The current set of research methods on ictresearchmethods.nl contains only one research method that refers to machine learning: the “Data analytics” method in the “Lab” strategy. This does not reflect the way of working in ML projects, where Data Analytics is not a method to answer one question but the main goal of the project. For ML projects, the Data Analytics method should be divided in several smaller steps, each becoming a method of its own. In other words, we should treat the Data Analytics (or more appropriate ML engineering) process in the same way the software engineering process is treated in the framework. In the remainder of this post I will briefly discuss each of the existing research methods and how they apply to ML projects. The methods are organized by strategy. In the discussion I will give pointers to relevant tools or literature for ML projects.
LINK
Background and Context: In order to fully include learners with visual impairments in early programming education, it is necessary to gain insight into specificities regarding their experience of and approach to abstract computational concepts. Objective: In this study, we use the model of the layers of abstraction to explore how learners with visual impairments approach the computational concept of abstraction, working with the Bee-bot and Blue-bot. Method: Six blind and three low vision learners from the elementary school level were observed while completing programming assignments. Findings: The model of the layers of abstraction, can overall be generalized to learners with visual impairments, who engage in patterns that reflect iterative actions of redesigning and debugging. Especially our blind learners use specific tactile and physical behaviors to engage in these actions. Implications: Ultimately, understanding such specificities can contribute to inclusive tailored educational instruction and support.
MULTIFILE
DOCUMENT
The main goal of this study was to investigate if a computational analyses of text data from the National Student Survey (NSS) can add value to the existing, manual analysis. The results showed the computational analysis of the texts from the open questions of the NSS contain information which enriches the results of standard quantitative analysis of the NSS.
DOCUMENT
In order to guarantee structural integrity of marine structures in an effective way, operators of these structures seek an affordable, simple and robust system for monitoring detected cracks. Such systems are not yet available and the authors took a challenge to research a possibility of developing such a system. The paper describes the initial research steps made. In the first place, this includes reviewing conventional and recent methods for sensing and monitoring fatigue cracks and discussing their applicability for marine structures. A special attention is given to the promising but still developing new sensing techniques. In the second place, wireless network systems are reviewed because they form an attractive component of the desired system. The authors conclude that it is feasible to develop the monitoring system for detected cracks in marine structures and elaborate on implications of availability of such a system on risk based inspections and structural health monitoring systems
DOCUMENT
From the article : "In this paper the implications of different research approaches and methods are illustrated by using two projects of the authors. Both projects take place in the same context: exploring participatory innovation within Small-to-Medium sized Enterprizes (SMEs). The main aspects coming forward when comparing the research characteristics of both projects are the importance of time and momentum, the structural set up of the project, people or participants and the abilities of the people involved. The research goal and the background of the researcher are main determinants for the chosen research methods. We hope with this paper to make researchers aware of the implications of the research methods and approach on the results of the project."
DOCUMENT
This paper shows an overview of design practices of the XR-lab at the Amsterdam University of Applied Sciences, The Netherlands. Over the course of six years, interdisciplinary teams of students have delivered 55+ prototypes in virtual, augmented, and mixed reality for a variety of 40+ clients. As human-computer interaction is entering a new evolutionary phase towards human-computer integration, new opportunities in extended reality (XR) have the potential to fundamentally alter human characteristics and abilities. Therefore, this paper begins with taking a philosophical stance on ‘being human’ and the anthropological concept of ‘liminality’ in XR-experiences. A further exploration of the concept of 'emotional rehearsal spaces' uses know-how from performance art, dance, architecture, and dramaturgy. Insights from tangible practices at the XR-lab show the cultural journey in XR-collaborations. This is made visible through a quick and dirty experiment on artistic thinking, design thinking, and system thinking, which shows how interdisciplinary collaborations are able to ignite new combinations of thought in design teams and individual professionals. Finally, we show an overview of specific design methods and tools that have been explored at the XR-lab over the years.
DOCUMENT
Background: While the interest of health promotion researchers in change methods directed at the target population has a long tradition, interest in change methods directed at the environment is still developing. In this survey, the focus is on methods for environmental change; especially about how these are composed of methods for individual change ('Bundling') and how within one environmental level, organizations, methods differ when directed at the management ('At') or applied by the management ('From'). Methods: The first part of this online survey dealt with examining the 'bundling' of individual level methods to methods at the environmental level. The question asked was to what extent the use of an environmental level method would involve the use of certain individual level methods. In the second part of the survey the question was whether there are differences between applying methods directed 'at' an organization (for instance, by a health promoter) versus 'from' within an organization itself. All of the 20 respondents are experts in the field of health promotion. Results: Methods at the individual level are frequently bundled together as part of a method at a higher ecological level. A number of individual level methods are popular as part of most of the environmental level methods, while others are not chosen very often. Interventions directed at environmental agents often have a strong focus on the motivational part of behavior change. There are different approaches targeting a level or being targeted from a level. The health promoter will use combinations of motivation and facilitation. The manager will use individual level change methods focusing on self-efficacy and skills. Respondents think that any method may be used under the right circumstances, although few endorsed coercive methods. Conclusions: Taxonomies of theoretical change methods for environmental change should include combinations of individual level methods that may be bundled and separate suggestions for methods targeting a level or being targeted from a level. Future research needs to cover more methods to rate and to be rated. Qualitative data may explain some of the surprising outcomes, such as the lack of large differences and the avoidance of coercion. Taxonomies should include the theoretical parameters that limit the effectiveness of the method.
DOCUMENT
Despite changing attitudes towards animal testing and current legislation to protect experimental animals, the rate of animal experiments seems to have changed little in recent years. On May 15–16, 2013, the In Vitro Testing Industrial Platform (IVTIP) held an open meeting to discuss the state of the art in alternative methods, how companies have, can, and will need to adapt and what drives and hinders regulatory acceptance and use. Several key messages arose from the meeting. First, industry and regulatory bodies should not wait for complete suites of alternative tests to become available, but should begin working with methods available right now (e.g., mining of existing animal data to direct future studies, implementation of alternative tests wherever scientifically valid rather than continuing to rely on animal tests) in non-animal and animal integrated strategies to reduce the numbers of animals tested. Sharing of information (communication), harmonization and standardization (coordination), commitment and collaboration are all required to improve the quality and speed of validation, acceptance, and implementation of tests. Finally, we consider how alternative methods can be used in research and development before formal implementation in regulations. Here we present the conclusions on what can be done already and suggest some solutions and strategies for the future.
DOCUMENT
From the article: Abstract Over the last decades, philosophers and cognitive scientists have argued that the brain constitutes only one of several contributing factors to cognition, the other factors being the body and the world. This position we refer to as Embodied Embedded Cognition (EEC). The main purpose of this paper is to consider what EEC implies for the task interpretation of the control system. We argue that the traditional view of the control system as involved in planning and decision making based on beliefs about the world runs into the problem of computational intractability. EEC views the control system as relying heavily on the naturally evolved fit between organism and environment. A ‘lazy’ control structure could be ‘ignorantly successful’ in a ‘user friendly’ world, by facilitating the transitory creation of a flexible and integrated set of behavioral layers that are constitutive of ongoing behavior. We close by discussing the types of questions this could imply for empirical research in cognitive neuroscience and robotics.
LINK
