PhD candidates and post-docs will present their research in the session 9.30-10.30h. The following presentations will be given:
Digitalization among SMEs
Erzsi Meerstra (Faculty of Economics and Business)
Theme: Digital Business
Abstract: Our research is strongly linked to the Digital Business theme. Yearly we conduct the innovation monitor among SMEs, this year's theme was digitalization. We have gathered data about the digital performance, digital readiness and digital businessmodel of SMEs in the northern region of the Netherlands. We would like to present our findings in more detail during the poster session.
Online shopping and offline shops: equal product purchases, less product returns
Christian Hirche (Faculty of Economics and Business)
Theme: Digital Business
Abstract: The ongoing transformation to a digital society encompasses a change in shopping patterns. Consumers increasingly choose the Internet over physical stores to purchase a wide range of products. Therefore, many businesses reallocate resources from physical to online channels. In our study, we find that online shopping behaviour is also influenced by offline shops, however, and therefore both need to be considered together. In particular, physical stores help by lowering online product returns for some types of products (by up to 13.91% - 16.26 %) and do not harm by not influencing online product purchases. Since high product returns are a burden for businesses and society alike – due to more waste, larger needed fleet of delivery vehicles and decreased profitability – we show that careful management of the digital transformation in retailing means to also be observant of the effects of the offline on the online world.
Which cognitive skills are important for learning computer programming?
Irene Graafsma (Faculty of Arts)
Theme: Digital Literacy
Abstract: There has been a growing interest in teaching students computer programming to prepare them for the demands of our increasingly digital society. Computational thinking and computer programming are often referred to as the literacy of the 21st century. Although the teaching of computational thinking & programming has long been underway across the globe, knowledge of the cognitive skills that underlie programming is still very limited. As a result, we lack critical understanding of how programming skills can best be taught. In my PhD I aim to close this gap in our knowledge by studying which cognitive skills are important when learning computer programming for the first time at a university level. The results of my first study show that algebra, logical reasoning and word learning skills are important predictors of students’ success in their programming course, and that students’ pattern recognition skills improve after a semester of learning to program. These results will inform better teaching methods and help us understand the broader benefits of incorporating programming into education, thereby contributing to the digital literacy of current and future generations.
A new path for mapping and understanding stakeholders' ecologies
Nuccio Ludovico (Faculty of Science and Engineering)
Themes: Digital Technology & Digital Literacy
Abstract: We present a methodology to study the Dutch energy transition hyperlink network. By combining web scraping techniques with computational text analysis, it is possible to find and classify stakeholers involved in the energy transition process in the Netherlands. Through this semi-supervised procedure, commercial, institutional and non-governmental actors have been automatically mapped in terms of area of activity, geographical position and degree of influence in the digital landscape. Furthermore, the analysis of the communication acts has made it possible to assess the degree of involvement of each stakeholder in the energy transition process. Thanks to its scalability and flexibility in application, this methodology can be a useful tool to support decision-making processes in the business and institutional environment.
Tracking water flow and traffic without sight and sound: combining AI and novel fluid sensing.
Ben Wolf (Faculty of Science and Engineering)
Themes: Digital Technology, AI/Data Science
Abstract: We developed and built a fluid flow sensor array using a novel sensing technique, which is based on a unique sensing modality found in fish and amphibians. With this array and AI methods, we are able to detect and localize nearby moving objects in water from their near-field fluid interaction. Furthermore, we demonstrate that we can detect and identify the shape of an object passing by the sensor array in a controlled setting. The poster would further highlight the relevant findings and applications to water and recreational waterway management.
3D-printed microsensors for biomedical flow sensing applications
Amar M. Kamat (Faculty of Science and Engineering)
Theme: Digital Technology
Abstract: The recent trends of personalized healthcare, autonomous sensor networks, digital technology, and the Internet of Things (IoT) have made microsensors essential to the healthcare industry. For example, accurate flow sensing is a key requirement in many biomedical applications including infusion pumps, insulin pumps, respiratory monitoring, and urine flow monitoring. This can enable continuous monitoring of vital patient signs, medicinal flow rate verification to avoid potential adverse drug events (ADEs), reduction in nurse workload through automated flow rate measurements, prediction of device malfunctions, and so on, paving the path towards ‘smart wards’ in hospitals powered by digital technology. Current methods of flow microsensor manufacturing involve standard microelectromechanical systems (MEMS) ‘cleanroom’ fabrication techniques which are expensive, tedious, and restricted to a limited number of materials, resulting in microsensors that are often prohibitively expensive for medical applications. To solve this problem, we propose the use of high-resolution 3D printing, soft lithography, and the use of piezoresistive nanomaterials to develop biomimetic flow microsensors inspired by the ultrasensitive ‘hair-like’ neuromast flow sensors found in the blind cave fish lateral line. Using designs compatible with high-volume manufacturing methods such as injection molding, our idea can thus be used to develop sensitive, low-cost (< €1), and disposable microsensors geared towards flow sensing applications in the healthcare industry.
Measuring Conversion Attribution by Machine Learning
Jan Piet Peeperkorn (Faculty of Economics and Business)
Theme: Digital Business
Abastract: In Digital Marketing literature there is a lack of proper attribution model to estimate advertisement effectiveness on an individual level. We constructed a Machine Learning method to estimate the probability that customers purchase a product online. The ROC of the prediction results have a rate of 97%, making it highly predictive. Conversion credits are then assigned to different marketing channels based on their relative contribution as defined by the Shapley value framework, which is proven to ensure fairness and is easy to interpret. This gives practitioners the ability to observe the contribution of each individual touchpoint, allowing for more precise measurements of return on investment. Moreover, marketing strategies could be adjusted, now that the company is aware of which customers are actually interested in their products.
Why social regulation online fails: more fudge is needed
Carla Anne Roos (Faculty of Behavioral and Social Sciences)
Abstract: In my PhD project, entitled "Social Regulation in a Digital Society", we investigate the way in which people handle disagreement while maintaining good social relations in text-based online as compared to face-to-face small group discussions. We find that the clarity and unresponsiveness in online communication hampers the successful navigation of disagreement. Based on these findings, we aim to design interventions to improve social regulation in online environments.
Planned obsolescence in hybrid products
Tim van Zuijlen (Faculty of Law)
Research shows that the lifespan of consumer electronics is getting shorter. This has a negative impact on our environment: 80% of our discarded smartphones, notebooks and printers is not being recycled, but dumped or burned as ‘e-waste’. According to the e-Waste Monitor, yearly, this waste equals the weight of over 4500 Eiffel Towers. Shorter product lifespans have a negative impact on the users of these products as well; they have to make more repeat purchases. The fact that manufacturers – apparently – are not able to develop more sustainable products has led to suspicion. Are lifespans of products artificially kept short, or even shortened, to force consumers to make repeat purchases more often? In other words: are producers ‘planning’ the obsolescence of their products?
Recently, Italian and French authorities accused some smartphone manufacturers of slowing down their old devices deliberately via an update in order to seduce or force users to buy new devices. Although these accusations are recent, smartphone users have been experiencing slowdowns of their devices for years. It has even gotten its own name: 'the slow iPhone phenomenon'. Although the Apple has become somewhat of a poster boy for this 'planned obsolescence', it is far from the only company that has been accused of these practices. Notably, printer manufacturers have been accused of incorporating software in their devices that let the printer indicate that the ink in the cartridge is depleted, while in truth the cartridge is still half-full. This leads to consumers buying new cartridges – or whole new printers – earlier than necessary. Are manufacturers allowed to artificially shorten the lifespan of their products in order to increase their revenue? To what extend does current law promote or dissuade the development of products with longer lifetimes? Are there legal measures that could contribute to longer lifetimes for products?
Practising news literacy How young people develop and express skills, competences and knowledge around news in the context of social media
Joëlle Swart (Faculty of Arts)
Social media have become pivotal for young people to orient themselves to the world around them. However, their frequent social media use does not necessarily mean they are news and/or media literate. Skills like recognizing political bias on news or understanding how algorithms impact news selection online are far from self-evident. This raises questions what competences, skills and knowledge students nowadays need to be taught to enable them to navigate the current news landscape. This study explores when news literacy becomes useful, starting from the practices of youngsters themselves. Using in-depth interviews, it asks how students in vocational education develop news literacy and how this impacts their practices, strategies and tactics when using news on social media. It finds that while most young people possess basic news literacy, they often lack more comprehensive knowledge about the news industry, media production, media effects and the democratic role of news and journalism that could help them navigate the contemporary news landscape more effectively. Moreover, it argues news literacy initiatives should move beyond enabling youth to use existing platforms and devices and enhance their agency to handle implications of future technologies. Finally, the study highlights the importance of going beyond protectionist approaches, equipping youngsters with the knowledge and skills to employ social media in a manner that empower them as citizens instead.
Transparent automatic genre classification of newspaper articles
Kim Smeenk (Faculty of Arts)
Systematic study of genre in newspapers sheds light on the development of journalism discourse. The genre conventions that can be discerned in a newspaper text signal the underlying discursive norms and practices of journalism as a profession. Digital newspaper archives provide the opportunity for large-scale empirical research. However, they do not contain the fine-grained genre information that is required for this purpose. The NEWSGAC project has adopted a machine learning approach to add genre labels to newspaper articles. Classifying genre in a standardized and reliable manner is challenging, though, because genre is a typical example of a ‘latent’ content category, which needs considerable interpretation. To ensure the reliability of the results and to evaluate the machine learning approach, it is crucial to make the methodological impact of various machine learning pipelines transparent.
Several posters on research in smart mobility
|Last modified:||27 September 2019 2.41 p.m.|