Hochschule für Gestaltung und Kunst Basel FHNW

Dauerhafte URI für den Bereichhttps://irf.fhnw.ch/handle/11654/11

Listen

Filter

2018 - 201912020 - 20232

Einstellungen

Dateien dazu vorhanden?: NeinThema: search.filters.subject.machine learning

Bereich: Suchergebnisse

Gerade angezeigt 1 - 3 von 3
  • Kein Vorschaubild vorhanden
    Publikation
    Algorithmic experience: visualising the Instagram machine learning process for end-users
    (Hochschule für Gestaltung und Kunst Basel FHNW, 2023) Szlachta, Anna Maria; Reymond, Claire; Oplatek, Jiri; Zeller, Ludwig
    Algorithmic experience (AX) is a Human Computer Interaction concept that applies to digital products where a significant part of the end-user experience is determined by the algorithms. In other words, it is not only the quality of the interface that is relevant, but also the algorithmic processes whose result is represented by the interface. Some examples of such software products are social media platforms like Facebook, Instagram, TikTok, YouTube, LinkedIn, and others. With the advancement of algorithms, machine learning and AI, the algorithmic experience that is delivered is increasingly personalised. Moreover, the tailored content means that the experience can be different for each user, depending on several factors. Digital product designers therefore face the challenge of researching with users about their algorithmic experience. However, when we speak of algorithms, we mean complex processes that are invisible to end-users. Typically, understanding algorithmic models and concepts also requires advanced mathematical or technical knowledge. So far, such research has been conducted by means of in-depth interviews, but hit many additional obstacles with, for example, the understanding of basic algorithmic vocabulary. During the thesis, it was proposed to overcome this barrier by using visualisation. Building a common ground between designers and end-users using visual language could deepen the quality of the interviews. This would enable UX researchers to provide more valuable insights to the data science team and also be responsible for shaping the algorithmic experience of the product. The popular social media platform Instagram was chosen as an example for visualisation. A series of images explored how to present the algorithmic process to non-experts. The process included not only image-making but also conversations with Instagram users in an iterative process: design – evaluation with five users during in-depth interviews – design – and next sessions with users. This made it possible to provide an interactive final visualisation that mainly focuses on inputs and outputs, elements in the algorithmic process that are familiar to users. Combined with Instagram’s familiar layout, this enabled discussion on multiple levels, not only referencing users’ own experiences of using the platform, but also learning how much and how users combine information in their mental model of the algorithm. The visual investigation also allowed for a broader consideration of privacy policies and data gathering by technology companies, and their real impact on users’ algorithmic experience. The illustrations opposite show the concepts tested during the design process. During the image-making process, an effort was attempted to combine, on the one hand, Instagram’s known layout for users and, on the other, to present what data is processed by machine learning and AI processes that determine the shape of the algorithmic experience. However, the main focus was on the input and output data in the input-black-box-output process.
    11 - Studentische Arbeit
  • Kein Vorschaubild vorhanden
    Publikation
    Making Arguments with Data
    (09.06.2022) Savic, Selena; Martins, Yann Patrick
    Whether we are discussing measures in order to ‘flatten the curve’ in the ongoing pandemic, or what to wear in face of the most recent weather forecast, we make arguments based on patterns and trends observed in data. What makes these patterns observable? Making arguments with data requires critical engagement with datasets, as well as computational processes to gather data, to organize and model their relationships. This article presents an approach to practicing ethics when working with large datasets and designing data representations. The arguments we make are based on the development and use of a computational instrument, and working with digital archives. We programmed and used web-based interfaces to sort, organize and explore a community-ran archive of radio signals. Inspired by feminist critique of technoscience and recent problematizations of digital literacy, we argue that one can navigate machine learning models in a multi-narrative manner, and that knowledge of radio signals or any other technical artefact transgresses domains. We propose visual explorations of complex data structures that enable storytelling and an understanding of datasets that resists extraction of discrete identities from the data. We hold that the main challenge to sovereignty comes from lingering forms of colonialism and extractive relationships that easily move in and out of the digital domain. Countering both the unbased narratives of techno-optimism, and the universalizing critique of technology, we discuss an approach to data and networks that enables a situated critique of datafication and correlationism from within.
    06 - Präsentation
  • Kein Vorschaubild vorhanden
    Publikation
    Application of Machine Learning Within the Integrative Design and Fabrication of Robotic Rod Bending Processes
    (Springer, 2018) Smigielska, Maria; De Rycke, Klaas; Gengnagel, Christoph; Baverel, Olivier; Burry, Jane; Mueller, Caitlin; Nguyen, Minh Man; Rahm, Philippe; Ramsgaard Thomsen, Mette
    This paper presents the results of independent research that aims to investigate the potential and methodology of using Machine Learning (ML) algorithms for precision control of material deformation and increased geometrical and structural performances in robotic rod bending technology (RBT). Another focus lies in integrative methods where design, material properties analysis, structural analysis, optimization and fabrication of robotically rod bended space-frames are merged into one coherent data model and allows for bi-directional information flows, shifting from absolute dimensional architectural descriptions towards the definition of relational systems. The working methodology thus combines robotic RBT and ML with integrated fabrication methods as an alternative to over-specialized and enclosed industrial processes. A design project for the front desk of a gallery in Paris serves as a proof of concept of this research and becomes the starting point for future developments of this methodology.
    04B - Beitrag Konferenzschrift