example Azure Cognitive services) up to custom built AI. An important part of getting started is to know what kind of roles are at play, this will be explained by use of a kitchen Analogy.

Forecasts of future trends in psychology show that psychology and technology will become ever more intertwined. In this talk I will present 4 developments illustrating how technical people and psychologists working together can have a positive impact on research and design.

The availability of large volumes of data enables psychologists to study detailed aspects of human behavior in a real-life environment rather than in a controlled lab environment only. For this, smart algorithms looking for patterns and exemptions in the data are required, as well as sophisticated visualization methods.

Technological developments are rapidly changing the kinds of products that are designed an made. At the same time, people have different wished and demands concerning the products and services they use. For instance, experience has become far more important than mere functionality. Together, technological people and psychologists can create more intelligent, pleasurable and personalized products and services and understand how these products impact well-being, inclusion, relationships, etc.

Constant checkers of social media may experience stress, absent presence is annoying, and online misuse and abuse happen regularly. Fortunately there are also positive effects of social media; it helps us keep in touch with friends and family far away, for instance. Technologists and psychologists have a shared responsibility in helping people have a good and healthy relationship with technology.

Evidence-based technological applications can support, enhance or sometimes even replace current medical treatments for mental health disorders, for instance Virtual Reality used for the treatment of anxiety, or mobile apps helping people control or treat their addiction to drugs, alcohol etc. Together, psychologists and technologists can design and develop reliable products supporting both the therapist and the client.

Our lives are influenced by AI in more ways than we can imagine. From which shows Netflix recommends to whether or not you can get a loan, AI is used to make decisions that can have a profound effect in our lives. It is easy to think that machines are impartial and make decisions based on facts, but in a world where facts are alternative the AIs we create are heavily influenced by our own biases.

In this talk, we’ll explore the ethical implications of how AI is shaping our lives, and whether we can ever have truly unbiased algorithms.

In his career Bram has come across many different companies looking for ways of leveraging analytics, big data pipelines and data science techniques to keep up with the huge demand and to stay ahead of the competition. He will take the audience through a business case in the customer helpdesk domain situated within a Scandinavian company, their main issue is dealing with the increasing number of emails they receive every day. Their main question was how to start responding automatically to certain incoming emails, to lower the workload for the employees at the helpdesk but still being able to deliver their personal touch to their customers. In this project we made a system that understands the content of an incoming email, answers the emails if possible and prioritizes the emails that should be handled manually. In this talk, Bram will talk about which techniques were used, which problems were faced along the way, for example dealing with a different and infrequently used language, and how these problems were finally overcome to create a solution that the client can use to improve their business and solve the problem they faced.

The Soft Factors of Organizing Beyond the Proof of Concept

More information about this talk soon!

The Nano4Sports EurReg project addresses the problem of sports injuries resulting from deficits in the running technique, the use of smart technology and meaningful feedback in sonification and music.

Many people listen to music while running for motivational reasons but also to keep up the pace (tempo). This implies a fundamental relationship between music and movement that has been studied broadly from different perspectives such as neuro physiology.

Using this knowledge combined with sensor data Nano4Sports aims to provide amateur runners with audio feedback in the form of adaptive music and sonification (of their footsteps) in order to help them to improve their running technique and so preventing sport injuries.

Because the application of persuasive sonic design is highly personal, machine learning is used to give feedback at an individual level. Furthermore the project aims to identify musical elements (in structure and expression) that trigger changes in movement. These insights are used to define design patterns for persuasive sonic design but also to explore the possibilities of controlling other AIs (such as Google’s Magenta) to create the music that is the most effective for the individual runner and specific circumstances.

Open Educational Resources (OER) are digital learning materials available for free on the Internet which provide the user the rights to reuse, adapt and redistribute. For teachers, OER provide more opportunities to diversify their lectures by reusing available OER. An exact number of OER available worldwide cannot be given, but estimations run into many millions. To better find these resources, in many cases metadata are added. However, adding metadata is a tedious task, best done by a professional (e.g. a librarian). These professionals are not always available. Also, standards used for metadata (e.g. IEEE LOM) do not provide topic specific information a teacher uses when searching for OER. Therefore, low findability of the right resources for specific contexts is experienced as a barrier for teachers, thereby hindering adoption of OER.

To overcome these barriers, machine learning techniques are developed to automatically add metadata to resources. In this talk an overview will be given of experiments in this area and the first results of these experiments.

"Being physically active in an era of low–cost technology generating massive amounts of personal and contextual data"

In the last five to ten years, developments in ICT and sensor technology have resulted in a wide range of application fields such as health related smartphone applications, activity trackers and sports watches. These innovations in smart wearable body sensors have created possibilities for people to become ‘experts’ in self-monitoring of health parameters and physical activity.

One of the game changers in these recent advances in technology is their ability to track behaviors in daily life situations over time, and across a large number of users. The capturing of temporally, behaviorally and ecologically contextualized data from their users generates large datasets. Through data analytics and data visualization (i) interesting patterns can be discovered, (ii) predictions can be made, and (iii) fundamental understanding is raised about behavioral patterns and the influence of individual, social and environmental factors, and systems and services designed to empower people to have an active lifestyle.

In this presentation, these developments will be discussed, and concrete examples of data analytics and visualizations and designed systems and products will be shown to highlight the opportunities that sensor- technology and data offer.

With some form of AI perennially at the peak of the ‘hype cycle’, how are we finding useful applications for machine learning? Practitioners face many roadblocks, such as the availability of data, implicit bias, privacy considerations and overblown expectations. Transparency (or ‘explainable AI’) is seen as a particular challenge. These concerns are not unfounded. We present ways of addressing these concerns in a technology agnostic manner, with concrete examples.

Deep learning has shown that it is able to achieve incredible performance on certain tasks. Lately, surfing on the hype wave surrounding chatbots, natural language understanding has been a huge focus for businesses investing in machine learning. One of our clients asked us to help automate their document processing pipeline by creating a deep learning module that would extract relevant information from contracts automatically. By training a deep learning model naïvely, a surprisingly good performance was already reached.

However, it was far from enough. After investigating, many mistakes made by the model could be explained by the ambiguity surrounding the documents it had to deal with.

Tables were difficult to find, words would be laid out physically on the sheet in a way that was difficult to translate into a sequence of words. These documents were created to be read by humans, not computers. It is often argued that the goal of artificial intelligence is to reach human-level cognition. But it doesn’t need to be such a one-sided effort. Humans can help bridge that gap by rethinking the way they design, by taking machines into account. For instance, road signs could be designed to be easily recognized by computer vision systems, fonts could be modified to improve optical character recognition, or products and packaging could be shaped in ways that robotic arms can more easily handle them. If artificial intelligence will help us in the future, we need to help it today.

Sociologists agree: generations are made during the so-called “formative years” between the ages of 10 and 25. Today’s children develop in two ways. Physically, but now also digitally. This is the first generation to come of age with this dual identity. What young people do online is a reflection of the uncertain and changing world in which this generation is growing up. Besides being leaders of this apparent dual life, they are the first generation to try to create a new order from this mix of realities. For them, the internet is both a meta-modernist amusement park where every reality can be created, and a place that is covered with a moral fog and where new rules of conduct arise. Generations do not get caught in a specific piece of technology. The character of a generation is never the technology itself. The education, the zeitgeist and everything they consciously experience in their youth play an important role.

Volume, value and influence
A new generation is coming to the fore. Substantial in size (32 percent of the world population), it has different values than its predecessors, and is already influential thanks to its use of influencer practices and technologies.

The synthetic generation
The task of making something of the world rests on the shoulders of every new generation. They must set their course in this rapidly changing and confusing world. Their mission and expertise are the synthesis. This is not a punk rock generation that wants to destroy things. It is a generation that introduces new values: post-materialistic, focused on self-expression and self-actualization.

Born influencers
The first generation of young people already has a business impact before they start their career. With their ambivalent identity they grow up as “influencers”. On TikTok or Boomoji the little ones learn to exert digital influence. They are experimenting with new forms of influencing like mukbang and ASMR. This generation pays attention to everything that is (or seems to be) authentic and real and turns away from the old world.

The synthetic future
Fake personalities on Instagram become so real that the distinction between reality and virtuality is no longer possible. Authentic images of political leaders are manipulated with synthetic media and forged into new truths. What is real and what is fake is increasingly difficult to determine.

Artificial Intelligence is about to become a commodity like electricity and internet. It will pervade our working and living environments, and reside in home equipment. It will be able to evolve in the local and personal context of the end user. Sufficiently advanced and locally evolved AI will develop behaviours that could be characterised with personality traits. It would be even better if AI designers would deliberately implement personality traits in their designs. This would enable people to relate to the AI in a more effective and more natural way. People project human-like personalities and emotions on animals and even dead objects anyway, so why not design for that? It would surely make the co-habitation of humans and computers more natural.

The impact of AI personality on human-technology relations is an innovation theme at Fontys ICT. Two examples are given. In HealthBots, the effect of tone of voice on the human-computer dialogue was studied. At the Dutch Design Week, FHICT presented De Dingen De Baas, an exploration of the interaction with smart objects with extreme characters. At first sight people hated it, but then some started wondering. Might they love it after all?

Designers and business decision makers want to have a final product that helps create a better world for our children and our common future. We want children to have the best opportunities in life. Digital products have the potential to enhance child development and wellbeing in supporting the natural process that children will go through. The potential of AI in smart toys for children has both utopian and dystopian outcomes.

Issues such as cognitive and physical development need to be strongly considered in a context of developing and changing evidence about the impact that exposure to digital media has on children. Additionally, the role of the family and the effect of a digital experience on a child’s world are to be taken into account when designing for their best interests.

Designers and creators of children’s products, services and content creators should become aware of the principles that cover the most relevant children’s rights in the context of design to ensure ethical design for our future generations. By examples of toys that make use of AI I will explore both promising and threatening aspects of these developments.